This paper introduces a design and optimization computer simulation program for autonomous hybrid PV/wind/battery energy system. The main function of the new proposed computer program is to determine the optimum size of each component of the hybrid energy system for the lowest price of kWh generated and the best loss of load probability at highest reliability. This computer program uses the hourly wind speed, hourly radiation, and hourly load power with several numbers of wind turbine (WT) and PV module types. The proposed computer program changes the penetration ratio of wind/PV with certain increments and calculates the required size of all components and the optimum battery size to get the predefined lowest acceptable probability. This computer program has been designed in flexible fashion that is not available in market available software like HOMER and RETScreen. Actual data for Saudi sites have been used with this computer program. The data obtained have been compared with these market available software. The comparison shows the superiority of this computer program in the optimal design of the autonomous PV/wind/battery hybrid system. The proposed computer program performed the optimal design steps in very short time and with accurate results. Many valuable results can be extracted from this computer program that can help researchers and decision makers.
Energy is considered as the most important factor in the modern life. In remote areas, especially in the desert of vast land of Saudi Arabia, the energy is valuable because the electric utility is not available and the transportation of fossil fuels required for small convention electric generation plants is very hard and expensive. There are wide areas in Saudi Arabia, especially in remote areas, that have no access to the electric utility. It is not economical to extend the national electric grid to all these wide remote areas. Most of these areas rely on conventional electric energy sources like diesel generators for their electric power supply. However, this conventional generation depends on the availability of fossil fuel that usually is quite expensive. Besides that, the engines usually operate at low efficiency due to the typical loads in remote areas that vary considerably during the day and night [
In a DC-coupled configuration, the different energy sources are connected to a DC-bus through appropriate power electronic converters interfacing circuits. The DC sources may be connected to the DC-bus directly if appropriate or may be a DC/DC converter is required to regulate the DC voltage. If there are any DC loads, they can also be connected to the DC-bus directly or through DC/DC converters to achieve appropriate DC voltage for the DC loads. The DC coupling scheme is characterized by simple construction and control and it has low efficiency and reliability: low efficiency because all power to the load should go through the inverter and low reliability because if the inverter does not work, the whole system will shut down. To avoid this situation, it is possible to connect several inverters with lower power rating in parallel, in which case synchronization of the output voltage of the different inverters, or synchronization with the grid, if the system is grid-connected, is needed which increase the system reliability but it will increase the cost. A proper power sharing control scheme is also required to achieve a desired load distribution among the different inverters [
Instead of connecting the entire hybrid energy sources to just a single DC or AC-bus, as discussed previously, the different sources can be connected to the DC or AC-bus of the hybrid system depending on the output of each of them. As a result, the system can have higher energy efficiency and reduced cost. On the other hand, control and energy management might be more complicated than for the DC- and AC-coupled schemes [
Choosing of the appropriate configuration depends on the type of output power of the most of the generation and loads. If most of the generation and some loads are DC, it is better to use DC-bus coupling or to use AC-bus coupling if the other case is valid. If the major power sources of a hybrid system generate a mixture of AC and DC power, then a hybrid-coupled integration scheme may be considered [
The proposed hybrid energy system.
In this paper, a hybrid RE system (HRE) in AC-bus connection is designed by using wind/PV/battery storage as shown in Figure
Wind energy is a form of solar energy produced by uneven heating of the earth’s surface. Unlike any conventional power source, wind power is less predictable. Although wind power source is less predictable than the solar power, it is typically available for more hours in a given day. Wind resources are influenced by the type of the land surface and the elevation of the land surface. Generally, if the land is in high elevation then it is good for wind energy conversion and vice versa. Also the generation of energy from WTs increasing considerably with the increasing the height of the tower but it increases the cost of the system. Also, the open area places are suitable for installation of wind energy systems. Since the wind speed is extremely important for the amount of energy a WT can convert to electricity (the power content of the wind varies with the cube of the wind speed), the choosing of wind energy site based on the wind speed data is very important. Many literatures have been done to choose the best possible site from many available sites [
The vertical wind speed gradient can be obtained from the following equation:
Due to the effect of wind speed on the generated power from WT, the availability of wind has a great influence on capital investment and on profitability of the entire systems, where, the characteristics of wind at the plant site play a significant role in the price of kWh generated from the whole system. So, the selection of suitable WT for certain site needs a lot of calculations and details to be studied and optimized.
