Study On Improved Genetic Algorithm For Train Energy-saving Operation Optimization

High speed train has the advantages of safety,high speed et al.And it plays a more and more important role in the railway transportation of China.But the energy consumption of the high speed train is huge.Therefore,reducing the energy consumption of high speed railway has a great significance for the development of green transportation system.This thesis aims to study the problem of the high speed train energy-saving optimization.The genetic algorithm is used to solve this problem.However,the effectiveness suffer from some potential drawbacks,such as low quality and slow convergence.Therefore,an improved genetic algorithm is proposed.Firstly,based on the train traction calculation,a high speed train kinematic model is established.Considering the constraints of ramp,speed limit,electrical phase,precision,accurate parking,the energy-saving optimization model of high speed train is established.Taking train operation switching points as variable,and traffic safety and punctuality as well as precise parking as constraints,the genetic algorithm is adopted to minimize the energy consumption of train operation.Furthermore,an improved genetic algorithm is proposed to achieve a better effect.The electrical phase operation is regarded as a gene,and the operation sequence fragment are inserted into the chromosome around the electrical phase operation.The content of the operation sequence fragment is {Maximum acceleration-Cruising by partial traction/braking force-Coasting}.According to the characteristics of train operation,the concept of recessive and dominant gene is proposed.The new gene is encoded with genetic traits and train operation switching points,and the crossover and mutation operators are improved effectively.In order to accelerate the population convergent rate and improve the solution quality,a new mechanism that can guide the evolution direction is introduced.Finally,the simulation data of CRH380 A EMU and three actual lines are used to simulate the algorithm that designed in this thesis.The demonstrations show that the algorithm is effective.In order to verify the improvement effectiveness of the algorithm,the simple genetic algorithm and the adaptive genetic algorithm are adopted for contrast.Comparison results show that the algorithm presented in this thesis can improve the convergence speed effectively and can save more energy consumption than the simple genetic algorithm and the adaptive genetic algorithm.