Progress and Challenges in Electromechanical Coupling of Radio Telescopes

,


Introduction
Many major astronomical discovery projects in recent years involve the radio telescope. For example, in the original gravitational wave discovery in 2014 [1,2], scientists used a 10 m aperture South Pole Telescope (SPT, located in Antarctica). In 2017, the European Southern Observatory used 4 Very Large Telescopes (VLTs) in Chile to observe the merger of two neutron stars [3]. Behind the frst black hole photo taken in 2019, a total of 8 telescopes (or telescope arrays) are combined into the Event Horizon Telescope (EHT) for global observation [4]; these telescopes include the Atacama Large Millimeter/submillimeter Array (ALMA, located in Chile with 54 12 m and 12 7 m aperture radio telescopes), the Atacama Pathfnder Experiment Telescope (APEX, 12 m aperture, located in Chile), the James Clerk Maxwell Telescope (JCMT, 15 m aperture, located in Hawaii), the Submillimeter Array, SMA, located in Hawaii with 8 6 m aperture radio telescopes), the Submillimeter Telescope (SMT, 10 m aperture, located in Arizona, USA), the Large Millimeter Telescope (LMT, 50 m aperture, located in Mexico), the Pico Veleta Telescope (PVT, 30 m aperture, located in Spain), and the SPT as mentioned earlier in the article. In 2020, the European VLBI Network (EVN) used eight other telescopes to observe multiple repeated fast radio burst sources [5], including the Efelsberg Telescope (100 m aperture, located in Germany), the Tianma Telescope (65 m aperture, located in Shanghai, China), the Medicina Telescope (32 m aperture, located in Italy), the Toruń Telescope (32 m aperture, located in Poland), the Irbene Telescope (32 m aperture, located in Latvia), the Jodrell Bank Mark II (25 m * 38 m, located in the UK), the Onsala Telescope (25 m aperture, located in Sweden), and a telescope in the Westerbork array (a total of 11 25 m aperture radio telescopes, located in the Netherlands).
In these discoveries, radio telescopes have played a crucial role. When making astronomical observations, celestial objects can not only be observed in visible light but also radiate radio waves [6]. As the device for receiving these radio waves, a large antenna body is the most obvious feature of the radio telescope [7]. As a typical electromechanical coupling device, the antenna body of a radio telescope consists of two parts, electrical and mechanical. An antenna is installed on the base of each radio telescope, and at least one receiving device is used to detect signals. Te structural part of the antenna is not only the carrier and guarantee for the realization of the electrical performance of the antenna but also often restricts the realization of the electrical performance [8]. Refector antennas are an ideal type of radio telescopes, which refect radio waves well to a single point. With the continuous development of radio astronomy, the high sensitivity and high-resolution requirements of the antenna system need to increase the efective receiving area of the antenna while ensuring surface accuracy [9], which means that the aperture of the refector antenna needs to be increased. With the increase of the antenna aperture, the weight of the antenna structure increases dramatically. Te large weight makes it increasingly difcult to ensure the stifness of the antenna structure [10]. Meanwhile, the improvement of the working frequency of the antenna also requires an unprecedented degree of refector surface accuracy [11]. Its design operating frequency is up to 115 GHz, and the surface root mean square error (RMSE) is required to reach 0.2 mm. Such high accuracy requirements have brought unprecedented challenges to the design of large aperture antenna structures.
Te mutual infuence and restriction of the electrical performance and mechanical structure of the radio telescope have made the development cycle of the large-aperture highperformance single-dish antenna relatively long. At the same time, the complex observation environment makes the operation and maintenance of the telescope quite difcult. Tus, this paper frst outlines the development of radio telescopes in recent years, including major constructions that are completed and under construction or in preparation for construction, then summarizes the key technologies of electromechanical coupling of radio telescopes from two aspects of design manufacturing and observation operating and analyze the key technologies one by one. On this basis, the research hotspots of radio telescopes in future developments are discussed at last to provide a technical reference for the efcient operation and maintenance of radio telescopes.

Development Trends of Radio Telescopes in Recent Years
Te large radio telescopes in service in the world, in addition to those mentioned in the introduction, also include the Green Bank Telescope (GBT, 100 m × 110 m aperture, located in West Virginia, USA), the Sardinia Radio Telescope (SRT, 64 m aperture, located in Italy), Nobeyama Telescope (45 m aperture, located in Japan), the Lovell Telescope (76 m aperture, located in UK), Parkes Radio Telescope (64 m aperture, located in Australia), Yebes Telescope (40 m diameter, located in Spain), as well as the 26 m Nanshan Telescope located in Urumqi, Xinjiang, China, the 40 m Telescope located in Kunming, Yunnan, and the 50 m Telescope located in Miyun, Beijing, etc. Tese telescopes have been summarized in detail in [6,7,9]; thus the authors give no further elaborations. In addition, the Arecibo Telescope (305 m aperture, located in the U.S. island of Puerto Rico), the second largest singledish fxed radio telescope in the world, crashed due to the suspended electrical receiving part after nearly 60 years of service, and the main body of the refector surface was also damaged, this telescope has ofcially retired at the end of 2020. Tis section will focus on the development of various radio telescopes that have been constructed and put into operation in recent years, as well as those under construction and be constructed in the future. Te authors have drawn a map to better exhibit the distribution of the major radio telescopes in the world according to their geographical distribution and types ( Figure 1).

Radio Telescope at Kashgar Deep Space Station, China.
Te Kashgar Deep Space Measurement and Control Station in Xinjiang, China, is one of the three deep space measurement and control stations in the Chinese deep space measurement and control network. Te other two are located in Jiamusi, Heilongjiang, and Argentina [12]. Te Kashgar Deep Space Station was built and put into use in 2013. At frst, there was only a set of 35 m refector antennas capable of measuring and controlling the three frequency bands of S/X/Ka (Figure 2(a)). By the end of 2020, the Kashgar Deep Space Station has built three new antennas with an aperture of 35 m (Figure 2 Figure 4). In 2017, the Argentine Deep Space Station was built and put into operation using a set of deep space measurement and control equipment with the abilities of working in three frequency bands of S/X/Ka [12]. After the construction of the Argentine deep space station, China has built a deep space exploration network with S/X/Ka multifrequency band measurement and control capabilities, integrating measurement and control, data transmission, interferometry, continuous TT and C, reliable communication, and precise navigation services with very high performance [15]. Te coverage rate of deep space spacecraft has increased to more than 90%, and the comprehensive level is at the forefront of the world [16].

Tianma 65 m Radio Telescope in Shanghai, China.
Te Tianma 65 m telescope ( Figure 5), located in Sheshan, Songjiang, Shanghai, was set up in the project in 2008 [17], and its design and construction began in 2009. Te frst stage of construction was completed in 2012, and it has VLBI observation capabilities in S/X, L, and C bands [18]. By 2016, receivers in the Ku, X/Ka, and K bands were added. Te main refector of the antenna adopts the best matching-shaped double refector. Te subrefector adopts the six-bar parallel mechanism adjustment technology to achieve high-precision adjustment of fve degrees of freedom [19]. In addition, Tianma Telescope is the frst radio telescope in China equipped with an active surface system. After the active surface system is turned on, the surface shape accuracy can be better than 0.3 mm. Te Tianma Telescope has completed the Chang'e series of missions and Mars exploration missions with high quality during the operation phase and completed the overall acceptance in 2017. It is currently actively serving domestic radio astronomy observations.

