A Simultaneous Approach for Optimal Allocation of Renewable Energy Sources and Charging Stations based on Improved GA-PSO Algorithm

The performance of the proposed method on the IEEE 33-bus system has been investigated to validate the effectiveness of the new GA-PSO method for optimal placement and sizing of RES and EV charging stations simultaneously. Output power of the charging station Output power maximum of wind power Output power maximum of solar power Output power hourly for wind power system Output power hourly for solar power Charged power rate of EV. According to the previous studies, most of the vehicles are in parking mode almost 95% of the day.

It can also be used to reduce network challenges by exploiting the capabilities of electric vehicles and PHEV1 charging stations [11,12]. In [19], a method for the optimality of electric vehicles and achieving the maximum benefit of the aggregator is given. By researching recent documents, the simultaneous optimal determination of the location and size of RES and the charging schedule of charging stations and electric vehicles has not been done.

In contrast to recent studies, a simultaneous determination of the location and capacity of RES and EV charging stations is obtained in this work. Moreover, optimal scheduling of the EV charging process is considered to use the vehicles for improving the electricity grid. In our model we use minimization of two terms, the first term, [𝑠𝑜𝑐𝑖𝑛𝑖𝑡+𝐸𝐷𝑒𝑚𝑎𝑛𝑑𝐵𝐶𝐴𝑃 ] gives us the expected charging demand of the.

In summary, the desired charging level of each EV driver cannot exceed the achievable amount of charging demand based on the duration of availability for charging, the charging speed and also the distance driven of the EV.

Fig. 1. Solar radiation profile [36].

The formulation and solution

Minimizing feeder losses is the desired goal from the distribution system operator's perspective. The penetration of renewable energy sources and electric vehicles into the electricity grid creates capacity to meet some of the demand, which helps the system reduce losses and improve voltage fluctuations. To increase the share of renewable energy sources in the supply of charging demand, the objective function was multiplied by the maximum power of each of these sources and divided by the hourly output of the sources.

The important point is the direct relationship between the amount of power provided in V2G and the battery depreciation cost, where it can increase the cost of the battery and thus prevent vehicle owners from participating in V2G. Therefore, the determination of location and capacity can affect the power losses, which are part of the objective function, as shown in Eq. In this method, implementation of the leading and lagging phase based on the power equations takes place and the line current directly is not used.

Given the objective function equation and the values ​​derived from the load current, the algorithm selects the best value for the objective variables. In this study, the idea of ​​a hybrid algorithm is covered, taking into account the structural and functional differences between different optimization algorithms. Considering that our decision variables in this study are two types of integers and floating-point numbers, considering the better performance of the PSO algorithm for continuous and infinite spaces and the performance of the GA algorithm for discrete spaces, we used GA-PSO to find the optimal solution.

At the end of the optimization process, each algorithm has its NOFE value, which indicates the speed of the algorithm in finding optimal solutions. 𝑃𝑐 and 𝑃𝑚 are the percentage of crossover and mutation operators in the genetic part of the proposed method, respectively. In each internal iteration of GA or PSO, if the generated answer remains stable for 40% of the answer population, the amount of 𝑃𝑚 and 𝑃𝑐 varies by multiplying the random variable 𝜀𝑖.

Finally, after 200 outer iterations, the best value for target variables and the best value for a set of parameters in the algorithm are calculated. The improved algorithm has a unique structure due to the population production process and elimination of the unwanted results. In this context, in addition to the creation of the best solution, the speed of the algorithm to solve the optimization problem is also increased.

Flowchart of the optimization algorithm and the flowchart of the proposed methodology are shown in Fig. respectively. 4 and Fig. 5 shown. If the generated populations remain stable for 40% of the solutions, population works on 𝑃𝑐 & 𝑃𝑚. - Application of selection process, updating the previous solutions with the new produced population to minimize the objective function.

Fig. 4.The GA-PSO method for finding optimal place and capacity of renewable energy sources and charging station

Simulation and results 1 IEEE-33bus system

Capacity and location of RES and charging station are the target variables in the optimization problem. The purpose of system study in the base scenario is to determine the system behavior before reconfiguration. In the first scenario, optimal placement and size of RES in the grid was achieved to determine the impact of these resources on the network operational parameters.

