This paper proposes a new load shedding method based on the application of a Dual Neural Network (NN). The combination of a Back-Propagation Neural Network (BPNN) and of Particle Swarm Optimization (PSO) aims to quickly predict and propose a load shedding strategy when a fault occurs in the microgrid (MG) system. The PSO algorithm has the ability to search and compare multiple points, so the proposed NN training method helps determine the link weights faster and stronger. As a result, the proposed method saves training time and achieves higher accuracy. The Analytic Hierarchy Process (AHP) algorithm is applied to rank the loads based on their importance factor. The results of the ratings of the loads serve as a basis for constructing the load shedding strategies of a NN combined with the PSO algorithm (ANN-PSO). The proposed load shedding method is tested on an IEEE 25-bus 8-generator MG power system. The simulation results show that the frequency recovery of the power system is positive. The proposed neural network adapts well to the simulated data of the system and achieves high performance in fault prediction.
1 August 2022 in International Journal of Electrical and Computer Engineering (IJECE)
This paper proposes an under-frequency load shedding (UFLS) method by using the optimization technique of artificial neural network (ANN) combined with particle swarm optimization (PSO) algorithm to determine the minimum load shedding capacity. The suggested technique using a hybrid algorithm ANN-PSO focuses on 2 main goals: determine whether process shedding plan or not and the distribution of the minimum of shedding power on each demand load bus in order to restore system’s frequency back to acceptable values. In the hybrid algorithm ANN-PSO, the PSO algorithm takes responsible for searching the optimal weights in the neural network structure, which can help to optimize the network training in terms of training speed and accuracy. The distribution of shedding power at each node considering the primary control and secondary control of the generators’ unit and the phase electrical distance between the outage generators and load nodes. The effectiveness of the proposed method is experimented with multiple generators outage cases at various load levels in the IEEE-37 Bus scheme where load shedding cases are considered compared with other traditional technique.
The rapidly drop of frequency in the microgrid (MG) system when it operates under island operation has caused frequency stability problems. In case, generators and renewable energy sources cannot supply enough power to restore the frequency, the load shedding will be proceeded. This paper proposes the load shedding method based on the coordinate multiple load shedding factors when the MG system operates in island mode. This proposed method carries out an approach through two main steps. Firstly, the optimal amount of load shedding power considering renewable energy sources and the ability to adjust the primary and secondary frequencies of the generator is calculated to restore the frequency to the allowed range. Next, distributing the optimal load shedding power for each bus is based on the combination of different criteria to meet the economic and technical goals. On the technical side, this paper also considers two criteria about the voltage distance, and the voltage sensitivity index. These two criteria contribute to reducing voltage drop and improving voltage stability when the microgrid operates in island mode. From the economic perspective, it considers the load importance factor, thereby minimizing the economic losses caused by load shedding. The AHP algorithm is applied in this paper to determine the importance of loads and the importance of the different criteria when they are combined. The effectiveness of the proposed method is demonstrated through simulation of IEEE 16-bus microgrid with 6-source.
13 October 2023 in Engineering, Technology & Applied Science Research
Optimizing the operational parameters and control of the power system in steady-state conditions is a crucial issue in reducing the costs of power generation and operation. In the case of long-term operation of a power system, besides aiming to minimize power generation costs, the cost of damage caused by load shedding also needs to be considered. This paper presents the optimization of the total cost of a power system including minimizing the generation cost function of power plants or power companies and minimizing the damage cost function caused to customers due to load shedding or power outages. At the same time, the objective function must also ensure the constraints on the operating conditions of the power system. This contributes to maintaining the continuity of the power supply to critical loads and minimizing damage. Base loads, priority loads, or loads that are not allowed to be shed are considered as constraints. The optimization problem is addressed by using the Particle Swarm Optimization (PSO) algorithm and the Cuckoo Search Algorithm (CSA). The IEEE 30-bus test system is applied to validate the reduction in total cost. The result comparison shows that when applying the CSA, the total cost is significantly reduced by 3.75% in comparison with the PSO algorithm. The algorithms are implemented in Matlab to demonstrate the efficiency and accuracy of the proposed method.
2 April 2023 in Engineering, Technology & Applied Science Research
The input feature is vital for power system stability classification. Feature selection can reduce the size of the input feature, making classifier training easier, and the small size of the input feature subset also reduces the cost of purchasing sensor measurement equipment. Therefore, feature selection works require an efficient seeking method. Binary Particle Swarm Optimization (BPSO) is a simple and easy-to-implement evolutionary calculation technique. While the classification accuracy of BPSO is essential, it is also needed to drive the algorithm to minimize the number of variables. The proposed approach is to apply a multi-objective function to help the BPSO algorithm identify the subset of features that can achieve the minimum number of variables and the highest classification accuracy. The k-nearest neighbor classifier is employed in the experiments to evaluate the classification performance on the IEEE 39-bus dataset.