Adaptive Control of the Mechanical Part of the Electric Drive System Using a Reference Model of Neuro-Fuzzy Logic

The introduction of new generation mechanisms in technological processes poses the tasks of ensuring their control, which require urgent solutions. For this reason, solving tasks that meet the arising challenges that have arisen and are aimed at improving the efficiency of control systems for electric drives of mechanisms becomes a priority. This paper presents a new approach to adaptive control of the mechanical part of the electric drive system of mechanisms operating in uncertain conditions. The proposed approach is modeled in the MATLAB software environment and provides high control speed, accuracy, adaptability and the ability to expand capabilities. This is achieved due to the fuzzy reference model used in the system and the coordinated operation of the PID controller, as well as the neural network's ability to learn quickly and accurately. The observations and studies necessary for the development of the control model were carried out for both the two-mass and three-mass mechanical parts of the electric drive system. This took into account the possible changes in connection stiffness that occur in mechanical parts and the ability of the control system to adapt to them. Tests of the model show that it suppresses the elastic vibrations of the mechanical part.