变异哈里斯鹰优化算法在气体泄漏溯源中的应用  

Application of mutant Harris hawks optimization algorithm in traceability of gases leakage

作　　者：夏志禹 徐正蓺[1] 李丹[1] 魏建明[1] XIA Zhiyu;XU Zhengyi;LI Dan;WEI Jianming(Center of Intelligent Information and Communications Technology Research and Development,Shanghai Advanced Research Institute,Chinese Academy of Sciences,Shanghai 201210,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院上海高等研究院,智能信息通信技术研究与发展中心,上海201210 [2]中国科学院大学,北京100049

出　　处：《传感器与微系统》2023年第11期160-164,共5页Transducer and Microsystem Technologies

基　　金：上海市2019年度“科技创新行动计划”项目(19DZ1202200);中国科学院青年创新促进会项目(2021289)。

摘　　要：针对气体泄漏事故的溯源问题,研究了提高精度和速度的新方法。首先,搭建气体泄漏应用算例,以传感器浓度监测数据和高斯烟羽模型计算数据的方差作为目标函数,将溯源问题转化为优化问题;其次,基于哈里斯鹰优化(HHO)算法和传统优化算法提出变异HHO(MHHO)算法,旨在更快速准确地解决该优化问题;最后,进行不同信噪比(SNR)条件下的试验与比较。研究结果表明:MHHO算法单次运行时长在0.51~0.53 s之间,所有参数的平均反算误差SNR为20时约为8.64%,SNR为50时约为6.41%,SNR为100时约为0.89%,在精度和速度方面相比其他算法具有明显的优势。因此,MHHO算法能更快速准确地反算泄漏源的三维坐标和强度。Aiming at the problem of tracing source of gas leakage accidents,a new method to improve the precision and speed is studied.Firstly,an application example of gas leakage is built,and the variance of the concentration monitoring data of the sensor and the calculated data of the Gaussian plume model is used as the objective function to transform the traceability problem of the gas leakage accident into an optimization problem.Secondly,based on Harris hawks optimization(HHO)algorithm and traditional optimization algorithm,a mutant HHO(MHHO)algorithm is proposed,which aims to solve the optimization problem more quickly and accurately.Finally,experiments and comparisons under different signal-to-noise ratio(SNR)conditions are carried out.The research results show that the single run time of the MHHO algorithm is between 0.51~0.53 s.The average back calculation error of all parameters is about 8.64% when SNR=20,6.41% when SNR=50 and 0.89% when SNR=100.It has obvious advantages over other algorithms in precision and speed.Therefore,MHHO algorithm can back-calculate the three-dimensional coordinates and the intensity of the leakage source more quickly and accurately.

关 键 词：变异哈里斯鹰优化算法 群体智能算法 高斯烟羽模型 源强反算 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] X928.5[自动化与计算机技术—控制科学与工程]