基于流体波信号和改进遗传算法的供水管网泄漏监测点位布局优化方法  

Layout Optimization Method for Leakage Detection Points in Water Supply Network based on Fluid Wave Signal and Improved Genetic Algorithm

作　　者：韩卓[1] 李炜[1] 岳宇 李丹丹 马子胜 任丰坤 李云飞 刘娜[4] 邢兰昌[3] HAN Zhuo;LI Wei;YUE Yu;LI Dandan;MA Zisheng;REN Fengkun;LI Yunfei;LIU Na;XING Lanchang(Technical Testing Center,Sinopec Shengli Oilfield Company,Dongying 257000,China;Testing and Evaluation Research Co.,Ltd.,Sinopec Shengli Oilfield Company,Dongying 257000,China;College of Control Science and Engineering,China University of Petroleum(East China),Dongying 266580,China;Shengli Oil Production Plant,Sinopec Shengli Oilfield Company,Dongying 257000,China)

机构地区：[1]中国石化胜利油田分公司技术检测中心,山东东营257000 [2]中国石化胜利油田分公司胜利油田检测评价研究有限公司,山东东营257000 [3]中国石油大学(华东)控制科学与工程学院,山东青岛266580 [4]中国石化胜利油田分公司胜利采油厂,山东东营257000

出　　处：《计算机测量与控制》2024年第12期73-80,共8页Computer Measurement &Control

基　　金：国家留学基金项目(202106455003);山东省自然科学基金项目(ZR2024ME090);中央高校基本科研业务费专项资金项目(20CX05005A);中石化集团公司科技攻关项目(P321043)。

摘　　要：针对炼化企业供水管道泄漏监测技术需求,提出了基于流体波信号和改进遗传算法的监测点位布局优化方法;以管道内流体波信号的传播模型为理论基础分析了泄漏引起的流体波信号的波速和衰减特性;在引入相邻监测点最短间隔、传感器成本、管道不平衡量、管道流量、管道风险等级等因素作为约束条件的基础上构建了监测点位布局优化模型;对传统的遗传算法进行了改进,解决了算法中重复编码的问题;采用管网仿真案例对所建立的监测点位布局优化方法进行了验证,首先利用最短间隔参数对监测点进行约束,避免监测点距离较近引起的监测范围重叠问题,然后以管网覆盖率及其变化率为性能指标获得最佳监测点数目,实现经济性布局的目标,最后利用改进的遗传算法对优化模型进行求解,获得监测点位的最优布局方案;仿真管网和实际管网测试的结果表明:在选择不同约束因素的前提下对监测点位进行优化布局,监测点均能够有效地分布于使优化模型目标函数值取值较大的管道区域,验证了所建立布局优化方法的可靠性。To meet the leakage detection requirements of water supplying pipelines in refining enterprises,a detection point layout optimization method based on fluid wave signals and improved genetic algorithm was proposed.Based on the propagation model of fluid wave signals in pipelines,the wave velocity and attenuation characteristics of fluid wave signals caused by leakage were analyzed.A detection point layout optimization model was constructed based on the introduction of factors such as the shortest interval between adjacent detection points,sensor cost,pipeline imbalance,pipeline flow rate,and pipeline risk level as constraints.An traditional genetic algorithm was improved to solve the issue of duplicate encoding in the algorithm.The layout optimization method for the established detection points was validated using a pipeline simulation case.Firstly,the shortest interval parameter was used to constrain the detection points to avoid the overlapping detection ranges caused by nearby detection points.Then,the pipeline network coverage and its change rate as performance indicators were used to obtain the number of optimal detection points to achieve the goal of economic layout.Finally,the improved genetic algorithm was used to solve the optimization model,and achieve the optimal layout results for detection points.Experimental results between simulation and actual pipeline networks show that under different constraints,the layout of detection points is optimized,it can be effectively distributed in the pipeline area where the optimization model is a higher objective function value,verifying the reliability of the established layout optimization method.

关 键 词：供水管网 泄漏监测 点位布局 流体波 遗传算法 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]