This paper is devoted to propose method for prevention of train collisions. Nowadays human factor plays asignificant role in control of railway system in a whole and a rolling stock in particularly. Some crashes with lethaloutcomes happened in Latvia in last three years. Main reason is the driver’s inattention passing the red signal. The task isto prevent such accidents by reducing the human factor. In this paper artificial neural network controller is proposed formotion control of rolling stock and for braking way calculation to stop the train timely and safely before the dangerouspoint. Mathematical models and algorithm for task solution is proposed. Results of experiment show the possibility touse neural network controller for speed control of DC drive depending on the distance to stop. The results show thepossibility of the developed systems to prevent accidents and to avoid different problems by intelligent diagnostic andcoordination devices. Neural network may be used to prevent breakdowns and accidents and such kind of controllersmay be integrated in working infrastructure for optimal speed control of railway traffic.
Keywords: railway transport, intelligent control, safety.
The paper is based on authors’ previousscientific work researching the intelligent devicesystems and its’ application in mechatronic systems.Intelligent devices are controllers, which has interfaceto work in global network and wireless networks andare programmed to use methods of the artificialintelligence. Intelligent devices have possibility tonegotiate with each other and to coordinate their workto get better decision.
2. Problem formulation
There are three main safety levels in transportsystems. The first is the safety of mechatronics systemof a train. That means, an intelligent diagnosticsystem for engine states is needed to separatedangerous situations by critical testimonies fromsensors from the regular states of the system.The second level is the safe control ofmechatronics system in a rolling stock. One of theprimary tasks is an intelligent speed control of a train,using with multi-criteria decision making, taking inaccount weather factors, state of the way andschedule.The third safety level includes an intelligentcontrol of the whole transport system. That is why thesolution of coordination task between all trains in thesystem is necessary.This paper is devoted to propose method forprevention of train collisions. Nowadays human factorplays a significant role in control of railway system ina whole and a rolling stock in particularly. Somecrashes with lethal outcomes happened in Latvia inlast three years. Main reason is the driver’s inattentionpassing the red signal. The task is to prevent suchaccidents by reducing the human factor.
3. Method of solution
Authors propose to solve this task usingintelligent devices system for all three safety levels.Method for diagnostics task solution was proposed inauthor’s previous work , based on neural networkand clustering, gives possibility to detect and warnabout changes in the engine, detect the problemimmediately, and to fix it in some cases withouthuman intervention.In this paper artificial neural network controlleris proposed for motion control of rolling stock and forbraking way calculation to stop the train timely andsafely before the dangerous point. Mathematicalmodels and algorithm for task solution is proposed
Download full report