Haoping 40 m Radio Telescope in Shaanxi, China.
Te Haoping Observatory of the National Service Center of the Chinese Academy of Sciences is in Luonan, Shaanxi Province. It has a 40 m aperture refector antenna ( Figure 6), which is mainly used to track and receive downlink signals from GNSS (Global Navigation Satellite System) satellites to achieve high GNSS satellite signal accuracy in observation International Journal of Antennas and Propagation and analysis [20], and will also be used to carry out radio astronomy observations such as pulsars and spectral lines etc. [21], and VLBI work is also planned in the future [22]. At present, based on the radio telescope, research on the time scale of pulsars is being carried out, and long-term timing observations have been carried out on many high-precision millisecond pulsars.

Wuqing 70 m Radio Telescope in Tianjing, China.
After the Zhurong Mars probe landed on Mars, the signal sent back by the probe was extremely weak due to the long distance of 400 million kilometers between Earth and Mars.
To ensure the steady operation of the Mars probe, the National Astronomical Observatory of the Chinese Academy of Sciences built a 70 m aperture antenna in Wuqing, Tianjin province, as a high-performance receiving system, mainly working in the S/X/Ku band. Its highly symmetrical umbrella support structure design makes the refector deformation more uniform, to achieve the efect of equal fexibility support [23]. Te overall installation of the refector was completed in April 2020. Te acceptance was completed in February 2021, and together with the Beijing Miyun 50 m antenna and the Yunnan Kunming 40 m antenna, it was in charge of receiving the return data of Tianwen-1 [24]. Te completion of the Wuqing 70 m antenna ( Figure 7) makes it the largest single-dish steerable antenna in Asia.

Five Hundred-Meter Aperture Spherical Radio Telescope (FAST).
After the retirement of the Arecibo telescope, FAST became the only "eye for the sky" in the world. Te 500-meter aperture spherical radio telescope in Pingtang, Guizhou Province, is currently the largest radio telescope in the world with the highest degree of sensitivity ( Figure 8). Its site selection and preresearch began in 1994 [25]. Te project was approved in November 2002, then started construction in 2011, and was completed in September 2016. After nearly 4 years of commissioning and trial operation, it passed the national acceptance in January 2020. Tis enormous project took 26 years to complete. It adopts a Gregorian feed cabin to collect signals. Compared with the Arecibo telescope, there are three main innovations. Firstly, the karst landform of the selected site is very consistent with the construction conditions of the giant single-aperture telescope. Secondly, it is active optics technology. Each panel of the refector is equipped with a real-time active controller. Te active control system can form a 300 m aperture instantaneous parabolic surface in the direction of celestial observation. Finally, the light cable-driven feed cabin technology of optical-mechanicalelectrical integration makes the fexible connection between the feed and the refector to achieve high-precision pointing and tracking. Over 100 new pulsars were discovered using FAST during the commissioning and trial operation phase [26]. FAST mainly operates between 70 MHz and 3 GHz, and compared with Arecibo, the scanning speed is three times faster in the L-band and the sky coverage range is two times wider [27]. Te successful construction and operation of FAST are playing a signifcant role in the feld of radio astronomy.

Greenland Telescope. Te Greenland Telescope (GLT)
is located in the northwest of Greenland. It is a 12 m aperture antenna working in the millimeter/submillimeter band ( Figure 9). Te antenna is the prototype of ALMA, originally located in Chile, and later gifted by the US National Science Foundation to the Academia Sinica Institute of Astronomy and Astrophysics (ASIAA) [28]. After upgrading, the construction was completed and put into operation in 2017. Te main refector is equipped with a deicing device. Te servo system is based on the ALMA antenna system using two high-precision tiltmeters, one on the azimuth bearing and the other on the support cone. Te antenna is currently mainly involved in the VLBI observation of the M87 galaxy [29,30].
2.1.9. Tai National 40 m Radio Telescope. Te 40 m Tai National Radio Telescope (TNRT) was proposed by the National Astronomical Institute of Tailand in 2017 and was assembled in 2020 ( Figure 10). TRNT has been put into use in early 2022. Its most important diference is its main focus on low-frequency or phased array feed systems. It is mainly used for astronomical observations such as pulsars, fast radio bursts, star-forming regions, galaxy, and active galactic nuclei, et al. As a single-dish instrument, TNRT is a perfect tool to explore time domain astronomy with its agile observing systems and fexible operation. Due to its ideal geographical location, TNRT will signifcantly enhance Very Long Baseline Interferometry (VLBI) arrays, such as the East Asian VLBI Network (EAVN), Australia Long Baseline Array (LBA), and European VLBI Network (EVN) [31]. Te antenna is an upgraded version of the Spanish Yebes radio telescope, working at 300 MHz-115 GHz [32].

Canadian Hydrogen Intensity Mapping Experiment
Telescope. Canadian Hydrogen Intensity Mapping Experiment Telescope (CHIME) consists of four 20 m × 100 m semicylindrical parabolic refectors ( Figure 11). It completed construction in September 2017 and began to operate for observation in September 2018. CHIME operates in the 400-800 MHz frequency band [33]. Each cylinder equipped with 256 dual-polarization feeds will continuously image the entire northern sky to measure the distribution of neutral hydrogen in the universe and study the properties of dark energy [34]. It is used to measure the expansion history of the universe by converting the collected cosmic signals into a three-dimensional map of hydrogen density [35]. CHIME could also be used to explore new fast radio bursts and pulsars [36].

Australian Square Kilometer Array Pathfnder.
Te Australian Square Kilometer Array Pathfnder (ASKAP) is a pioneer project of the frst Square Kilometer Array (SKA) (Figure 12). It is located in the Murchison Radio Astronomy Observatory in Western Australia and has a total of 36 refector antennas with a diameter of 12 m [37]. ASKAP began construction in 2009 and completed the deployment and assembly of all antennas by 2012 [38]. ASKAP's prototype is the Boolardy Engineering Test Array, the frst aperture synthetic radio telescope to use phased array feed technology [39]. Te success of early science projects using ASKAP could be seen in all areas of astrophysics, indicating that ASKAP's scientifc prospects are very appealing [40].

Murchison Widefeld Array.
Te Murchison Widefeld Array (MWA), also located at the Murchison Radio Astronomy Observatory, is one of SKA's pioneering projects ( Figure 13). Te frst phase of the MWA was commissioned in 2013 and a total of 128 aperture arrays were constructed [41]. Ten, it began the full operation. Te second phase was completed in 2017 and ofcially put into use in 2018; the construction of the second phase doubled the number of antennas [42]. Te MWA is an international efort by different countries and its data are publicly available [43]. MWA's primary scientifc goals consist of four key scientifc themes: searching for the redshift of 21 cm radiation in the reionization era of the early universe, galactic and extragalactic sky surveys, time-domain astrophysics, as well as the study of scientifc problems and space weather in the sun, heliosphere, and ionosphere [44].
2.1.13. Karoo Array Telescope. Te Karoo Array Telescope (MeerKAT) is a medium-sized pioneer telescope instrument group in SKA [45], located in South Africa, consisting of 197 ofset refector antennas with a diameter of 13.5 m ( Figure 14). Te concept of MeerKAT was frst proposed in 2009 [46], and seven prototypes (KAT-7) were initially built [47]. It was not until January 2022 that the construction and networking of 197 antennas were fnally completed [48]. After ten years of design and construction, the observation of the central region of the Milky Way has achieved promising results after MeerKAT was ofcially put into operation [49].      Figure 15). Te HERA aims for a high-signifcance power spectrum measurement [50]. Te main goal of HERA is to detect cosmic signals from the reionization era [51]. In the later stage, HERA will deploy a total of 350 antennas to form an array. At the same time, HERA is also the last of the four pioneer projects of SKA [52].

