Prediction of Evaluation Index of Tie Line Power Control Based on LSTM

Abstract

With the large-scale wind power connected to the power grid, the volatility and uncertainty of wind power output will cause fluctuations in the power of the tie line between the interconnected regional power grids, resulting in the limit of the power of the tie line, affecting the assessment of the power system and even threatening system security. Therefore, this paper proposes an evaluation index prediction method of tie line power control based on LSTM. Use historical data of a grid tie line as input, and establish a prediction model and predict the tie line power in advance. The CPS standard for the tie line control performance evaluation index is introduced, and the CPS indicator is predicted based on the tie line power prediction. The results of the example show that the power of the tie line predicted by LSTM is basically the same as the actual power, and the CPS indicator is more reliable. Comparing the prediction method with the results of other prediction methods, the error is small, and the feasibility and effectiveness of the above prediction method for the evaluation of tie line control performance are verified.