This paper shows an analytic hierarchy process (AHP) algorithm-based approach for load shedding based on the coordination of the load importance factor (LIF), the reciprocal phase angle sensitivity (RPAS), and the voltage electrical distance (VED) to rank the load buses. This problem is important from a power system point of view, and the AHP method is able to support the decision-making process in a simple and intuitive way in a three-criterion environment. This satisfies the multicriteria decision-making to meet economic-technical aspects. The ranking and distributed shedding power at each demand load bus are based on this combined weight. The smaller overall weights of the load buses show the lesser importance of the load bus, the smaller reciprocal phase angle sensitivity, and the closer voltage electrical distance. Therefore, these load buses cut a larger amount of capacity, and vice versa. By considering the generator control, the load shedding consists of the primary and secondary control features of the generators to minimize the load shedding capacity and restore the system frequency value back to the allowable range. The efficiency of the suggested load-shedding scheme was verified via the comparison with the under-frequency load shedding (UFLS). The latter result is that the load shedding power of the suggested approach is 22.64% lower than the UFLS method. The case studies are experienced on the IEEE 9-generator; the 37-bus system has proven its effectiveness.

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.