The actual WT output power with the wind speed is shown as
The speed of the wind is continuously changing, making it desirable to describe it by statistical methods. One statistical quantity which we have mentioned earlier is the average or arithmetic mean [
Consider
A method depending on the accurate statistical analysis for obtaining Weibull parameters has been used in this computer program [
The capacity factor of the WT in certain site can be obtained from the following equation:
The orientation of the PV array against the movement of the sun determines the intensity of the sunlight failing on the modules surface and, therefore, it will affect the system power output. The tilt angle and the azimuth angle are the two parameters that need to be considered. The tilt angle is the angle between the plane of the PV array and the horizontal whereas the azimuth angle is the angle between the plane of the PV array and due south (or sometimes due north when in southern hemisphere). The orientation will be facing south for the case of Saudi Arabia due to its location in the north of the equator. The PV array surface should be positioned in a way that it is aimed directly perpendicular to the sun’s rays. This will capture the maximum amount of sunlight to be converted into electricity. This tilt angle and orientation can be easily achieved using a tracking system that follows the sun’s trajectory at a particular time and day. However, a tracking system is costly and requires high maintenance. Therefore, various studies have been conducted on the optimum tilt and orientation angle for fixed surfaces. In this research, PV array is assumed to have optimum monthly tilt angle. However, the optimum monthly tilt angle chosen should provide maximum power output from the system during the month. The tilt angle can be fixed at a certain angle all year round, seasonally or monthly changed. The monthly optimum tilt angle will be used in this study because of its simplicity and its economic benefits. There has been some earlier literature that suggested positioning PV arrays to face south at certain tilt angle for sites in the north of the equator [
The position of PV array is defined by its tilt angle and orientation, expressed as the azimuth angle;
PV module facing South and tilted at angle
The angle of declination
Monthly collectable radiation on a tilted surface for a given month (
Equation (
Monthly average daily extraterrestrial radiation on a horizontal surface (
The cell temperature can be obtained from the following equation:
The calculation of optimum PV area, PVA can be obtained from energy balancing between the load and the generation from PV system. PVA may be different than the average value due to the approximation and due to the discarded surplus power output from the PV system in case of full battery and the output greater than the load requirements.
Battery storage system has been used to store the extra generated energy than the load requirements and supply the load with the energy when the generated power is not sufficient for the load. The maximum limit that the battery can store is
Consider
Power electronics converter has been used in hybrid system to convert DC power to AC and from AC to DC to be suitable for the bidirectional power flow. Modern PWM converters have improved efficiency with typical value ranging from 90% to 95% with investment cost of $800/kW and $750 for replacement, $8 for annual cost and maintenance, and 15 years life time [
The inverter power (
Consider
Many researches introduce economical techniques to determine the cost of the generated kWh depending on many assumptions. Some detailed techniques [
The initial capital cost of the hybrid system includes the whole components of the system including the civil work, installation cost, and electrical connections and testing. The price of each component of the hybrid system and its life time and efficiency are shown in the Table
The detailed economic factors of each component of the hybrid system [
Item | Price $ | Replacement |
OMC $ | Lifetime |
Scrap |
Number of replacements | Salvage times |
---|---|---|---|---|---|---|---|
WTG, kW | 800 | 640 | 10 | 20 | 20 | 1 | 2 |
PV, kW | 1140 | 1026 | 5 | 30 | 10 | 0 | 1 |
Inverter kW | 700 | 630 | 5 | 15 | 10 | 1 | 2 |
Batteries |
200 | 160 | 10 | 5 | 20 | 5 | 6 |
Civil work |
20% | 1 | 30 | 20 | 0 | 1 | |
Civil work |
40% | 1 | 30 | 20 | 0 | 1 |
The replacement cost of the wind generator, the batteries, and the inverter have to be included in the cost analysis of the hybrid system. Considering the inflation rate of component replacements (
The present value of the operating and maintenance cost, OMC, is representing the maintenance cost all over the life time of the project for all components in the time of installation. This is the sum of all yearly scheduled operation and maintenance costs. OMCs include such items as an operator’s salary, inspections, insurance, and all scheduled maintenance. Some researchers used a fixed percentage of the total cost of the system for maintenance such as in [
The present scrap value, PSV, can be obtained from (
The new proposed computer program (NPCP) has been designed to perform sizing and optimizing the whole component of the hybrid system. The NPCP used the model of each component explained in the previous sections. This computer program is flexible which is not available with market available software. The results obtained from NPCP have been compared with HOMER software. The results obtained from NPCP are very near with the results obtained from HOMER software which prove the superiority of this computer program. NPCP has been designed in modular form and has been written in visual FORTRAN software. The block diagram of the proposed computer program is shown in Figure
Summarized block diagram of the proposed computer program.
