Study on Optimization Model ofPipeline Network Scheduling Based on Gas Load Prediction

Aiming at the scheduling problem of local gas pipe network, an optimal scheduling model based on gas load prediction is proposed in this paper. Firstly, the model and solution of gas load prediction are built based on short and longtime memory method (LSTM). Then, according to the optimal scheduling results of the scheduling layer, the production plan is made, and the step-type correction strategy for dynamic flow and pressure parameters is studied and established to ensure the schedulable capacity between pipe networks. Tabu search algorithm (TS) with strong global search ability is selected to solve the scheduling layer optimization model. Finally, one year operation data of a gas company is selected for simulation experiment, and the results show that the model and optimization method are effective.