Low-Frequency Array.
Low-Frequency Array (LOFAR) is a large distributed radio telescope ( Figure 16) [53]. LOFAR was ofcially put into operation and observation in December 2012, using the omnidirectional dipole antenna as the array element, working in the frequency band of 10 MHz to 240 MHz. Since it is a fxed antenna, LOFAR uses aperture synthesis software for imaging [54]. Te lower operating frequency band of LOFAR enables it to greatly enhance the observation efciency of pulsars in the low radio frequency band [55].

QiTai 110 m Radio Telescope, Xinjiang, China.
Te ultra-large aperture fully steerable radio telescope QTT (QiTai Telescope) ( Figure 17) is located in Qitai County, Xinjiang, China. Its aperture is as wide as 110 m, and its operating frequency covers 150 MHz-115 GHz, which is highly complementary to FAST [56]. Te overall refector surface structure of FAST is fxed on the ground, and the formation of the refector surface needs to be controlled by the actuators during observation, while QTT can not only rotate in azimuth and pitch but also is equipped with the      International Journal of Antennas and Propagation active surface system, making it the largest single-dish fully steerable antenna in the world. Te refector adopts the shaping technology so that the main and secondary refectors can be converted between the shaping and the standard surface to realize the focus multibeam. By using fast antenna measurement and shaping technology, the closed-loop control of the active surface is realized, and the deformation of the antenna mechanical structure caused by gravity and external load is corrected in time. Te preresearch of QTT has also been carried out for nearly ten years. Recently, the Xinjiang Astronomical Observatory of the Chinese Academy of Sciences completed the bidding for the antenna construction at the end of September 2022. Te construction period is expected to be 5 years. Te completion of QTT will play an important role in the fundamental scientifc research felds of gravitational wave detection, black hole discovery, and the origin of galaxies [11].

Jingdong 120 m Radio Telescope, Yunnan, China.
In August 2020, the Yunnan Provincial Government and the Chinese Academy of Sciences formally signed the "Cooperation Agreement on the Joint Construction of the Jingdong 120 m Pulsar Radio Telescope (JRT)" ( Figure 18). In September, the project's launching ceremony was held in Jingdong, Pu'er City, Yunnan Province, marking the ofcial start of construction of the world's largest omnidirectional rotatable low-frequency radio telescope. Te construction period is expected to be three years [57]. Te operating frequency of JRT is concentrated at 0.3-10 GHz, and the main observation target is pulsars. It will carry out cuttingedge scientifc research such as gravitational wave detection and strong gravitational feld relativity testing. Unlike QTT, which uses solid panels for the overall refector surface, JRT only uses solid panels in the central 40 m diameter position, and mesh panels are used for the rest due to low-frequency operation and overall weight reduction.

Shigatse and Changbaishan 40 m Radio Telescope,
China. In the upcoming fourth phase of the lunar exploration mission, to meet the requirements of accurate VLBI orbit measurement, the Chinese Academy of Sciences plans to build two 40 m aperture radio telescopes measurement and control station in Shigatse, Tibet, and Changbaishan, Heilongjiang Province. Since 2018, the Shanghai Astronomical Observatory has traveled to Tibet seven times to investigate the site selection. Finally, the junction of Nie Rixiong Township, Sangzhuzi District, Shigatse City, and Chaxiu Township, Sakya County, was used as the construction site of the 40-meter radio telescope for the new lunar exploration project. In 2019, the Shanghai Astronomical Observatory (SOA) signed a cooperation agreement with the Shigatse City People's Government, which means that a new refector antenna will be constructed in the central and southern parts of Tibet to help China Aerospace go further. Moreover, in 2021, the SOA also confrmed the construction site of the Changbaishan 40 m radio telescope and advanced the construction of the "six stations and one center" of the fourth phase of China's lunar exploration.

Nanjing Zijinshan Observatory 60 m Submillimeter
Telescope. To promote China's submillimeter-wave astronomical observation capabilities, the Chinese astronomical community plans to build a 60 m aperture submillimeterwave radio telescope, which is still in the preresearch and preparation stage. Te plan is to develop submillimeter wave adaptive optics, manufacture phase reference artifcially, use the phase reference transmitter of the main refector to transmit microwave signals, observe with a scientifc observation receiver, and correct the active surface in real time according to the real-time observation results, so as to realize the real-time measurement and correction of the antenna wavefront. In 2018, the Nanjing Zijinshan Observatory will take the lead to hold a seminar on key technologies for largescale submillimeter-wave telescope antennas, and conduct technical exchanges and discussions on the design, construction, and operation of 60 m aperture submillimeterwave telescope antennas.

Large Submillimeter Telescope in Japan.
After the international astronomy community proposed the concept of the next-generation large-aperture single-dish telescope in 2008, the National Astronomical Observatory of Japan proposed the design concept of the Large Submillimeter Telescope (LST) in 2014 and published the frst related paper in 2016 [58]. Tis telescope is expected to operate at 70-950 GHz. Its key technology is an active surface control technology. Each adjustable panel of the refector can accurately compensate for gravity and thermal deformation and achieve high surface accuracy of each panel. Te actuator on the back of the panel can adjust and correct the panel deformation in real time. LST was listed in 2020 as one of the large academic projects in the feld of astronomy and astrophysics in the 2020 Master Plan by the Japan Science Council.

Atacama Large Aperture Submillimeter Telescope.
Te Atacama Large Aperture Submillimeter Telescope (AtLAST), which is expected to be built in Chile, is also a 50 m aperture refector telescope that also works in the submillimeter wave band. Relative design concepts were proposed in 2015 [59], and more details were introduced in the AtLAST project in 2020 [60]. AtLAST uses an advanced cryogenic receiver, which is boxed in a cryogenic thermostat with a diameter of 1 m. Te cryogenic thermostat is bolted to the ceiling of the equipment room at the focus of the refector. It is obvious that the AtLAST project led by the European Union has the same aperture as the prementioned LST project, both working in the submillimeter wave band. Terefore, the Japanese Science Committee and the European Union are considering merging the two projects, which is expected to be completed in 2024 [61].

