Research Status of AC/DC Arc Welding Inverter (3)
4 Intelligentization of AC and DC arc welding control
Microcomputer controlled arc welding power supplies have been widely used in the field of welding. The traditional control method of the inverter power supply relies on the description of the mathematical model and parameters of the power system. The development of power inverter technology shows that the model structure and parameters of the inverter power supply are closely related to many complicated factors. The change in model structure and parameters is mainly caused by changes in the continuous or discontinuous mode of the inverter power supply, the circuit parameters related to the operating temperature, the voltage of the output grid, and the random non-linear load. In addition, the welding process is a complex process. The output characteristics of the welding power source are nonlinear, time-varying, and uncertain controlled objects. It has characteristics such as time-varying, nonlinear and multi-interference factors, so it is difficult to establish an accurate mathematical model. It is often difficult to establish a mathematical model that accurately and comprehensively reflects the static and dynamic characteristics of an inverter power system using transmission control theory. The development of modern control technologies, especially intelligent control technologies, provides a way to solve these problems. At present, the more mature methods mainly include expert systems, neural networks and fuzzy control methods [14, 15].
The expert system is based on the professional knowledge and experience of control theory, and uses the knowledge representation and reasoning technology of the artificial intelligence expert system to derive the action control system. In a sense, an expert system is a system of knowledge acquisition and utilization. At present, it has achieved some results in the design, modeling and waveform control of inverter power supply. However, its acquisition is slow and it is difficult to meet the real-time control requirements of fast time-varying systems.
The neural network control method can be seen as a computing system for parallel input and parallel output distribution. Due to the special form of nonlinear parallel processing, its most prominent advantage is its ability to approximate any L2 nonlinear function. Using the neural network method, the control problem can be seen as a pattern recognition problem. It is generally believed that systems designed with neural networks are adaptable, intelligent, and robust, and can handle the control problems of complex industrial processes with high dimensionality, nonlinearity, strong interference, uncertainty, and difficult modeling. It has been used in arc welding power supply and welding process control. The main problems are how to improve the learning speed, avoid local optimization and achieve global optimality.
The rapid development of fuzzy control theory and technology provides a new and effective tool to solve the above problems. Fuzzy control is a kind of computer numerical control method based on fuzzy set theory, fuzzy linguistic variables and fuzzy logic reasoning. It is a kind of nonlinear control, which belongs to the category of intelligent control and is an important and effective way to realize intelligent control. form. The key part of fuzzy control is the fuzzy inference synthesis operation. The application of fuzzy automatic control technology to high-power inverter power has unique advantages, so that the design process can establish fuzzy control rules and other processes without relying on the mathematical model structure and parameters of the power system, and can comprehensively improve the high-power inverse. The purpose of variable power control for fastness, stability, accuracy and robustness. The United States, Japan, Germany and China have successively introduced fuzzy chips, fuzzy microcontrollers and related development tools. This implementation directly implements fuzzy logic operations and reasoning through hardware, which can greatly improve the speed of inference and control accuracy. However, its performance price is not ideal enough. The number of input, output and fuzzy rules of the fuzzy chip is limited, which is not flexible enough compared with the software implementation. However, with the rapid development of fuzzy microprocessor technology, it will bring vitality and vitality to this hardware implementation.
5 Conclusion
The development of computer technology, network technology, control technology and power electronics technology has guaranteed the development of intelligent arc welding power sources. The use of computer storage functions and high-speed, high-precision data processing capabilities can promote the development of multi-functional and intelligent welding machines.
In the arc welding power supply, modern control methods such as expert system, neural network control and fuzzy control are introduced to optimize the parameters and control the welding quality to further improve the performance and adaptability of the welding machine. Through digital and intelligent control technology to ensure the stability, reliability, high efficiency and welding quality of arc welding power supply is the development direction of modern AC and DC arc welding power supply.
Previous Next
Aluminium Coil,Aluminium Coil for Construction,Mill Finish Aluminium Coil
Huawei Aluminum Co., Ltd. , http://www.hnaluminumcoil.com