High-speed train self-adaptive control method and system based on multi-particle model

Abstract

The invention relates to a high-speed train self-adaptive control method and system based on a multi-particle model. The method comprises the steps of determining the actual state quantity of a high-speed train through a multi-particle model of the high-speed train; determining a target state quantity of the high-speed train by utilizing a reference model of the high-speed train; determining a tracking error according to the actual state quantity and the target state quantity; determining an adaptive rate according to the tracking error; determining a self-adaptive controller according to theself-adaptive rate; and controlling the high-speed train according to the self-adaptive controller. According to the invention, high-precision tracking control of the high-speed train is realized.

Description

technical field [0001] The invention relates to the field of modeling and tracking control of a high-speed train running process, in particular to an adaptive control method and system for a high-speed train based on a multi-mass model. Background technique [0002] Since the beginning of the new century, China's high-speed railway has developed rapidly and has achieved world-renowned achievements. At present, the development of high-speed railway is in a transition period, and intelligentization will be an inevitable trend. Apply advanced technologies such as artificial intelligence and intelligent control to the development of high-speed railways, realize automatic driving of high-speed trains, and automatically realize basic driving functions such as starting, traction, inertia, and braking of high-speed trains. Continuously updating and developing technologies related to automatic driving of high-speed trains. On the basis of precise modeling and effective control of hig...

AI Technical Summary

Problems solved by technology

However, on the basis of not affecting the performance of high-speed trains, many researchers have simplified the high-speed train model and tried various modeling methods, such as: data modeling, dynamic characteristic modeling, ANFIS model and other modeling methods. Design the control strategy on the basis of the high-speed train to achieve the expected performance of the high-speed train, but in essence, the above modeling is a kind of single-mass modeling, for the complex high-speed train with distributed power, it is not enough to describe the internal force relationship between adjacent cars

[0004] There have been many researches on specific control methods for realizing high-speed train tracking, such as: PID control, fuzzy control, neural control, predictive control and other algorithms, but they all have their own defects. PID control is not conducive to smooth operation, and the parking accuracy is not high; fuzzy control It is more dependent on actual experience and model establishment, and cannot achieve high-precision tracking; neural network control has poor stability and slow learning speed; predictive control has a small amount of calculation and simple algorithm strategy, but the system stability is not strong, so it is not suitable for high-speed trains Stable operation

Images