Cerro Chajnantor Atacama Telescope. Te Cerro
Chajnantor Atacama Telescope (CCAT), originally known as the Cornell Caltech Atacama Telescope, was designed jointly by Cornell University, Caltech, and the Jet Propulsion Laboratory. Te three institutions jointly study the construction of a 25 m aperture submillimeter telescope on a 5000 m peak in northern Chile [62]. Te antenna's back frame and main refector will be constructed of carbon fber reinforced plastic [63] as to meet the surface RMS error requirement of 10 microns [64]. However, due to the problem of funding, although the designing concept was proposed in 2006, the construction of the antenna has not yet started.   International Journal of Antennas and Propagation

Lunar Crater Radio Telescope.
Te far side of the moon has an excellent environment for astronomical observation because the moon can act as a physical barrier to block signal interference from the Earth or the sun. In 2018, NASA proposed the construction of the Lunar Crater Radio Telescope (LCRT) [65], and by using the existing craters on the moon, crawling robots were used to arrange the wire mesh to form a parabolic refector with a diameter of 1 km [66]. Te telescope will work in the ultralow frequency range of 3-30 MHz [67], and the main scientifc goal is to explore the evolution of the interstellar medium when the frst stars in the universe were formed.
2.2.9. BINGO Radio Telescope. Te Baryon acoustic oscillations from Integrated Neutral Gas Observations (BINGOs) project aim to build a radio telescope to detect baryon acoustic oscillations [68]. It will consist of two fxed 40 m aperture telescopes, equipped with a feed array with 50 receivers. Te concept of the bingo project was proposed in 2012, and with the gradual progress of the project, the main structure of the telescope, the optical part, and the receiver prototype were completed around 2019, and the overall construction is expected to be completed around 2022 [69].

Hydrogen Intensity and Real-Time Analysis Experiment Radio Telescope. Hydrogen Intensity and Real-time
Analysis eXperiment (HIRAX) is a radio telescope array located in Karoo, South Africa. It has the same working frequency and scientifc goals as CHIME. CHIME is mainly used to detect the northern sky, while HIRAX is used to detect the southern sky [70]. Tey complement each other. HIRAX will consist of an array of 1024 6 m aperture refector antennas [71]. A prototype array consisting of 8 antennas was frst built in 2017 for preliminary experimental verifcation, and a pathfnder array with 128 antennas will be built in the future. It is expected to expand to the fnal 1024 antenna arrays in 2022 [72].

2.2.11
. Square Kilometer Array. Square Kilometer Array (SKA) is a large-scale cooperation project between governments around the world. Te related concept was frst proposed in the early 1990s [73]. SKA is planned to be constructed in South Africa (mainly constructing midfrequency arrays and dish arrays) and Australia (mainly constructing low-frequency arrays). After completion, it will have a total collecting area of more than 1 square kilometer. Te approval for the start of construction from its governing Council occurred in June 2021 [74].

Conclusion.
It can be seen from the above development trends that in the feld of radio astronomy, refector antennas can achieve high bandwidth, high gain, and high resolution at low cost [75], thus becoming indispensable. At the same time, it is because of the higher and higher gain and resolution requirements that the antenna aperture is getting larger and larger, and the electromechanical coupling problem caused by the mutual infuence and restriction between electrical properties and structure is also becoming increasingly prominent. Terefore, how to solve the problem of radio telescope in electromechanical coupling is crucial. Te next section summarizes the research progress of key electromechanical coupling technologies in radio telescopes from the perspectives of design manufacturing and observation operating.

Research Progress of Electromechanical Coupling of Radio Telescopes
At this stage, the characteristics of various electromechanical coupling technologies can be summarized into three aspects, as shown in Figure 19. In terms of design and manufacture, electromechanical coupling modeling and design include theoretical models, environmental load characteristics analysis, antenna panel, and structure design and tolerance allocation of antenna morphology. In terms of observation operation, the antenna performance assurance under the two kinds of loads of slowly varying loads and rapidly varying loads is analyzed. Tis section introduces the development status of three aspects of electromechanical coupling technology around Figure 19 in detail.

Electromechanical Coupling Modeling and Design
3.1.1. Electromechanical Coupling Modeling. For large radio telescopes, Reference [76] point out that due to the large aperture of the refector, the complex structure, and longterm working in the outdoor environment, environmental loads such as self-weight, sunshine, wind, and snow also cause negative efects. Te systematic errors caused by these factors and the random errors generated during the installation and manufacture of the telescope will afect the ultimate performance of the telescope (such as pointing, gain, and others.). To accurately and quickly analyze the efects caused by these errors, it is necessary to develop an efcient and accurate computational tool [77], that is, the electromechanical coupling model of refector antennas.
For the feedforward refector antenna or double refector antennas commonly used in engineering (the subrefector and feed can be equivalent to feed to the virtual focus of the subrefector by the equivalent feed method), Reference [78] points out that the main structural factors afecting electrical performance can be summarized into three aspects: main refector error, feed pointing error, and feed position error. Terefore, the infuence on the ideal pattern can be analyzed from these three aspects, and fnally obtaining the electromechanical coupling model.

International Journal of Antennas and Propagation
For the main refector, its random errors can be described as diferent distribution functions on the main refector, and then superimposed on the infuence of systematic errors on the deformation of the refector. Te systematic error is a deterministic error, which can be obtained by fnite element analysis. Te main refector area can be regarded as the far area of the feed. When the deformation of the main refector is small (compared with the overall size of the antenna), the efect of the main surface error on the electromagnetic feld amplitude of the aperture surface can be ignored, and it is considered that the main surface error will only cause relative errors on the electromagnetic feld in the aperture surface [79,80]. As for the feed source pointing error, that is, when there is an angular error between the feed source pointing and the central axis of the refecting surface, the irradiation path from the feed source to the aperture surface will change, which will bring amplitude errors to the electromagnetic feld of the aperture surface and will cause the sidelobe level to rise. For the feed position error, the ofset of the feed position will change the phase center, thereby causing the phase change of the electromagnetic feld of the aperture surface [78].
Combining the error efects of the previous three aspects and the ideal pattern formula of the refector antenna, the fnal electromechanical coupling model of the refector antenna can be obtained [81] as follows: where f 0 (ξ − Δξ(δ(β)), ϕ ′ − Δϕ ′ (δ(β))) is the efect of the feed pointing error on the phase of the electromagnetic feld on the aperture surface, φ f (δ(β)) is the efect of the feed position error on the electromagnetic feld phase of the aperture surface, and φ s (δ(β)) and φ r (c) are the main surface deformation and the efect of the main surface random error on the electromagnetic feld phase of the aperture surface, respectively.

Characteristic Analysis of Environmental Load.
After the electromechanical coupling model is established, to accurately analyze the efects of environmental loads in the designing phase (such as wind, temperature, and gravity), thereby giving an evaluation of the electric performance of the telescope's pointing gain, an accurate analysis of the infuence mechanism of various environmental loads on the performance of the telescope is also of great necessity.
Te wind in the natural environment can be divided into two types: steady-state winds and transient winds. Te period of steady-state wind is generally much longer than the natural vibration period of the structure, and its infuence on the structure can be regarded as a static force that does not change with time, thus applied to the structural for analysis as a static load. For instance, Reference [82] analyzes structural static deformation of the antenna under 8-level wind speed. Reference [83] analyzes wind pressure distribution on the refecting surface of the antenna at a wind speed of 14 m/s through wind tunnel experiments. While the period of transient wind is short, its action properties are dynamic and will cause structural vibration. Structural vibration will lead to the change of the antenna pointing. Reference [84] analyzes the antenna vibration and pointing error changes under the infuence of transient wind. For the larger aperture surface of the radio telescope, whether it is steady wind or transient wind, a certain torque will be generated on the antenna. Under the infuence of this  torque, the antenna will undergo a certain torsional deformation, which will cause a change of the antenna pointing. References [85,86] analyze the torque of the telescope antenna under wind load. Both thermal and gravitational loads can cause deformation of the refector surface of the antenna, thereby afecting the fnal pointing and gain performance of the telescope. Any material will be deformed to a certain extent under an uneven temperature feld. Te telescope will be afected by the sun because it is always in the natural environment. Te sun's irradiation will create a shadow area on the surface of the refector surface and the back frame, which will cause the antenna to not be heated evenly, eventually causing the deformation of the refector surface and the back frame. Reference [87] analyzes the thermal stress and deformation of the refector surface of the antenna under diferent solar incident angles. Any substance will be afected by gravity, so the infuence of gravity load is unavoidable, but since gravity will not change, as long as the attitude of the antenna does not change, then the gravitational deformation will not change, which can be analyzed by simulation or various measurement methods to obtain the deformation of the antenna under gravitational loads, such as total station [88], photogrammetry [89], and laser scanner [90].

