ELECTRIC VEHICLE SLIPPAGE PREVENTION SYSTEM BASED ON NEURAL NETWORK CONTROLLER

А functional diagram of the skidding prevention system was built, a mathematical model an asynchronous electric drive vehicle and neuro-regulator synthesized. The motor is selected according to equivalent power method, standard urban WLT cycle taken as basic cycle. mechanical part built taking into account possibility simulating slippage each driving wheels separately with different coupling coefficients. consists converter, battery, speed regulator, torque motor, braking resistor, unit for generating set signals, mechanics measuring units. internal control on basis DTC vector using Matlab blocks. This work uses NARMA-L2 block, which included in Neural Network ToolboxTM. A simplified (object model) neural network parameters such number hidden layers, discretization, samples, epochs were training. trained linearized object, reflects qualitative type real processes system. Regardless linearization system, output signal minimal error (about 1%) corresponds input. An analysis obtained results training carried out. Simulation operation without one simulation are compared during training, namely discrepancy between input signals. does not take side drift, so only wheel speeds linear change can be observed. traction method intelligent networks safety shown. conclusion made about operability efficiency neurocontroller prevent possible modes.