水流量标准装置的流量和压力快速控制

Flow and Pressure Fast Control of Water Flow Standard Facility

针对流量和压力调节相互耦合导致控制系统快速性降低的问题,以水流量标准装置为例,分析了流量和压力的耦合关系;以调节阀开度为流量主控制量,变频器工作频率为压力主控制量;提出了神经网络结合PID的快速控制方法;建立了神经网络模型,并进行了模型预测精度验证和控制特性实验验证。预测结果与实验结果表明,变频器工作频率预测相对误差在±0.6%以内,调节阀开度预测相对误差在±5%以内;BP神经网络结合PID的控制效果良好:与串行PID控制相比,流量和压力的调节时间减少38.5%～87.3%;与并行PID控制相比,流量和压力的调节时间减少25.4%～83.7%。

Mutual coupling between the flow and pressure regulation reduces rapidity of the control system.To address this,taking water standard facility as an example,the coupling between flow and pressure regulation is analyzed.Regulating valve opening and inverter frequency are selected as the main control variable of the flow and the pressure respectively.And a quick flow and pressure control method based on back-propagation(BP)neural network and PID control is proposed.The established neural network model achieves a high predicting precision.The relative error of valve opening prediction is within±5%and that of inverter frequency is within±0.6%The experimental results indicate that the control combining BP neural network and PID can achieve a faster flow and pressure regulation:compared with the serial PID control,the regulation time is decreased by38.5%~87.3%,and compared with the parallel PID control,that is decreased by 25.4%~83.7%.