Structural Innovation Design Based on Electromechanical Coupling.
Although the infuence of gravity load cannot be avoided, it can be reduced by means of structural conformal design [91]. Te simplest and most efective way is to reduce the weight of the antenna back frame. However, if the weight of the back frame is reduced by simply reducing the number of trusses, it will inevitably lead to a reduction in the overall strength and stifness of the antenna. On the premise of ensuring the rigidity and strength of the structure, replacing some truss structures with prestressed cables can not only reduce the weight of the antenna back frame but also change the stress distribution during the operation of the structure to reduce structural deformation [92]. Reference [93] uses parts of the tension and members of the antenna are replaced with prestressed cables. Under the same accuracy, the weight of the 8 m aperture refector antenna is reduced by 15% through the cable-truss composite structure. In addition, the antenna structure can also be designed with equal fexibility, that is, it is necessary to optimize the topology of the antenna structure to ensure the consistency of its node deformation. Equal fexibility design can be seen in antennas such as 37 m HUSIR [94], 50 m LMT [95], 64 m SRT [96], and 110 m Efelsberg [97]. Meanwhile, Reference [98] also uses an equal fexibility design to optimize the support structure of the QTT 110 m antenna refector that is to be constructed. Te 16-point equal fexible support structure used in the support structure greatly improves the deformation of QTT. For radio telescopes, the driving power required for the antenna base, whether it is a wheeled table or a roller type, is relatively large. No matter how the traditional optimization of the back frame or the base frame is carried out, the efect of reducing the weight of the antenna is not signifcant enough. Only by innovating the structure can the weight of the antenna be greatly reduced. Compared with the traditional series drive structure, the parallel drive structure has the advantages of high rigidity, high accuracy, large bearing capacity, and small motion load. For example, Reference [99] uses the 6-DOF Stewart platform to control the pointing of the AMiBA telescope. Among radio telescopes, the Stewart platform is mostly used to control the attitude of the subrefector and is not suitable for large antenna structures. In [100], the authors propose an innovative design of the antenna mechanism, including a cable-truss combined backup structure and parallel drive antenna, and used this design to verify the 26 m aperture antenna. Te new design reduces the weight of the antenna by 54.3%, and the driving power is reduced by 63.3%.

Design of Antenna Panel Block.
Te ultimate purpose of the previously mentioned conformal design of the antenna structure is to ensure the surface accuracy of the telescope. With the increasing requirements for surface accuracy of large radio telescopes, the manufacturing accuracy of the antenna panel has become of vital signifcance. Considering the machining, transportation, installation, and manufacturing costs of the antenna, it is necessary to design a reasonable block of the main refector panel of the antenna. Te antenna panel is machined by molds. Terefore, although the smaller the panel, the higher the manufacturing accuracy; but the number of corresponding molds will also increase, thus increasing the cost. At the same time, the smaller the antenna panel, the more support structures are required for the overall installation of the antenna refector, which will lead to increasingly complex back frame structures and increased self-weight. Terefore, the smaller antenna panel does not solve the problem and the mold of the antenna panel needs to be reasonably designed. For the QTT 110 m antenna, Reference [101] reduces the original 23 types of molds to 14 types of molds under the premise of meeting the requirements of the electrical performance of the antenna through reasonable block design and mold sharing strategy.

Tolerance Allocation of Antenna Morphology.
Te structural design of a radio telescope is a systematic project. Its structure consists of multiple substructures, such as wheel rails, mounts, back frames, feed bins, main refectors, and subrefectors. Tese structures together determine the ideal performance of the telescope. In practical applications, the performance can also be controlled and compensated by the control system of the antenna. Reference [95] points out that the structural design of the antenna needs to fnd the optimal balance between structure and control because the fnal performance of the radio telescope is directly linked to the surface accuracy [102]. Meanwhile, the surface accuracy of the antenna is jointly determined by each substructure, which means that engineers need to reasonably allocate and control the tolerances of each substructure system. In the past, this process was usually performed by engineers using their own design experience. If the fnal performance did not meet the standard, then an adjustment was made and the process is considered a trial error. After the electromechanical coupling model is established, the infuence of each structural error on the fnal electrical performance can be analyzed through this model, to obtain the infuence degree of each substructure of the antenna on the performance, and then allocate the tolerance accuracy more reasonably according to the infuence degree. For example, Reference [103] analyzes the infuence of axis errors such as the antenna azimuth axis and pitch axis and the orbital irregularity of the antenna wheel rail on the fnal pointing accuracy. Reference [104] lowers the accuracy requirements for the main refector by using subrefector compensation technology, which can provide certain guidance for the tolerance allocation of the antenna's main surface panel in the design stage. Studies in References [79,104] show that the accuracy of each ring panel on the main surface of the antenna has diferent efects on the fnal electrical performance. Together with Reference [101], it is explained that the accuracy of the panel near the center of the refector has a greater impact on the electrical performance than that of the outer panel, so the machining accuracy of the inner panel can be improved during the process, and the machining of the outer panel can reduce its precision to a certain extent.

Temperature Field Reconstruction.
Temperature load is one of the important loads of radio telescopes, and its most distinguishing feature is that it is greatly afected by sunlight.
Tere are 2 methods to obtain the antenna temperature feld distribution: fnite element simulation and temperature sensor measurement. Finite element simulation is subject to the limitation of calculation speed and cannot guarantee the real-time acquisition of the temperature feld. Due to the limitation of the structure of the antenna, engineers cannot put too many temperature sensors on the antenna for measurement. To ensure the accuracy and real-time performance of the antenna temperature feld acquisition, it is necessary to accurately reconstruct the overall temperature feld distribution of the antenna by using temperature data measured by a small number of temperature sensors through an appropriate method. Reference [105] combines two methods of sensor measurement and fnite element simulation. Firstly, the temperature feld database of the antenna at diferent times is obtained by using the fnite element simulation analysis, and then the temperature feld is obtained by looking up the measured temperature data in the database to obtain the overall temperature feld distribution. Finally, the temperature feld is corrected by the measured ambient temperature. An interpolation algorithm is used in [106] to interpolate the temperature data measured by a small number of sensors to obtain the overall temperature feld distribution of the antenna.

