With the continuous development of social economy and the continuous updating of science and technology, large-scale numerical control equipment with high technical quality has been widely used in many fields.
Therefore, in order to meet the corresponding workpiece processing requirements and ensure the stability of the PID control system, it is necessary to improve the accuracy of its synchronization control. In this paper, the two-axis synchronous control of neural network PID controller is simply analyzed and discussed, hoping to be helpful to the research of relevant personnel. The PID controller is mainly composed of input circuit, operation circuit and output circuit. The essence of the work of the PID controller is to carry out proportional operation, differential operation and integral operation on the deviation signal, and then adjust the results so as to realize the control of the whole system. Therefore, the operation is the technical core and key link of the PID controller. By establishing and using the cross-couple contract step-by-step control scheme, the errors in the work between axles can be introduced into the corresponding compensation unit. Compensation unit can be said to be a data analyst of motor control regulator to a certain extent. It calculates position error and speed error continuously by using its own working principle, and then transmits these error data to motor controller, so that the motor can make up the corresponding error and adjust the technology according to the actual situation. Achieve more accurate, more effective and less error synchronization control.
As far as cross-coupled synchronous control is concerned, it is very important to design the compensation unit scientifically and reasonably.
Compensation unit is a key link in cross-coupled synchronous control, and also one of the most critical parts in the whole cross-coupled system. The compensation algorithm adopted by the compensation unit controller directly affects the effect of synchronous control. In order to ensure the quality of synchronous control and improve the control effect, the compensation unit must introduce the PID algorithm. Usually, the general PID control has certain instability, especially when the system is disturbed by external environment and other factors, it will have a great impact on its synchronization control. In order to ensure the control performance of the PID control, the performance of the neural network control can be used to further develop the advantages of the neural network control, and adjust the parameters of the PID according to the actual situation of external interference, so that the synchronization control is more stable and efficient. In the proportional link, the static error can be reduced by increasing the proportional coefficient. However, to increase the proportion coefficient, it is necessary to increase the proportion coefficient too much in a certain range, which will increase the overshoot to a certain extent, and then cause oscillation, which is not conducive to the stability of synchronous control.
In the integral link, increasing the integration time constant can adjust the overshoot, thereby reducing the oscillation, reducing the corresponding steady-state error, and so on. The stability of control is improved, but the time of eliminating static error is increased to some extent by increasing integral time constant. In the differential link, ensuring differential time constant can improve the dynamic performance of the whole system and effectively restrain the variation of deviation.
However, if the differential time constant is too large, the adjustment time of the system will be longer and the anti-interference ability of the whole system will be reduced to a great extent. Normally, it is difficult to satisfy the practical requirements of the two-axis synchronous control by using neural network control alone. Only when combined with ordinary PID control, can the technical difficulties in the two-axis synchronous control be improved.
In addition, the parameters of proportional link, integral link and differential link in PID control must be adjusted in real time according to the change of load, so as to make the control process more accurate, faster and more stable. Neural network can be divided into input layer, output layer and hidden layer. Simulink is needed to build the simulation model. By using Simulink, users can see the corresponding simulation results, which to a certain extent facilitates the user to adjust the relevant parameters, and also promotes the performance of the system and the user’s improvement of the system.
According to the actual situation, thermostatic element to test the performance of synchronous control, it is necessary to compare the cross-coupling scheme of the neural network PID compensator used in the simulation process with that of the conventional PID. Through the corresponding comparison and the simulation results, the synchronous control performance of the neural network PID controller is further studied. Detection to determine system performance.
The application of neural network PID control algorithm in the compensation unit of two-axis synchronous control is an effective means to improve the function and stability of the system. By taking the synchronous speed error generated by the two-axis motion as the reference signal and putting it into the neural network PID compensation unit, the relevant functions of the neural network are used to correct the parameters of the PID in real time, so as to realize the reasonable compensation of the synchronous error.