At present, the control of most actual servo systems still uses PID control because this controller has the advantages of being intuitive, simple to implement, and robust. However, for the actual system, especially the actual servo system with variable load and strong interference, the control of this control strategy is affected because of its difficulty in establishing the mathematical model, the uncertainty of the model and the nonlinearity. In this paper, neural network and PID controller are combined to complete the PID adaptive control. Simulation and: 3 PID adaptive controller design 3.1 System requirements In use, the system has large changes in the moment of inertia, there is a large unbalanced torque and not The balance moment changes greatly. At the same time, the required position accuracy of the system is 0.3mard, the speed error is 2Cmard, and the equivalent sinusoidal error is 4tmard ((a=37*/s2,v=24*/s). Therefore, it is difficult for the conventional PID controller to meet the system requirements. The summation is replaced by a summation and the differential is replaced by a finite difference, ie the above equation is: where T is the sampling period. For the above equation, a two-layer linear neural network is used to construct the controller as shown in Fig. 1: Gradient method is available for the neural network PID control coefficient correction: For the initial value and learning step, it is not only related to whether it reaches the global minimum point , but also affect the length of learning time Tao Yonghua and so on. The new PID control and its application 丨M Bo Machinery Industry Publishing (on page 8) fuzzy controller better solve the fuzzy control of the problem of large static error, and its dynamic performance is better than the classic PID control and pure Quantitative fuzzy control. 6 Conclusions In this paper, a filter identification method for on-line identification of time delay is proposed. Based on this, the proposed fuzzy controller overcomes the influence of time delay on the steady-state performance of the fuzzy control due to the introduction of the prediction, due to the introduction of class integrals. The defect of the fuzzy control itself has been overcome well, and the fuzzy controller has improved the steady-state performance of the fuzzy controller from both internal and external factors. Li Zhanming, Li Juan. An incremental fuzzy controller that effectively overcomes static errors. Industrial instrumentation and automation devices. 2000(4)::6. Li Zhanming, Li Juan. A Variable Structure Adaptive Model for Overcoming Static Errors Bai Jianguo, Hu Kejian. A new method to compensate for the pure time-lag process - intelligent sampling adjustment. J. Chemical automation instrumentation. 1988, Wang Dan, Hu Xiaojing. Intelligent resistance furnace temperature Control System | J |. Automation Chen Xiaohong. Application of Fuzzy Controller in Temperature Control System of Resistance Furnace IJ. Measurement and Control Technology. 1966 is like Chang. Predictive Fuzzy Controllers and Applications IJ. Fuzzy Systems and Numbers 1. Attractive small-sized high power LED spot light, suitable for architectural, landscape and accent lighting applications. Spot Light,LED Spot Light,Anti Surge LED Spot Lights,High Power Spot Light,RGB Spot Light,IP66 Spot Light StrongLED Lighting Systems (Suzhou) Co., Ltd. , https://www.strongledcn.com
2. Cree or Lumileds LED in RGB or mono color and various beam angles.
3. Available in several LED configurations with various power outputs range to meet various projection requirements.
4. High brightness contrast level for consistent and smooth color change effects.
5. Precision die cast aluminum, clear tempered glass diffuser and optional anti-glare hood.
6. Tilt and planar rotation adjustable mounting base for easy and fast installation and beam projection setup.
7. AC110V~220V or DC24V input voltage with integrated driver. Ingress protection at IP66.