Inversion of Refector Surface Deformation.
Te refector surface of the radio telescope will deform under the infuence of gravity load and temperature load, in order to accurately analyze and compensate for the fnal performance of the telescope in real time (such as the need to quickly calculate the adjustment amount of the actuator, subrefector and feed source pointing, and others); it is of vital signifcance to quickly obtain the shape information of refector surface. Reference [107] constructed infuence factor matrix for fast calculation of thermal deformation of the antenna. Reference [108] uses precise difraction algorithms to evaluate repetitive deformations of refector surfaces. Reference [109] proposes a method for fast calculation of antenna panel deformation based on an angle sensor. Reference [110] proposes a method of ftting the refector surface based on the least squares method. Tis method calculates the six key parameters required for the ftting equation of the refector surface based on the principle of the least squares method and the integral extreme value theorem, and then obtains the ftting equation of the refector surface and completing the reconstruction of the refector surface.

Performance Compensation of Deformed Refector
Surface. When the main or subrefector of a radio telescope is deformed, it will cause problems such as reducing the antenna gain, elevating side lobes, and pointing defection. Terefore, it is necessary to compensate for this part of the electrical performance loss through a certain compensation method, such as active surface adjustment, subrefector adjustment, or feed adjustment. Reference [81] shows that the gain loss is related to the RMS error of the refector surface, so to compensate the gain, it is necessary to minimize the surface RMS error of the antenna. Tis is the purpose of the active surface adjustment technology. Tis technology actively adjusts the main refector by installing actuators on the back of the antenna panel and uses the actuators to move the deformed panel to a specifed position to reduce the surface RMS error. Active surface technology was proposed in the 1990s [111]. Until today, many large radio telescopes in the world have applied this technology. Reference [112] used active surface technology to efectively improve the observation efciency of an antenna. Reference [76] proposes an active surface adjustment method based on electromechanical coupling theory. Tis method also uses the least squares principle to calculate the distance between the deformed surface and the optimal working surface and then outputs the distance information to the actuators to complete the active surface adjustment. Reference [113] used edge sensors to measure the deformation of the telescope and actuators to ensure the regular operation and highquality data output of the radio telescope. Te subrefector adjustment refers to adjusting the subrefector of the antenna, such as subrefector pose adjustment, active subrefector adjustment, and array subreverse adjustment. Te pose adjustment of the subrefector usually uses a Stewart platform with 6 degrees of freedom to adjust the overall position of the subrefector, so that the subrefector can better match the main refector. For example, References [114,115] describe in detail the Stewart system in the subrefector of the 26 m antenna in Nanshan, Xinjiang, and the adjustment position of the subrefector is determined by maximizing the gain. Te adjustment of the active subrefector is similar to the previously mentioned adjustment of the active surface, but the actuator is installed on the back of the subrefector. Te electrical performance is compensated by adjusting the surface type of the subrefector surface. Active subrefector technology can be seen in radio telescopes such as the 43 m GreenBank Telescope [116], 37 m Haystack Telescope [117], and 100 m Efelsberg Telescope [118]. Te adjustment of the array subrefector surface is essentially a kind of electronic compensation. Te traditional subrefector surface is replaced by the array unit. By controlling the excitation amplitude and phase of each unit, a conjugate matching focal plane feld is constructed. Tus, the isophase plane is achieved on the aperture plane. Reference [119] used the microstrip refector array as a subrefector, the phase error of the aperture surface caused by the surface distortion of the main refector is compensated, thereby signifcantly improving the antenna performance.
Feed adjustment is mainly divided into feed pose adjustment and array feed adjustment. In simple terms, the pose adjustment of the feed is to adjust the position and attitude of the feed to the theoretical position corresponding to the best-ft paraboloid focus, which is similar to subrefector pose adjustment. Reference [120] proposes a method to determine the adjustment amount of refector antenna feed based on far-feld information. Array feed adjustment is also a kind of electronic compensation technology. Te array feed is used to replace the traditional waveguide feed, and the amplitude or phase of the received signal is changed by electrical modulation to achieve the antenna beamforming, to achieve compensation. References [121,122] introduced a dual-polarized phased array feed with a low noise amplifer, which greatly improves the measurement speed of the antenna.

Shaped Surface Antenna Profle Description.
In the application of refector antennas, there is often a need for special beam coverage, or efective omnidirectional radiation power is required for the target area. At this time, the beam of the antenna needs to be shaped, and this type of antenna is called a shaped beam antenna or shaped surface antenna. Te beamforming of the antenna can be realized by changing the antenna waveguide feeding device or the antenna surface structure. Tanks to the development of active surface technology, radio telescopes with the active surface system can achieve surface switching under the operating state of the antenna, to meet diferent observation requirements. Taking FAST as an example, FAST could be considered a special shaped surface antenna. Te body of FAST is a spherical refector with an aperture of 500 m. When observing, the position of the panels needs to be adjusted by the actuators so that some of its panels are combined into a paraboloid with an aperture of 300 m [123], which involves the conversion from spherical to parabolic. It can be seen that there are two technical contents in the process of switching the profle of the shaped surface: the profle description of the shaped surface antenna and the switching between the paraboloid and the shaped surface. Shaped surface antenna could be considered as a type of antenna with a special surface. Tere is no standard equation to describe the surface of this type of antenna. Terefore, the target surface needs to be described frst when switching. Reference [124] uses the Jacobian polynomial to describe the surface of the shaped surface antenna, and the fnal surface description equation is obtained by using the particle swarm optimization algorithm. In [125], the geometric information of the antenna surface under the specifed beam shape is obtained by using the geometrical optics method. In [126], a method of using Zernike polynomials to describe shaped surface antennas is proposed. Te surface shape switching of the antenna is generally carried out under normal working conditions. It will be afected by external loads, which requires the calculation of the adjustment amount from the deformed surface to the target-shaped surface. Te authors can adopt the idea of the optimal ftting surface and uses the least squares method to calculate the required displacement of each actuator node, thereby achieving the switching process from the deformed paraboloid to the shaped surface. Te switching of the shaped surface to the paraboloid can be considered as an inverse process. First, the shaped surface and the best working surface (usually a paraboloid) are described, and then the nodes displacement between the shaped surface and the best working paraboloid is calculated, thereby achieving the switching process.

Match and Call of Compensation Database.
During the operation of the radio telescope, there is often a real-time requirement for the compensation of electrical performance, to improve the observation efciency. For the infuence of slowly changing loads such as temperature load and gravity load, the fnite element simulation clearly cannot meet the real-time requirement as the fnite element model calculation is time-consuming. For gravity loads, the self-weight deformation of the antenna at any altitude can be obtained through advanced simulation. Tis is the same for temperature loads (as described in the frst part of this section). When the deformation information is known, the corresponding adjustment amount can be calculated in advance as the adjustment amount can be directly matched and called by the look-up table method during the antenna observation process. In other words, the compensation amount can be quickly obtained by establishing a compensation database in advance. Te key lies in the accuracy of database matches and calls. Reference [127] points out that the Italian 64 m SRT antenna stores the antenna compensation information corresponding to the self-weight deformation in the database in advance, and realizes real-time compensation by looking up the table during the operation of the antenna. Reference [128] uses the optimized best-ftting surface parameters and the matching degree of the main and auxiliary surfaces to call the adjustment value stored in the subrefector database in real time to compensate for the gravitational deformation of the main refector. Reference [129] establishes a temperature database based on transient temperature feld simulation and then corrected the simulated temperature feld extracted from the temperature database by using real-time measurement data and the lookup table method. Combined with the fnite element model of the telescope, the thermal feld can be calculated in real time and used for thermal deformation compensation. Reference [130] proposes a method for fast match and call of the compensated database, which guarantees the accuracy by matching the root mean square error, similarity regions, and critical regions of the data. Reference [131] used a thermal deformation database determined by numerical simulation according to the observation schedule in advance to compensate for thermal deformations in real-time.

