优化遗传算法在天然气管网流量分配过程模拟计算中的应用OA北大核心CSTPCD

In the flow process of natural gas pipeline network,the gas flow is not uniform at the connection of multiple pipelines,and the loss of flow distribution is difficult to be accurately calculate,which brings calculation errors to the natural gas parameters of downstream pipelines.Based on the actual data of the pipe network,a multi-objective optimization model of flow distribution is constructed,the control equations are processed by linearization method,the multi-objective optimization model is simplified into a single-objective optimization model by selecting boundary conditions.The control matrix is adapted to a three-point format that can be solved using the T DMA method.The genetic algorithm is used to solve the parameters after flow distribution,and the model is corrected in time by monitoring the calculation error,and the subsequent flow distribution process of the pipe network is calculated.The results show that the method can accurately simulate the flow distribution process of natural gas pipeline network,and the maximum errors of pressure,temperature and flow are 0.15％and 0.31％,respectively,and the temperature error is 0.16 K,which meets the requirements of energy measurement.