The required data for the computer program have been summarized in the following points. The data of the market available WTs such as the cut-in wind speed, The data of the market available PV modules such as module efficiency, module area, output voltage and current, nominal operating temperature, and lifetime. The data of market available batteries such as the rated capacity, depth of discharge, lifetime, charging and discharging efficiency, and operating voltage. Hourly wind speed for available sites and hourly solar radiation for available sites. Site locations and elevations. Economic data such as the price of each component, its replacement price, operation and maintenance price, scrap (salvage) price, interest rate, inflation rate, and life time proposed in this paper.
The purpose of this subroutine is to determine the Weibull, scale and shape parameters,
The flowchart of the first subroutine for determining the Weibull parameters.
The purpose of this subroutine is to determine the capacity factor,
In this computer simulation program, it is assumed that the tilt angle of the PV modules should be changed monthly to the best tilt angle. The monthly best tilt angle should be calculated from (
The flowchart of the modifying solar radiation on a tilted surface.
Energy balance part of the proposed computer program is the main part. To compute the optimum number of WTs and optimum solar module area required, an energy balance between the loads and the output from wind energy and PV systems must be made. Many salient factors will be calculated from this part such as optimum number of WTs, optimum area for PV modules, optimum capacity of the battery, optimum size of the inverter, the yearly energy, the energy contribution from PV and wind separately, many other valuable data can be extracted from this part of the computer program. if if if if
This part of the computer program uses the average number of WTs, ANWT, and the average area of PV modules to calculate the energy produced from each supply and check if these values are just enough for the load requirements or if the program should increase it. The total energy generated should be combined and compared with the load power. In case the generated power is greater than the load requirements, the surplus power will charge the battery. In the other case, where the generated power is less than the load requirements, the defect power will be obtained from the battery system. The highest value of the accumulation of the power in the batteries determines its capacity as will be explained in the following logic. The total energy generated should satisfy the load requirements; otherwise, the size of the wind energy or the size of PV system should change by certain value. In the other case, if the total energy generated is greater than the load requirements by considerable limit, the size of the wind energy or the size of PV system should be reduced by certain value. The cycle starts again until the generated energy just satisfies the load requirements through the year. The logic behind the energy balance part is shown in Figure
The total energy generated from the WT is the summation of power generated from the wind through the year and can be obtained from the following equation:
The flowchart of energy balance subroutine in the new proposed program.
If
If
If
If
If
If
The maximum value of energy that the battery can discharge from its full charging condition or the maximum value of energy that the battery can charge from its minimum charging condition is the maximum value of
A detailed economic technique has been explained in economic analysis section. This technique is discussed in details and it takes into consideration all economic factors to get an accurate price for the generated kWh to easily compare between different options. The output results from energy balance program should be transferred to this subroutine such as the optimum number of WT, NWT, the optimum PV array area, PVA, and the required capacity for the batteries,
The flowchart of the economic analysis model.
Hybrid optimization model for electric renewable (HOMER software) software is one of the most famous softwares developed by National Renewable Energy Laboratory (NREL). This system can be used in HRES with backup like batteries or gas turbine autonomous or interconnected with electric utility. The HOMER software also can optimize the system with fuel cells and hydrogen tanks and electrolysis. This software can handle systems with MPPT tracking for PV system and it can compensate the solar radiation of the horizontal surface to be as the radiation on sun trackers with one or two axes or it can modify the solar radiation to become on the daily or monthly or yearly best tilt angle. Also the input wind speed can be modified to the hub height of WT. This software has a detailed economic calculation which takes account of all economic factors. But the detailed calculation is not shown and it is just a black box with a limited flexibility in changing the input data without an ability to check and change the economic calculation technique. Also, there is one more shortage with this software which is that the system cannot provide the user with the optimum sizing for each component but it can only use the available options that the user introduces to the computer to choose one of these options. So, this program can be used to compare with the results obtained from the proposed computer program in its optimum results. The simulation by using HOMER software is the same as the one used in NPCP.
Many computer simulation programs may be used to do the same job as HOMER software but with less flexibility and they are not famous and effective like HOMER software. These market available computer softwares are HGYBRID2, HYDROGEMS, HOGA, TRANSYS, SOLSIM, SOMES, RAPSIM, ARES, INSEL, and HYBRIDS. A detailed comparison between these software packages is shown in [
Simulation has been carried out by using the NPCP and HOMER software to check and validate the results obtained from the new proposed computer program. In our new proposed computer program, the program can design the optimum size of each component depending on the minimum cost of energy. But the main limitation of HOMER software is that this feature is not included in the software however HOMER software needs the user to introduce many possibilities for each component to select the best option from these possibilities. So, the lowest price obtained from HOMER software is not the optimum solution but it will be the best possibility from the available possibilities entered to each component as a data. The optimum solution from the NPCP will be entered to HOMER software to compare the results.