Performance Assurance Technology under Rapidly
Varying Loads

Approximate Dynamics Modeling.
For radio telescopes in the natural environment, wind disturbance has become the most complex and difcult problem in various environmental loads due to its randomness and time variation. Te wind load will not only cause the deformation of the antenna but also cause the vibration, resulting in unstable pointing and poor observation performance. Terefore, it is necessary to study how to prevent wind disturbance. To describe the pointing deviation caused by the deformation (fexibility and rigidity) of the refector surface under disturbance, estimate the pointing direction of the telescope, and facilitate the design of a control system with good control performance, it is necessary to establish a high-precision dynamic model of the antenna structure. Te dynamic modeling of the antenna could be established with theoretical modeling. Teoretical modeling can be divided into rigid modeling, fexible modeling, and rigid-fexible hybrid modeling. For example, Reference [132] used rigid modeling to model the 110 m GBT. Te advantage is that the model order is small. However, with the increase of the antenna aperture, the fexible deformation of the antenna structure cannot be ignored. However, rigid modeling cannot refect the antenna fexibility information; therefore the actual control efect of the designed controller is not so ideal. Reference [133] used fexible modeling to perform fnite element analysis on large aperture antennas, which can estimate the structural deformation. Furthermore, it also carried out a modal analysis on the antenna based on fnite element analysis, which can select the modal retention that has a greater impact on the antenna performance, carry modal polycondensation, reduce the degree of freedom, and combined it with the rigid body model to obtain the overall dynamics model of the antenna to facilitate the further design of the controller. Reference [134] establishes a dynamic model of the antenna structure based on the model superposition method, and the dynamic model is poly-condensed according to the infuence degree of each vibration order on the pointing, and further established a rigid-fexible hybrid model. Te established dynamic model of the large-scale refector antenna for pointing control is not only benefcial to the design of the controller but also can refect the infuence of fexible deformation on the pointing.

Parameter Identifcation of Servo Control System.
Large-scale radio telescopes are complex in structure, and due to the nonlinear efects such as wind load and friction, it is difcult to carry out accurate theoretical modeling. In this case, system modeling can be used to establish an accurate system model. Te model uses the parameter identifcation algorithm to obtain important parameters that can characterize the behavior of the system, establish a model that can imitate the behavior of the real system, and use the current measurable system input and output to predict the future evolution of the system output, which is advantageous to the design of the controller. Classical system identifcation methods include the step response method, impulse response method, frequency response method, correlation analysis method, spectral analysis method, least square method, and maximum likelihood method. Te parameter identifcation of a servo control system mainly includes two aspects, that is, the identifcation of parameters such as internal motor inductance and the identifcation of external mechanical parameters including load inertia, friction damping, and shaft stifness. For example, Reference [135] used the method of online identifcation of the system to collect the input and output data of the control object in real time. Te system mathematical model is established by using the fourth-order controlled autoregressive moving average model and selecting the quadratic index to perform least squares ftting on its parameters. Reference [133] presents the process of deriving the velocity loop model from feld test data. It includes descriptions of antenna white noise testing, test conditions, input and output signals, open-loop testing, closed-loop testing, instrument setup, and system identifcation testing. Te analytical context for data description and data processing is also included. Furthermore, the correlation between the antenna model and its elevated position and the diameter of the dish antenna is discussed, and fnally the analytical model and the identifcation model are compared.

Wind-Resistant Servo Control Algorithm.
For largediameter and high-frequency antennas, the pointing error caused by wind disturbance is still one of the most difcult and insurmountable problems due to its randomness and time-varying. To improve the pointing accuracy of the antenna and efectively suppress the pointing error caused by wind disturbance, it is necessary to study the efective error analysis and control compensation method according to its characteristics, and design suitable servo control algorithms and servo control systems. For the LMT 50 m antenna, Reference [136] compares the antiwind disturbance capabilities of the antennas with four control algorithms, including PI-PI, PI-LQG, LQG-PID, and LQG-LQG algorithms, and the simulation results show that LQG-LQG has the best control efect. Reference [137] aims at the control method of radar antenna to suppress wind disturbance, under the premise that the antenna is regarded as an ideal rigid model, the design of the controller is divided into two independent stages according to the design theory of disturbance observer and the principle of fuzzy control, that is, performance design phase and interference suppression phase. Te results show that the fuzzy PID controller has better dynamic performance than the traditional PID controller, and the introduction of a disturbance observer can efectively reduce the pointing error caused by wind disturbance for the infuence of wind load on pointing.
Reference [138] deduced uncertain models suitable for diferent models for diferent types of beam waveguide antennas, and an H-infnity controller is designed in combination with the model. Te results verify the efectiveness of the controller design method in suppressing wind disturbance in diferent antennas. Te pointing performance based on the H-infnity controller is slightly better than the error compensation efect based on the LQG control. In addition, the frequency of gusts is much higher than the response frequency of the control system. When the wind direction and wind speed change, the control system has no time to adjust its control strategy, so the control system cannot compensate for the pointing error caused by wind disturbance efectively. Terefore, a threshold-based control strategy will be adopted in the future to be equivalent to the interference of the wind on the antenna for a period of time [7,139]. In other words, for the wind over a period of time, an adjustment amount is used. In some long-term high wind felds, the active surface (under the threshold) and the servo dual-system control strategy can be used to achieve stable control.

Wind-Resistant Pointing Error Compensation Based on
Wind Prediction. Wind disturbance has the characteristics of short-period pulsation, and the estimation process or measurement process of the measured wind has a time delay, which means that when the wind acts on the antenna refector surface, the control system cannot make real-time adjustments according to the received interference signal. Tis requires reasonable prediction of incoming wind information and calculation of the adjustment amount in advance. An adjustment amount can be calculated at every moment, but it still takes time for the control system to act on the antenna to adjust. Terefore, it is necessary to calculate an optimal adjustment amount (threshold) for a certain period of time, and the antenna should be adjusted in advance to reduce the infuence of wind disturbance. Te establishment of the wind speed prediction model provides the possibility to compensate for the antenna pointing error in real time. Diferent from the traditional compensation strategy based on error feedback, predictive control is a state feedback compensation strategy that reduces future disturbances by generating compensation in advance. At the same time, it is a complex process to reduce the pointing error caused by the random wind disturbance throughout the antenna rotation angle generated by the compensation efect. To obtain the desired pointing performance, it is necessary to realize the matching between the compensation elements of each link. Te specifc manifestations are matching between prediction parameters and controller parameters (to realize the optimal selection of adaptive control parameters under variable-time forecast wind speed), matching between predicted values and performance (to avoid cumulative prediction errors introduced by predicted values), and time margin matching with system time delay (to avoid compensation lag caused by long-time iteration of complex control algorithm). Reference [140] verifes that it is possible to accurately estimate the radial and tangential velocity components of the wind feld in a realtime manner, and verify the validity of the steady-state wind speed prediction model.

