Predictive Control of a Synchronous Generator Network Using Neural Networks

It appears that the current global upswing is in tandem with a strong push toward electrical energy as an industrialization strategy. Power plants are an essential aspect of any modern infrastructure, and this project is the best option for backup and industrial generators. In this paper, we provide a novel approach based on neural network technology long-short-term-memory (LSTM), and particle swarm optimization (PSO) methods to address this issue. Predictive control of a synchronous generator system is the focus of this research. In the first phase, we tried a few different approaches to solving the issue, including a support vector machine, a random forest, and a decision tree. Then, we realized that the results we had been getting weren’t cutting it, so we set out to create a new LSTM-based technique. Based on the results of the traditional LSTM, a new study applies the LSTM-based PSO algorithm to solve the issue. The findings obtained are compared, and it is shown that the suggested LSTM-based PSO is superior to the previous investigations.