|Alternative Title||Research on laser thermal load temperature control system based on fuzzy neural network|
|Place of Conferral||北京|
Laser thermal load acts on the surface of the materials or components with the laser as a heat source to simulate the heating process in the service environment. It is an important way to test the thermal fatigue performance and life prediction of heated components. The temperature fluctuation in the process of the laser thermal load directly affects the transient thermal stress response and the thermal fatigue crack evolution process of the specimen. The effect of temperature control has a significant impact on the accuracy of the final results. With the increasing demand for the stability, high precision and intelligence of temperature control system in industrial applications, the existing conventional temperature control methods are difficult to meet the demand, which has become the bottleneck restricting the development of laser thermal load technology. Based on fuzzy control and neural network, a new type of temperature controller is designed and implemented in this paper, a temperature control system platform is built, and the human-computer interaction control interface and control software are compiled to improve the stability of the temperature control system and the controllability of temperature loading. It is of great significance for the thermal load experiment to simulate complex temperature loading environment and accurately evaluate material performance and life prediction.
Through the theoretical research, design simulation and experimental verification, the temperature control target in the process of laser thermal load is achieved in this paper. The related work and corresponding conclusions are as follows:
1. For the temperature of the laser thermal load in the high-cycle stage, some control strategies are studied and developed. By designing several different types of controllers, the simulation experiments are carried out in the simulation environment Simulink of MATLAB respectively. In the experiment, the corresponding disturbances and changes are designed, and the system of the designed controller is tested for the dynamic and static characteristics, anti-interference, robustness and other control performances. The results show that the fuzzy neural network PID controller can realize the self-tuning of parameters, and has better control performance than the conventional PID controller and fuzzy PID controller.
2. Based on the experimental platform of pulse laser thermal load, a temperature control system of laser thermal load is developed. The design and integration of the external control circuit and temperature sensor system of the laser are completed. The control software of the host computer with the flexible and friendly man-machine interface is programmed, and the fuzzy neural network PID control algorithm is realized. In order to realize the precise closed-loop control of temperature in the process of pulse laser thermal load, the foundations of software and hardware are established.
3. According to the technological characteristics of laser thermal load, the experiments of laser thermal load in the mode without the controller and the mode with the fuzzy neural network PID controller are carried out respectively. The whole process of the variable-amplitude thermal cycle is realized by changing the parameters of the pulsed laser. The temperature of the central point of the laser heating spot is detected by the infrared thermometer, and the pulse current is adjusted by the designed controller at the high cycle stage to realize the closed-loop temperature control in the process of laser thermal load. The experimental results show that the controller makes the temperature stable and controllable in the high cycle stage, and the median value of temperature oscillation has no overshoot and steady-state error, thus achieving the desired temperature control target.
A closed-loop temperature control system based on the fuzzy neural network with self-setting PID parameters is designed realized in this paper. The system has good dynamic and static characteristics, anti-interference and robustness. It is suitable for stable and fast temperature control in the laser manufacturing process including laser thermal load. The results have broad application prospects in the fields of laser thermal load performance testing or life evaluation of aerospace materials and components of the engine combustion chamber.
|李青宇. 基于模糊神经网络的激光热负荷温度控制系统研究[D]. 北京. 中国科学院大学,2019.|
|Files in This Item:|
|李青宇-学位论文终版.pdf（6713KB）||学位论文||开放获取||CC BY-NC-SA||Application Full Text|
|Recommend this item|
|Export to Endnote|
|Similar articles in Google Scholar|
|Similar articles in Baidu academic|
|Similar articles in Bing Scholar|
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.