Several models of intelligent inquisition data CNC control and simulation

Control process setting curve PFC1PFC2PFCn controller scheduling process output multi-model PFC control system schematic 2.1 Model matching degree determination Multi-model PFC control, it is important to change the model when. According to engineering experience, it may be assumed that when the system is running, its transfer function is G1, and the actual transfer function is 1G'. The following methods can be used to judge whether the model is mismatched.

If J ≤ C (C is a constant, which can be determined according to engineering experience), it means that the model basically matches, otherwise the model mismatch must be switched. It should be pointed out here that because the adaptive control of the predictive function is strong, the accuracy of the system parameters is not high. Therefore, as long as the C is not too large, the control effect can always meet the engineering requirements.

The identification of the system parameters may be assumed that when the transfer function is G1, the system starts from the start to the time K, during which time, the calculated J1 ≤ C; when running to the K+1 time, the calculated J1> C, the description The model has been mismatched. At this time, the system is identified by the input and output data from the K+2-N time to the K+1 time, and the identified transfer function is G2. That is to say, the system switches at time K+1, and the model function after switching is G2, and G2 is used as a new prediction model, and the system is controlled by PFC, and so on until the end of the system operation. It is not difficult to understand that if the identified model function has high precision, at least in the N sampling periods of K+2-N to K+1, G2 is used as the prediction model, and the calculated J is obtained. As mentioned earlier, in a short time, many industrial control systems can use a first-order inertia plus lag link to represent their transfer function. For such a system, there are various methods for parameter identification, such as least squares method, frequency domain method, maximum likelihood method, etc., each having advantages and disadvantages. In order to overcome the "data saturation" phenomenon in the identification process, the recursive least squares method with variable forgetting factor is used to identify the parameters of the system online <10>.

Here, the value of ρ is adjusted in time with the change of the dynamic characteristics of the system. When the system has a sudden change, select a smaller ρ to improve the sensitivity; when the system tends to be stable, select a larger ρ to enhance the memory length. The calculation formula of the recursive least squares method of the forgetting factor is as follows: ε(i) is the residual between the measured value and the calculated value of the estimated model. According to the input and output data of N time and the above formula, the transfer function of the system during this period can be identified.

Further research on the system control process is based on the above identification method to find a new prediction model, and PFC is used to control the system. Similarly, at each sampling instant, the matching degree performance index J is calculated according to formula (11). If J ≤ C, it indicates that the model mismatch is small at this time, and the model can continue to be used, otherwise it should be re-identified immediately. Repeat this process to control the system until the end of the system. In order to facilitate understanding, the control process of the entire system is now represented by a flow chart.

NYNY begins to set the initial transfer function and each parameter uses N data identification Gi(s)i=i+1PFCi control Ji≤C end of operation

(Finish)

Jeepney

Isuzu Jeepney Modernization,Dongfeng Jeepney Mini Bus,Jeepney Passenger Van Truck,Isuzu Jeepney Passenger Van Truck

HUBEI FUYA VEHICLE CO., LTD. , https://www.fuyavehicle.com