A Novel Technique for High-Performance Grid Integrated with Restricted Placement of PV-DG considering Load Change

,


Introduction
Tere is no doubt that the electricity demand has reached a high mark due to the progress of human civilization.Terefore, the structure of the traditional electrical distribution network (EDN) has been modifed by integrating the distributed generation (DG).Photovoltaic (PV), wind turbine (WT), and biomass (BM) are popular renewable DG technologies [1,2].Furthermore, there are technical, fnancial, and environmental advantages to integrating DGs into an already-existing EDN [3,4].Because of the many benefts of this technology, EDN was being integrated with DG technology.However, placing DG units, particularly large ones, in certain locations may have detrimental efects on the EDN, such as voltage increase, instability, and circuit breaker failure because of the bidirectional power fow [5].Finding the ideal size and location of DG units is crucial for achieving the benefts of DG and avoiding any potential drawbacks [6].Several researchers have presented eforts in literature with the main aim of harnessing the maximum benefts of optimal DG placement with a defned specifc capacity in diferent scenarios with diferent techniques.
Numerous optimizations have been ofered in literature, by optimizing their place with a defnite specifc capacity to obtain the main purpose of maximizing the profts expected from connecting DGs to EDN [7].A Pareto optimal concept-based novel multiobjective quasi-oppositional grey wolf optimizer (MQOGWO) algorithm has been proposed in [8].To fnd the optimal location and capacities of DGs, the stud krill herd algorithm has been implemented in [9].Te author in [10] used the self-adaptive levy fight-based black widow optimization algorithm to obtain the DG units' optimum location and size.Te PV-DG sizing and locations using the loss reduction sensitivity factor and biogeographybased optimization algorithm in a distribution network are implemented in [11].With the main objective of reducing the cost of connecting a new renewable DG unit to public medium voltage distribution systems, the author in [12] provided an analytical approach for efectively distributing renewable distribution generators and capacitor banks.
To improve a certain target or a grouping of purposes, the main advantages that have been researched in the literature include a reduction in power losses and grid performance enhancement [13][14][15][16][17][18].A voltage sensitivity index and a voltage profle improvement indicator (VPII) were adopted for DG placement to minimize power loss [19].In [20], a hybrid binary particle swarm optimization with ant lion optimizer (BPSO-ALO) is used to defne ideal system reconfguration while simultaneously allocating and sizing DGs to decrease power loss and enhance voltage profle.Te author in [21] introduced a diferent optimization technique that combines the benefts of network reconfguration options with the DG allocation to lessen technical losses and improve the bus voltage profle in EDN.In [22], the enhancement of the voltage profle by using a genetic algorithm is implemented.Te equilibrium optimizer (EO) is used to realize the voltage profle enhancement by reconfguring with DG placement in the distribution system [23].Choosing the best placements and sizes for renewable DG units, considering the variation of load demand had been presented in works [24,25]; a multiobjective optimization problem was constructed to identify the best locations and sizes of renewable DG units by considering the hourly change of load demand and the generation profle of renewable sources in [26].Te author in [27] introduced a methodology for DG unit placement in the distribution system that considers changes in the active power load.Te evolutionary programming (EP) is used for fnding the optimal placement and sizing of multidistributed generation (DG) units including diferent load models as presented in [28].Te grey wolf optimizer (GWO) and loss sensitivity factor (LSF) are used to determine the best location and size of DG in distribution systems at diferent load levels (light, nominal, and peak) [29].A new imperialistic competitive algorithm ICA-based algorithm has been introduced in [30] to obtain the optimal location and size of DG units, reduce losses, and enhance voltage stability considering load variations.Te authors in [31] proved that PV-DG fragmentation was one of the best solutions to enhance the grid performance under the restricted placement of PV-DG but without considering the load change in their study.In this research, DG grid integration was studied but with a predetermined PV placement and size which on the availability of the land plot, conditions of the proper location, cost budgetary, and a need for a modifcation considering the realistic change in load profle.Te purpose of this study is to enhance the grid's performance with previously mentioned conditions using a novelty technique that breaks up PV-DG unit into smaller units and connects them to every bus of the grid through distribution lines.Te paper is organized as follows.In Section 2, the proposed method is discussed using IEEE 15 bus system modeling based on the MATLAB environment.In Section 3, the research method is described.Te results and discussion are presented in Section 3. Finally, the conclusion is presented in Section 4.

Proposed Method
2.1.Problem Defnition.Grid performance enhancement is expected using DG integration.PV-DG grid integration is used in this research considering realistic conditions that represent the problem of this research.Tese conditions are predetermined PV size and placement based on land plot availability, proper site conditions, cost budgetary, and available transfer capacity (ATC) of the existing EDN for how far the existing transmission lines can bear the PV-DG surpluses.Te research also considers the change in the load profle.

