Source Location In Complex Environment Based On CEOA

With the development of industrialization,harmful gas leakage has threatened people’s production and life.At present,there are many researches on the location of harmful gas leakage source,but most of them assume that the leakage source is in a stable environment,so it cannot well cope with the challenges in reality.In order to solve the problem of locating gas leakage source in complex environment,this paper proposed a baseline fitting algorithm for spectral signal region to improve the accuracy of gas signal extraction,proposed a new intelligent optimization algorithm to improve the efficiency and reliability of leak source location,and finally located the leakage sources in complex environment based on the proposed algorithm.The main contents of this paper are as follows:1)Spectral baseline fitting algorithm research: A baseline fitting algorithm for spectral signal region,QGS(Quadratic polynomial fuse Gradual Suction),is proposed.The experimental results show that the QGS fitting algorithm has high accuracy and stability under different baseline types and different SNR.The relative error of the overall fitting accuracy is only 47.0% of the piecewise quadratic polynomial fitting,35.6% of the Air PLS fitting,and 20% of the Wavelet fitting.And the algorithm has good real-time performance because only the spectral signal region baseline is fitted.2)Intelligent optimization algorithm research: A new intelligent optimization algorithm,CEOA(Chief Executive Officer Election Optimization Algorithm),is proposed.It is inspired by the fact that different people have different choices the process of company CEO election.The performance of CEOA was evaluated by 21 classical test functions,29 CEC2017 conference functions and 6 specific engineering design optimization problems.The results show that CEOA algorithm has both efficient exploitation ability and effective exploration ability compared with the classical DE,GWO and advanced Optimization algorithms such as EO,MPA and AFEA.It achieves the overall best performance in both unimodal functions and multimodal functions,and in both low dimension functions and high dimension functions.3)Leakage source location algorithm research: A leakage source location algorithm based on CEOA in complex environment is proposed.The algorithm not only uses the real-time information measured by multi-robots as particles,but also uses the historical information,it once sensed as supplementary particles.The robot moving direction is gotten using CEOA algorithm and K-means clustering,and the multi-robot cooperation strategies such as decentralized search and local obstacle avoidance are proposed.In order to test the effectiveness of the algorithm,four different simulation scenarios and different numbers of robots are used for experiments.When the number of robots in different scenarios is equal to 3,the success rate of localization is over 90%,while when the number of robots is greater than 3,the success rate of localization is 100%.The results show that the leakage source location algorithm based on CEOA has strong robustness.It lays a foundation for the follow-up study of multiple leakage source location and leakage source location in real environment.