Research On Control System Of Electric Locomotive After Meeting Obstacles Based On CGWO

Mining electric locomotive is an important part of coal mines and play a very important role in mining transportation.However,due to driver fatigue and failure to find obstacles in time,collision accidents occur in electric locomotive transportation,resulting in casualties and property losses.To solve this problem,this paper designs an electric locomotive detecting and avoiding obstacles system based on CGWO,and edge computing,deep learning and population optimization algorithm.Based on the actual transportation environment of the electric locomotive,this article has established an electric locomotive detecting and avoiding obstacles system architecture,and analyzes the specific functions and key technologies of the system.In order to improve the real-time nature of the system,the edge cloud collaborative architecture based on edge calculation is designed.The edge nodes of the system were constructed through the embedded microprocessor on the electric locomotive,and the target detection and control decisions were completed.Cloud computing provides storage space and other resources for edge computing,which further improves the system performance.In order to allow electric locomotive to detect obstacles such as workers,materials and equipment on the rail,a kind of improved target detection model is proposed and deployed on an embedded terminal with limited computing resources.In this paper,Mobile Net V3-Large is used to improve SSD to complete the detection task of obstacles on the rail.On the premise of ensuring the accuracy of detection,the SSD model is improved by lightweight,the model size is reduced,and the performance of the improved SSD target detection model is tested.The parameters of the existing electric locomotive controller are constant,resulting in the problem of non-matching parameters when controlling the electric locomotive under different circumstances,resulting in a decrease in the dynamic response capacity of the motor.In order to improve the control performance of the motor control system,a variable-universe fuzzy controller based on CGWO is designed,Fuch chaotic mapping and normal cloud model are used to enhance the global searching ability of the algorithm,and the searching ability of Grey Wolf optimization algorithm is improved by adaptive adjustment of cloud model parameters,ensuring the accuracy of the scaling factor of the domain.The experimental results show that this design improves the control performance of PMSM and realizes the fast and accurate control of electric locomotive.The method proposed in this paper and the designed system realize the target detection function in the embedded terminal of the electric locomotive.Compared with the method of video data transmission to the cloud platform,this method ensures the real-time and reliability and realizes the autonomous obstacle control of the electric locomotive.The research work of this paper provides guarantee for t he safety of coal mine transportation and production.Figure [33] Table [10] Reference [80]...