Conclusion and Future Challenges of Electromechanical Coupling
4.1. Conclusion. Tis paper takes several major radio astronomy discoveries in recent years as the starting point and shows the key role that radio telescopes play in these major discoveries. From the major VLBI networks, the authors can fnd a global development trend in radio telescopes. Te guarantee and improvement of the performance of a singledish radio telescope is the foundation of observation, and the network observation of radio telescopes between countries is also a method that can maximize the performance of telescopes. Te authors then describe the development of radio telescopes built in the world in recent years and the development of telescopes or telescope arrays that are under construction or to be constructed. It can be seen that China has built a relatively large number of single-dish radio telescopes in recent years and has also actively participated in the construction of some international radio telescope arrays, such as the SKA. In Europe and the United States and other countries, the focus has been on the construction of telescope arrays in recent years. At the same time, single-dish antennas have been transformed, upgraded, and networked for observation. Finally, the authors summarize the application of electromechanical coupling technology in radio telescope design and challenges from three perspectives of electromechanical coupling modeling and design, performance assurance technology under slowly varying loads, and performance assurance technology under rapidly varying loads.

Future Challenges of Electromechanical Coupling.
In future, with the development of radio telescopes in the direction of higher apertures or higher frequency bands, the author believes that the following technical challenges should be addressed.

Uncertainty Analysis and Processing of Radio Telescope
Errors. For electronic equipment with integrated structure and function as a radio telescope, the mechanical structure is the carrier of its electrical performance, the device performance is the guarantee of its electrical performance improvement, the material parameters are the constraints of its electrical performance breakthrough, and the environmental factors are obstacles to the stability of its electrical properties. With the development of radio telescopes in the direction of high frequency and high gain, the infuence of the above four factors on performance is becoming increasingly obvious, and there is still a phenomenon of mutual coupling between these four aspects. At the same time, in engineering, there are uncertain factors such as manufacturing and installation errors, device performance temperature drift, material parameter changes, and environmental load fuctuations. If these uncertain factors are treated as deterministic information, this might lead to contradictory or very unreasonable results. Tus, in the follow-up research process, a certain uncertainty analysis method can be used for the error. On the basis of sorting out the infuence mechanism of uncertainty factors on the telescope, a coupling model and performance guarantee methods that are more in line with the actual situation are proposed.

Wind Regulation Technology for Large Radio Telescope.
Most of the existing radio telescopes use passive methods such as wind resistance control or the construction of a radome to reduce the impact of natural wind. Wind resistance control is likely to be inefective in the high-frequency wind or strong wind conditions. Te radome is expensive to build and afects the electrical performance, making it impractical for large radio telescopes. Tus, the authors need to start from another point of view to weaken the infuence of wind by actively reducing the wind speed or changing the wind direction, which is called wind regulation technology. Wind regulation is to design of a wind regulation device by using the topographic and geomorphological features of the telescope site without changing the state of the telescope to achieve the purpose of actively reducing the near-feld wind speed of the antenna and guiding the wind fow to bypass the antenna, thereby reducing the wind disturbance of the radio telescope and increasing the high-frequency stable observation time of the telescope. Te idea of wind feld regulation is to weaken the wind energy, reduce wind speed, and change the wind direction to achieve the efect of stabilizing observation and increasing operation time. Te difculty of regulation lies in how to regulate the wind at a height of 100 meters. On the one hand, the focus of wind feld regulation is to design the device structure with a good shading efect and great consumption of wind energy. On the other hand, it is the synergy with terrain-based wind prediction and antenna servo system compensation.

Application of Low-Cost New Materials in Antenna
Structure Design. At present, radio telescopes basically use metal materials to build the mechanical structure of the antenna. Te high density of the metal material itself leads to a large self-weight of the antenna, and there is a natural structural limit. New materials could be used to design the structure of telescopes. Because of its excellent mechanical properties, carbon fber composite material is a composite material with resin as the matrix and carbon fber as reinforcement. It has the characteristics of lightweight, high modulus, and low thermal expansion coefcient. At the same time, this material itself has electrical conductivity, so the antenna made of carbon fber composite material can not only ensure the electrical performance of the antenna, but also light in weight, strong in shock and vibration resistance, and can withstand harsh environments such as high and low temperature, thermal vacuum, and others. Carbon fberreinforced resin composite materials have the characteristics of low expansion, lightweight, and high strength. For highprecision antennas serving in extreme site environments, carbon fber composite materials are very ideal refector materials. Te surface of carbon fber composites needs to be coated with metal coatings by magnetron sputtering to achieve an efcient refection of electromagnetic waves. On top of the metal coating, the magnetron sputtering method is used to coat the dielectric layer to protect the metal layer, which not only plays the role of antioxidation and anticorrosion but also ensures the thermal performance of the refective surface during operation. Epoxy resin coating is currently the most widely used, but the anticorrosion ability of pure epoxy resin is limited. Adding fllers to epoxy resin coatings to prepare epoxy resin-based composite anticorrosion coatings is an efective way to improve the anticorrosion ability of epoxy resin coatings. Silicon nitride has a low dielectric constant and dielectric loss, as well as a series of excellent properties such as high strength, high hardness, electrical insulation, thermal shock resistance, corrosion resistance, and wear resistance and is recognized as the most promising new wave-transmitting material. Te previously mentioned new materials have broad application prospects in radio telescopes, but due to factors such as cost, they are rarely used in antennas at this stage.

Performance Assurance Technology under the Simultaneous Action of Transient Wind and Temperature Loads.
Due to its special outdoor working environment, largeaperture radio telescopes are inevitably afected by external loads such as gravity, temperature, and wind loads. Te infuence of these three loads on the surface accuracy of the refector and the electrical performance has been analyzed in the previous section separately, but the actual service environment of the antenna cannot have only a single load. For example, the temperature uniformity of the refective surface will afect its structural deformation due to the action of the wind feld. Terefore, the infuence mechanism of the unsteady wind and temperature felds on the electrical performance of large antennas should be studied in the future.

Moon-Based Radio
Telescope. Te aperture of the radio telescope on the Earth has reached its limit due to the infuence of gravity. Te gravity on the moon is only 1/6 of that on the Earth. Naturally, it is possible to consider establishing a radio telescope with a large aperture (120 m and above) on the moon, for example, the Lunar Crater Radio Telescope mentioned earlier in this article. However, the telescope is a mesh antenna, and it can only perform lowfrequency observations after completion. If high-frequency signals are to be received, then a solid panel must be used to build the antenna. In 2021, China and Russia jointly issued the Joint Statement of the China National Space Administration and Roscosmos on Cooperation in the Construction of an International Lunar Research Station. Tis means that China and Russia will work with other international partners to jointly build an international lunar scientifc research station, of which the lunar astronomical observatory is one of the key points. Terefore, it is foreseeable that the moonbased radio telescope will be a major research hotspot in the future. Among them, lunar morphology analysis, telescope selection, energy supply, environmental protection technology, thermal control technologies (heat dissipation on lunar day and heat preservation on lunar night), assembly, measurement and commissioning calibration, telescope operation and maintenance, and remote monitoring will all be areas for breakthroughs.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest
Te authors declare that they have no conficts of interest regarding the publication of this paper.