In the second and third scenarios, EV penetration to the network, considering the impact of vehicle charging management and V2G, is simulated. The fourth scenario is simply investigated for a short-term price impact on consumption pattern, pricing based on usage in the moment is an indirect way to control network consumers. In the fourth scenario, the network structure of the third scenario is used and only the pricing method has changed.

The value of the objective variables obtained from the optimization process is given in Table 5. 9 shows the effect of the simultaneous deployment of G2V and V2G on average voltage profile in the network buses. In the third and fourth scenarios, voltage level improvement, approaching the voltage magnitude to 1pu, and the voltage fluctuation reduction are obtained.

11 shows the network voltage fluctuations of the end buses (which have the lowest voltage level among other buses in the baseline scenario) within 24 hours for different scenarios. According to the results, the lowest voltage fluctuation occurred in the base case and the first scenario. However, in terms of voltage magnitude, the highest voltage level of the end buses is obtained from the fourth scenario.

Since this pricing structure in the fourth scenario provides high price of energy at high load (peak) times, it makes consumers to inhibit electricity consumption to recharge their EVs at that time and also prevents grid overload due to the charging requirement. 13, in the second scenario, the system demand due to the presence of EVs and their random charging schedules faces high volatility and effectively overloads are imposed on the grid at peak moments. By using the real-time pricing model in the fourth scenario, load shifting and peak shaving are realized, so the demand curve compared to the third scenario is more favorable.

To investigate the effectiveness of improved GA-PSO algorithm in such an optimization process, the operation of the algorithm is compared with the DE algorithm given in [15]. Consequently, the objective function value calculated by GA-PSO is lower than the DE results except for 𝐹4.

Fig. 7. System load profile [36].

Conclusion

1] Fazelpour F, Vafaeipour M, Rahbari O, Rosen MA, "Intelligent optimization to integrate a plug-in hybrid electric vehicle smart parking lot with renewable energy sources and improve grid characteristics," Energy Conversion and Management, vol. 3] Amini MH, et al., “Effect of Electric Vehicle Parking Lot Charging Demand as Dispatchable Loads on Power Loss,” IEEE International Conference on Electro-Information Technology (EIT), May 19-21. 6] Katrasnik T, "Energy Conversion Phenomena in Plug-in Hybrid Electric Vehicles," Energy Conversion and Management, vol.

7] Kempton W, Tomic J, "Vehicle-to-grid power fundamentals: Calculating capacity and net revenue," Journal of power sources, vol. 8] Amini MH, et al, "ARIMA-based demand forecasting method considering the probability of electric vehicle parking lots," Power and Energy and Energy Society General Assembly, IEEE, 26-30 July. 11] Honarmand M, Zakariazadeh A, JadidSh, "Self-scheduling of electric vehicles in an intelligent parking lot using stochastic optimization," journal of the franklin institute, vol.

12] Liu Z, Wen F, Ledwich G, "Optimal Planning of Electric Vehicle Charging Stations in Distribution Systems," IEEE Transactions on Power Delivery, vol. 15] Moradi MH, Abedini M, Tousi SMR, Hosseinian SM, "Optimal Siting and Sizing of Renewable Energy Sources and Charging Stations Simultaneously Based on Differential Evolution Algorithm," International Journal of Electric Power and Energy Systems, vol. 16] Turton H, Moura F, "Vehicle-to-Grid Systems for Sustainable Development: An Integrated Energy Analysis," Technological Forecasting and Social Change, vol.

17] El-Zonkoly A, “Intelligent energy management of optimally localized renewable energy systems integrating PHEV”, energy conversion and management, vol. 19] Han S, Han S, Sezaki K, "Development of an optimal vehicle-to-grid aggregator for frequency regulation", IEEE transactions on smart grid, vol. 20] Kempton W, “A test of vehicle-to-grid (V2G) for energy storage and frequency control in the PJM system”, is the result of an industry-university research partnership, 2008.

21] Moradijoz M, ParsaMoghadam M, Haghifam MR, Alishahi E, "A multi-objective optimization problem for parking space allocation in a distribution network", International Journal of Electric Power and Power Systems, vol. 22] Amini MH, Parsa Moghaddam M, Karabasoglu O, “Simultaneous allocation of electric vehicle parking spaces and distributed renewable resources in smart power distribution networks”, Sustainable Cities and Society, vol. 23] Amini MH, Sarwat AI, “Optimal reliability-based deployment of plug-in electric vehicles in a smart distribution network”, International Journal of Energy Sciences, vol.