IEEE-15 Bus
System.Te IEEE 15-bus, 11 kV redial distribution system was selected to be the test bed in this study.Te test bed is selected because it has characteristics of a long grid lightly loaded requiring the application of voltage regulators to satisfy voltage standards, which is similar to structures of medium voltage systems in Egypt.Te total load of the IEEE-15 bus system is 1226.4KW and 1251.11KVAR.Te distribution line and load data of the test system are given in [25].Te IEEE-15 bus system is simulated in MATLAB/Simulink environment to obtain all buses per unit voltages, power fow data, and system efciency for performance judgment.

PV-DG
Connected to the System.Te PV-DG unit is integrated into the IEEE-15 bus system test bed with a restricted specifc size and location to study the grid performance and how to enhance it as shown in Figure 1.Te size of the PV-DG unit 400 kW is selected to be in the range of 25% of the system load which represents a huge percent value of the system under this study.Also, the PV-DG unit location is selected to be beside Bus 1 as a restricted placement in which there is high incident solar radiation and the least weather factors.

Proposed
Techniques.Tis research proposes two techniques.First, by connecting the whole size selected PV-DG unit through the transmission line to the optimum bus considering the available transfer capacity (ATC) of the test bed and load change, this technique is called here through the paper "PV-DG without fragmentation."Second, by breaking up PV-DG unit into smaller units to be connected to all buses together overcoming the problem of the ATC of the network, this novelty technique is called "PV-DG with fragmentation."

First Technique: Optimal Bus Connected to PV-DG Unit without Fragmentation.
Te PV-DG unit without fragmentation adjacent to Bus 1 is connected to each bus to determine the optimum bus taking the existing EDN's available transfer capacity (ATC) into account.Only the available buses 1, 2, 3, 4, 6, and 11 can be connected to the PV-DG unit using distribution lines as indicated in Figure 2. Tese buses have interconnected distribution lines that can support the PV-DG surpluses.
Tree diferent scenarios were considered to study the EDN performance of integrating the PV-DG without fragmentation and determination of the optimum bus in which it is to be connected through a distribution line with diferent load schemes: Te performance analysis of the EDN is carried out in this study by studying the voltage deviation obtained from each bus when the PV-DG unit without fragmentation is inserted and by determining the resulting efciency of the EDN.Te equations used to obtain the two parameters are as follows: where V dev = voltage deviation, V max = maximum voltage obtained when PV-DG unit was at bus i, and V min = minimum voltage obtained when PV-DG unit was at bus i.And the reason for using equation ( 1) to minimize the voltage dip percent.where c = efciency, P L = power losses when the PV-DG unit was at bus i, and P in = power demanded when the PV-DG unit was at bus i.

Case A: At Standard Load (Scheme 1). A MATLAB/
Simulink environment is used to simulate the IEEE-15 bus In this case, Table 3 shows the results of all bus voltages when the PV-DG unit was connected to each available bus.It is evident from the table that without PV-DG unit fragmentation, bus no.11 gives the minimum voltage deviation and hence optimal voltage profle.Also, the results shown in Table 4 indicate a better enhancement in the system efciency of 96.6% when connecting the PV-DG to bus no.11. 5 is used to simulate the IEEE-15 bus system.In this case, Table 6 displays the results of connecting the PV-DG without fragmentation to every available bus and displays that bus number 3 provides the best voltage profle.Additionally, the results in Table 7 show an improvement in system efciency of 98.8% when the PV-DG is connected to bus number 3. Journal of Electrical and Computer Engineering

Case C: At Load Change (Scheme 3).
In this case, Table 8 shows another description of the load change which is used to simulate the system of the IEEE-15 bus.By carrying out the load fow analysis PV-DG without fragmentation is connected to every available bus.Table 9 gives the voltage profle of the system which presents that bus number 6 is the best voltage profle.Te PV-DG unit's connection to bus number 6 results in a 98% increase in system efciency as presented in Table 10.

Second Technique: PV-DG with Fragmentation.
Te fragmentation here refers to the spread out of the small capacity of PV-DG unit installations across a larger area.Tis approach is more advantageous than a single largecapacity of PV-DG installation; which overcomes the problem of the available transfer capacities of the existing EDN.In this work, the PV-DG with fragmentation into small units had occurred by dividing the PV-DG unit into 14 equal parts each 28 kW approximately as shown in Figure 3, and connecting to each loaded bus with distribution lines parameters as described in Table 11.
Tree diferent scenarios were considered to study the performance of the EDN when the PV-DG units were fragmented under diferent load schemes as presented when the PV-DG unit was not fragmented.

