A Safety Braking Model Of Heavy Haul Train Control System Under Complex Conditions

The Shuohuang Railway is an important heavy haul railway for coal transportation in China.The current fixed block method with three display have been unable to meet the rapidly increasing traffic requirements.The research of moving block technology based on wireless communication on the Shuohuang railway is carried out,to achieve the purpose of shortening the interval between trains and improving transportation capacity.Speed protection is a typical model-based control,obtaining a guaranteed braking process model is a prerequisite for speed protection.The two main factors affecting the safety brake model are the train dynamic characteristics and the line conditions.It should be specifically applied to the complex conditions of Shuanhuang railway,and study the safety braking model to ensure the implementation of speed protection under moving block conditions in Shuohuang railway.The characteristics of heavy haul trains and the complexity of the line make it difficult to obtain accurate mathematical models.The model parameters are numerous and coupled to each other.This paper proposes a safety braking model for heavy haul trains,and establishes a parameter identification framework for this model,the speed protection process is gradually transformed into different levels of nonlinear parameter models.Different identification methods of various model parameters are studied,which makes the speed protection more targeted and suitable for different operating scenarios.In this study,by introducing model decomposition and filtering techniques,the finite data is used to solve the parameter identification problem of colored noise interference in unit consisting trains.The distributed model of combined trains is established,and the coupling identification method is used to solve the problem of model splitting and multivariable parameter identification.The multi-innovation estimation strategy is used to solve the anti-interference identification running on the undulating slopes.Aiming at the influence of the long downhill cycle braking condition on the braking system,the parameter identification under the slow time-varying disturbance of the braking system is solved.This paper also evaluates the application results of the safety brake model and quantitatively estimates the uncertainty of the speed protection results,and finally it is evaluated whether the speed protection may cause longitudinal impulse anomalies.These methods ensure the safety and fault tolerance of the speed protection function.The main innovations of this paper are as follows:1.The safety braking model of heavy haul trains is established.The identification framework and experimental environment for the Shuohuang railway are designed for the model parameter identification.For the basic unit train consisting,the Hammerstein model with nonlinear input and output with autoregressive colored noise is established.A filter iterative method based on model decomposition is proposed,which can better fit the dynamic braking process of the unit train.2.A distributed model is established for combined train using wireless reconnection control system,and a partial coupling identification method for multivariable systems is proposed.The variation of the undulating slopes is converted into the unknown load disturbance of the output,and an anti-interference identification algorithm based on multi-innovation estimation strategy is derived.The model parameters and disturbance parameters are identified.The speed protection effect is significantly optimized.3.Aiming at the influence of the long heavy down grade cycle braking condition on the actual braking rate,a state space model of combined train based on multi-agent is established.The state change of air brake system is convert into unknown timedependent disturbance,and derive the subspace anti-interference identification algorithm based on orthogonal projection.Aiming at the condition of dynamic braking applied during cyclic braking,the nonlinear state space model of locomotive dynamic braking is established by Kalman filter-based iterative algorithm,which significantly improves the accuracy of speed protection execution.4.The quantitative evaluation of the uncertainty of execution effect is obtained.The probability distribution model of key parameters is obtained.Monte Carlo simulation of speed protection execution is performed by changing the parameter random variables,and the distribution function of braking distance is fitted.An evaluation method of longitudinal impulse state based on unknown input observer is proposed.Based on the convergence condition of the observer,an abnormal state recognition algorithm based on residual is designed to identify the surge state of longitudinal impulse.