Study On Urban Rail Transit Train Energy-saving Control Methods Based On Genetic Algorithm

In recent years, the urban mass transit system is developing rapidly in China. Although the mass railway transit could alleviate traffic congestion and provide convenience for the resident, at the same time it has produced enormous energy consumption and extremely high operational cost. Train operation energy consumption is an important part of urban mass transit system operational energy consumption, and the energy-efficient train operation is an effective way to reduce train operation energy consumption. Therefore, the research on train energy-efficient driving strategies has the vital practical significance as long as it satisfies the urban mass transit system operational requirements.Based on the researches by fellow researchers, this paper establishes an optimization model of energy-efficient train operation considering the impact of preceding train, designs a real-coded multi-group parallel genetic algorithm against the characteristics of this model and uses the simulation method to find the result. The following components are involved:(1) Establishment of the energy-efficient train operation optimization model on urban mass transit system. Fully considering the line conditions and the impact of preceding train on train operation energy consumption, establish the modified optimization model of energy-efficient train operation by using train operation sequence as control variable, using minimum train operation energy consumption as objective function and using train travel time, distance, speed, condition conversion and more as constraint conditions.(2) Design of the real-coded multi-group parallel genetic algorithm on energy-efficient train operation optimization model. Use real matrix encoding mode to express the train operation sequence, introduce chromosome-length mutation operator, and conduct the multi-group parallel optimization and individual migration between the sub-groups. Take probability control by using probability factor based on evolution phase in order to achieve quickly and efficiently convergence of the algorithm(3) Simulation implementation of the genetic algorithm. Based on the Microsoft Visual C++6.0, embed the real-coded multi-group parallel genetic algorithm into the Urban Train Movement Calculation System. Design the simulation process based on genetic algorithm, as well as the data structure and the key modules. Finally use the simulation method to find the result of the real-coded multi-group parallel genetic algorithm and the energy-efficient train operation optimization model.(4) Case study and analysis of simulation results. Conduct the train operation simulations under three kinds of cases, namely, the existing lines and train conditions, impact of preceding train conditions and different technical equipment conditions. Verify the effectiveness of the models and algorithms, optimization performance and applicable scope. Study the energy-saving control methods of the tracing train with the effect of the leading train. Then conduct research on the variation trend of the train operation energy consumption under the condition of different technical equipment and the influence of different control strategies on energy consumption in order to provide some useful references on track design, vehicle selection and train control method for urban rail transit system operation and management.