The simulation has been done in the beginning with one of the best sites in wind energy which is Dhahran site where its average wind speed is 4.7 m/s at 10 m height from ground level. This site is located in the eastern part of Saudi Arabia at the coast of Arabian Gulf in 26° 18′N and 50° 8′E. One hundred WT data are introduced to the program to choose the best WT suitable for this site and the optimum contribution from wind and PV. Three PV modules have been introduced to choose the best one for this site. The load profile for small village known as (ADDFA) in the north of Saudi Arabia were collected over several years and used as a load profile in our analysis. The load data for this village is multiplied by 10 to compensate the future extension and to design a system that may supply several remote villages. The penetration ratio of wind and PV in feeding the load is changing in step of 1% and the optimum value will be selected according to the minimum price for kWh. The life time has been chosen to be 30 years, interest rate for Saudi Arabia is 2% [
Comparison between the results from NPCP and HOMER for Dhahran site.
|
|
LEC | Energy contributions from wind | Energy contributions from PV | |||||
---|---|---|---|---|---|---|---|---|---|
NPCP | HOMER | NPCP | HOMER | NPCP | HOMER | NPCP | HOMER | NPCP | HOMER |
5.185 | 5.18 | 3.11 | 2.99 | 10.5 | 10.7 | 60 | 63 | 40 | 37 |
The detailed results for Dhahran site in HOMER software are shown in Figure
The energy and cost contributions for each component for Dhahran site in HOMER software and NPCP.
Item | HOMER software | NPCP | ||||||
---|---|---|---|---|---|---|---|---|
Wind | PV | Battery | Converter | Wind | PV | Battery | Converter | |
Energy contribution % | 63 | 37 | 60 | 40 | ||||
Cost contribution in initial cost ($ millions) | 16.28 | 21 | 6 | 3.5 | 14.4 | 22.006 | 7.929 | 4.792 |
Cost contribution in initial Cost % | 34.8 | 44.9 | 12.8 | 7.5 | 29.3 | 44.8 | 16.14 | 9.8 |
The cash flow summary by component for Dhahran from HOMER software.
The same results obtained from NPCP which are shown in Table
The cash flow summary by component for Dhahran from NPCP.
It is clear from the results of HOMER software and NPCP that the price of kWh generated does not change too much, only 2% deviation in these two values. NPCP predicts that the best contribution is 60% from wind and 40% from PV but HOMER software predicts that the contribution is 63% from wind and 37% from PV.
Table
The cost contributions for each cost type for Dhahran site in HOMER software and NPCP.
Item | HOMER software | NPCP | ||||||
---|---|---|---|---|---|---|---|---|
IC | REP | OMC | Salvage | IC | REP | O & M | Salvage | |
Cost contribution in total cost ($ millions) | 46.78 | 20.16 | 18.644 | −6.92 | 49.13 | 18.09 | 8.81 | −2.011 |
Cost contribution in total cost (%) | 59.5 | 25.63 | 23.7 | −8.8 | 66.4 | 24.4 | 11.9 | −2.71 |
Same results can be obtained from NPCP where the capital cost is $49,126,780 (66.4%), $18,089,010 (24.4%) for replacement costs, $8,809,040 (11.9%) for operating and maintenance cost, and $−2,011,168 (−2.71%) for salvage price.
Hybrid RE system can economically feed loads in remote areas in Saudi Arabia and any other places in the world. Hybrid renewable energy system does not need fuel transferee and it reduces the pollution associated with generation from conventional generating stations. The proposed hybrid system uses wind and PV to feed the loads with help of batteries and power electronic converters. The proposed system used the hybrid AC and DC coupling to increase the efficiency and reliability. A new proposed program has been introduced in this paper to optimally design each component of the system. This proposed program has a detailed analysis for each part of the hybrid energy system. This computer program produced accurate results in a flexible fashion. This computer program does not need predefined sizes of each component as many market available softwares to choose the best one of these combinations but it optimally designs each part of the system. The computer program can select the best site from many available sites and suitable WT and PV type for each site and the optimal contribution from wind and PV system. The results from this computer program have been compared with the results obtained from HOMER in Dhahran site in Saudi Arabia with real load data for remote location. The comparison results show the superiority of the proposed computer program.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors acknowledge the College of Engineering Research Center and Deanship of Scientific Research at King Saud University in Riyadh, Saudi Arabia, for the financial support to carry out the research work reported in this paper.