Research On Operational Control Strategies Of Heavy Haul Trains Based On Fuzzy Neural Network

As a major artery and important infrastructure of the national economy,the freight capacity of railways to a certain extent affects the country’s economic development and people’s livelihood needs.Heavy haul railways fully leverage the advantages of centralized,bulk,medium to long distance,and all-weather transportation,improving transportation efficiency and reducing transportation costs.Heavily loaded trains have heavy loads,complex line conditions,and long operating times,which require higher performance requirements such as safety and stability during train operation.At present,manual operation is still the main driving method for heavy haul trains in China,which not only increases the workload of labor,but also no longer meets the development requirements of safety,efficiency,energy conservation,and environmental protection for heavy haul railway transportation in China.Therefore,conducting research on optimization strategies for heavy-duty train operation control has important theoretical and practical significance for train operation safety and stability.Based on the analysis of the impact of internal and external interference factors on train operation and control,a multi particle dynamic model of the train is established.In response to the problem that some control strategies are prone to falling into local optima and the optimization effect is not significant,a heavy haul train operation control strategy based on fuzzy neural network is proposed.The tracking simulation of the target speed curve is carried out to achieve effective control of heavy haul train operation and manipulation.The main achievements are as follows:(1)Analyzed the dynamic characteristics of heavy-duty trains during operation,explained the generation mechanism of internal and external forces,and constructed a longitudinal multi particle dynamic model based on this,accurately presenting the impact of internal forces on the train,and established its state space equation.(2)By combining neural networks with self-learning and adaptive characteristics,as well as fuzzy control algorithms that have good control effects on complex systems such as time-varying and hysteresis,the established membership function and rule base are mapped to the neural network.The neural network is used for fuzzy inference and adjustment of the established fuzzy control rules,and a heavy-duty train operation controller based on fuzzy neural network is designed.(3)On the basis of the above theoretical analysis and research,taking the operation control of heavy haul trains on the Shuohuang Railway as an example,a designed heavy haul train operation controller is used to conduct simulation experiments on tracking train speed curves.Evaluate the performance advantages of the designed controller by comparing and analyzing the changes in speed tracking error,control force,and coupler force.The simulation results show that fuzzy neural network control can effectively achieve the goal of stable control of train operating speed and has good regulation ability and stability.(4)A multi-objective model for optimizing the speed curve was established,and the target speed curve of the train in the experimental section was obtained.Considering the problem of falling into local optima during the parameter adjustment process of fuzzy neural networks,an improved particle swarm optimization algorithm was introduced to further optimize the initial parameters of the fuzzy neural network.The improved algorithm was used to control the train for tracking simulation of the target speed curve,verifying the effectiveness and superiority of the proposed control strategy in tracking the accuracy of the target speed curve and ensuring safe and smooth train operation.The simulation results show that the improved controller has improved in all aspects of performance compared to the previous PID controller.