Case A1: Fragmentation at Standard Load (Scheme 1).
A MATLAB/Simulink environment is used to simulate the IEEE-15 bus system using the load data shown in Table 1.Table 12 displays voltages buses using "PV-DG with fragmentation" versus the "PV-DG without fragmentation" optimum bus through T.L. (Scheme 1).Results shown in Table 13 indicate a signifcant enhancement in the system efciency of 99.4% in this case compared to the system efciency of 96.6% when connecting the PV-DG without fragmentation to bus no.11.
Figure 4 shows the simulation results of grid voltages buses' profle at load Scheme 1 when there is no PV-DG connection and when the two proposed techniques are applied: "PV-DG without fragmentation" fnding the optimum bus to be connected and "PV-DG with fragmentation."Results show that the two proposed techniques enhance the performance more than no connection of PV-DG, while the second applied technique of PV-DG with fragmentation is the best.Furthermore, comparing the total losses in this standard load (Scheme 1) for the system without PV-DG penetration 35.6 KW [31], with optimal bus-connected unfragmented PV-DG penetration 28.7 KW, and with fragmented PV-DG penetration 4.7 KW, a very high performance can be noticed using PV-DG with fragmentation.

Case B1: Fragmentation at Load Change (Scheme 2).
In this case, Table 14 shows all obtained bus voltages of "PV-DG with fragmentation" versus "PV-DG without fragmentation" optimum bus through T.L. (Scheme 2).Results shown in Table 15 also indicate a signifcant enhancement in the system efciency of 98.9% in this case comparing the system efciency of 98% when connecting the PV-DG without fragmentation to optimum Bus no. 3.
Figure 5 shows the simulation results of grid voltages buses' profle at load Scheme 2 when there is no PV-DG connection and when the two proposed techniques are applied: "PV-DG without fragmentation" fnding the optimum bus to be connected and "PV-DG with fragmentation."Results show that the obtained optimum bus considering load change scheme 2 (Bus no. 3) when applying the frst technique difers from load change scheme 1 (Bus no.11), which is not doable in reality to change the connected bus.In the meantime, still, the two proposed techniques still enhance the performance more than no connection of PV-DG, while the second applied technique of PV-DG with fragmentation is the best.

Case C1: Fragmentation at Load Change (Scheme 3).
In this case, Table 16 shows all obtained bus voltages of "PV-DG with fragmentation" versus "PV-DG without   Figure 6 shows the simulation results of grid voltages buses' profle at load Scheme 3 when there is no PV-DG connection and when the two proposed techniques are applied: "PV-DG without fragmentation" fnding the optimum bus to be connected and "PV-DG with fragmentation."Results show also that the obtained optimum bus considering load change scheme 3 (Bus no. 6) when applying the frst technique difers from load change scheme 1 (Bus no.11) and scheme 2 (Bus no.3), which is not doable in reality to change the connected bus.In the meantime, still, the two proposed techniques still enhance the performance more than no connection of PV-DG, while the second applied technique of PV-DG with fragmentation is the best.

Conclusions and Future Work
Te main goal is to look into ways to increase grid performance while still adhering to the restrictions for PV sizing, PV location, and various realistic loads.Te research applied two proposed techniques considering the available transfer capacity (ATC) of the EDN into account to fnd the optimally connected bus.Results show grid performance enhancement by applying the frst technique where the optimal connected bus is changed by changing the load.Unfortunately, the validation of this action is not realistic.Simulation results indicate that the PV-DG with fragmentation into small units and connecting them to all buses as a novelty second technique was the best realistic action plan for improving grid performance.In future work, the authors will apply the proposed approach to various IEEE buses or any redial system and consider an economic analysis.

Figure 1 :
Figure 1: PV-DG unit connected to the IEEE-15 bus system.
DG-PV Unit at bus 1

Table 2 :
Distribution line data for PV-DG without fragmentation connected to grid-available buses.

Table 3 :
Bus voltages when connecting PV-DG without fragmentation to available buses through T.L. (scheme 1).

Table 4 :
Total demand, losses, and generation with PV-DG without fragmentation connected to bus 11, (scheme 1).
system using the standard load data shown in Table1.Table2displays the distribution line parameters TL P for various connections between the PV-DG without fragmentation and available connected buses.

Table 6 :
Voltages buses when connecting PV-DG without fragmentation to available buses through TL, (scheme 2).

Table 7 :
Total demand, losses, and generation with PV-DG without fragmentation connected to bus 3 (scheme 2).

Table 9 :
Voltages buses when connecting PV-DG without fragmentation to available buses through T.L. (scheme 3).

Table 10 :
Total demand, losses, and generation with PV-DG without fragmentation connected to bus 6 (scheme 3).

Table 11 :
Transmission lines data for the PV-DG with fragmentation connected to the grid buses.

Table 12 :
PV-DG with fragmentation voltages buses compared to PV-DG without fragmentation optimum bus (scheme 1).

Table 14 :
PV-DG with fragmentation voltages buses compared to PV-DG without fragmentation optimum bus (scheme 2).
fragmentation" optimum bus through T.L. (Scheme 3).Results shown in Table17also indicate a signifcant enhancement in the system efciency of 98.5% in this case comparing the system efciency of 98% when connecting the PV-DG without fragmentation to optimum Bus no. 6.

Table 16 :
PV-DG with fragmentation voltages buses compared to PV-DG without fragmentation optimum bus (scheme 3).