Research on the Strategy of Multi-Path Selection and Release of Power Data Based on Load Balance

In order to improve the efficiency of the electricity information collection system in releasing full collection data to the marketing business application system, implement load balancing strategies, and plan the selection of power data release paths, this paper proposes a multi-path transmission strategy optimization method based on LSTM prediction and MPC. Firstly, a multi-path data transmission state prediction model is constructed using LSTM to obtain the predicted values of the data transmission state for each path at the next moment; Then, input the predicted value for the next moment into the LSTM model to obtain a data transmission prediction for a period of time; Finally, based on the set minimum release time as the objective function, MPC rolling optimization and feedback correction are adopted to obtain the transmission control quantities of each path that meet the objective function. In comparative experiments with multiple baseline methods, it has been shown that the method proposed in this paper can achieve multi-path transmission strategy optimization for full data publishing and improve transmission efficiency. This article provides a new solution for the electricity information collection system to send full data to the marketing business application system, further shortening the publishing time of full meter reading data and ensuring the stability and security of meter reading data publishing and operation.