Study And Design Of Adhesion Control Methods Of Heavy-Haul Locomotive

With the rapid development of railway transportation industry in china, the higher requirements for traction power of heavy-haul locomotive, the more complex states of locomotive operation, it makes more requirement for the traction of heavy-duty electric locomotive. To optimize the traction power and the transport capacity of heavy-duty locomotive, it is important to improve the adhesion utilize, which is an important factor of locomotive traction. With the study of adhesion control theory at home and abroad, the effective adhesion control method which can improve the adhesion utilization of locomotive is designed. Therefore, in-depth study of adhesion control method is important to develop new type heavy-duty electric locomotive.The rail surface will influence the characteristics of adhesion, so it is important to identify the condition of rail-surface effectively which is necessary to optimize the adhesion utilization. This thesis designs two methods for identifying the condition of rail-surface, which are based on locomotive operation experience and the amount of actual adhesion data. When there is few adhesion data for a new heavy-haul locomotive, the identification method of rail-surface is based on fuzzy system. With the increase of adhesion data of the locomotive, the identification method of rail surface will be based on SVM, the optimum classification function and the classification hyperplane can be acquired through training model based on the mass of experimental data. The accurate rail surface can be distinguished through those methods.To optimize the utilization adhesion of wheel-rail, a comprehensive adhesion control method combined with adhesion control method is proposed which is based on improved disturbance observer and hybrid adhesion control method. Aiming at the fixed adjustment value of traction and the wrong operation by the small scale variation of adhesion caused the loss of the traction. A optimized adhesion control method which based on disturbance observer is designed; the traction adjustment function is optimized by adhesion coefficient and its differential value which can be estimated by the disturbance observer; the method can improve the adhesion utilization. Because of the fixed identification threshold of idling of different rail surface in traditional hybrid adhesion control method, more traction loss of locomotive is caused; an adhesion control method based on SVM identification, which can switch the idling discriminant parameters and the output function of traction torque; it can reduce the loss of locomotive traction and improve the utilization of adhesion. A comprehensive adhesion control method that combined with the advantage of improved adhesion control method based on disturbance observer can quickly control idling, and the universality and stability of hybrid adhesion control method, is proposed; it can more quickly control the slip of axis and has better control stability.The processor-in-loop(PIL) co-simulation platform is designed. The adhesion control core is; the control object is locomotive model that built on.The locomotive simulation experiments is implemented on the PIL co-simulation, and verified the validity of heavy-duty locomotive adhesion control methods.