基于闪电搜索算法的水泵流量-扬程样本曲线校准

Calibration of pump flow-head sample curves based on lightning search algorithm

通常水泵的流量-扬程曲线是通过实测获得,但是并联水泵单泵流量-扬程曲线的实测需要安装很多仪表,且测试过程会影响水厂的正常供水。针对此问题,提出了一种利用水泵开启组合数据和出厂流量、压力数据来校准样本曲线的方法。分析了调速泵曲线的数解法和水泵调速运行下并联的基本原理,并建立了曲线校准的数学模型。采用闪电搜索算法,结合各水泵的实际运行数据,在曲线参数的范围内,对每台水泵的流量-扬程样本曲线进行校准。结果表明:模拟误差大多在±1%以内,该方法得到的水泵流量-扬程曲线可信,可以用于指导泵站的生产和调度。

Usually the flow-head curves of the pump are obtained by the actual measurement,but the measurement of the flow-head curve of the parallel water pump needs to install a lot of instruments,and the test process will affect the normal water supply of the water plant.In order to solve this problem,a new method of calibrating the sample curves using the combination data,flow and pressure data of running pump was proposed.Basic principle of numerical solution of the timing pump curves and parallel connection under the speed regulation operation of the pump was analyzed,and mathematical model of the curve calibration was established.Based on the lightning search algorithm and the actual operation data of each pump,the flow-head sample curve of each pump was calibrated in the range of curve parameters.The results showed that most of the simulation errors were less than±1%,and it could be considered that the pump flow-head curve obtained by this method was credible,and could provide guidance for the production and scheduling of pumping stations.