1 Introduction
The welding process is a complex process, which has the characteristics of time-varying, nonlinear and multi-interference factors, and it is difficult to establish an accurate mathematical model. With the rapid development of modern production, the requirements for welding quality are getting higher and higher, which requires adaptive control and intelligent control of the welding dynamic process to ensure the stability of the welding process, improve the welding quality and the level of welding automation. Fuzzy control allows real-time control of the welding process without the knowledge of skilled and skilled welders without an accurate mathematical model. Experts and scholars in the welding industry at home and abroad realized earlier that fuzzy control has broad application prospects in the welding process, and actively used fuzzy control for welding tracking, welding quality and control of welding power equipment.
2 Fuzzy control of weld seam tracking
The automatic tracking of the weld is to obtain the deviation information between the torch and the center of the weld through the sensor. After processing the information, different control algorithms are used to obtain the control signal to drive the torch to center the weld. For this reason, sensors such as machinery, arc and vision have been developed at home and abroad. With the advancement of sensors and signal processing technology, multi-sensor information fusion will be combined with arc welding robot technology and widely used in automatic seam tracking.
The principle of the arc sensor is to obtain the lateral and high deviation information of the weld from the change of the arc current and voltage. When the distance between the torch and the workpiece changes, the current changes accordingly to maintain the original melting rate. Therefore, the change of arc current reflects the change of the height of the torch. The edge of the weld is scanned by the arc vibration, and the information of the lateral alignment of the torch can be obtained from the characteristics of the current waveform. The relationship between arc current and torch height variation is a time-varying nonlinear relationship, and its precise mathematical model is difficult to establish. Although domestic and foreign scholars have studied the dynamic and static models of some arc welding processes, the adaptability and robustness of these models are limited due to the strong electromagnetic interference at the welding site. Fuzzy control has good robustness and nonlinear mapping capability, so it is suitable for arc sensing tracking control.
JWKim et al. developed a set of arc sensors for CO2 gas shielded welding, using simple fuzzy control and self-organized fuzzy control methods for weld tracking. Tests show that the self-organized fuzzy controller still has strong tracking ability when the deviation angle is 10°.
Japanese scholars measure the arc current (I), voltage (U) and wire feed speed (V) to calculate the distance (H) between the groove and the torch, ie H = F (I, U, V) to control the welding Sew tracking, fuzzy logic is used for tracking control of such arc sensors.
S. Murakami et al. studied the fuzzy control of the weld tracking of arc welding robots, and adopted fuzzy rules and fuzzy controllers based on linguistic rules.
Yao Haiqing of Hohai University studied a height control system for CO2 gas shielded welding torch. The arc torch duty arc sensor was used to detect the height of the torch. The height of the torch was controlled by a fuzzy controller. The control system selected the arc duty cycle. The deviation e and the variation of the deviation ec are used as fuzzy input variables, and the torch height adjusts the number of output steps u of the stepping motor as the fuzzy output value. The test shows that the system has good control effect.
The research on the self-adjusting scale factor Fuzzy-P controller in the weld seam tracking system was carried out by Hu Shengwei of Tianjin University. Based on the theoretical analysis and experiment, the fuzzy control rules of the weld seam tracking system were determined by non-contact ultrasonic sensing. The self-adjusting scale factor Fuzzy-P controller in the seam tracking system was developed, and the hardware and software of the controller were designed. The experiment proves that the controller improves the response speed and tracking accuracy of the weld seam tracking system, which can meet the needs of practical engineering applications.
In recent years, optical sensors have received widespread attention at home and abroad due to their large amount of information, small inertia, and non-contact characteristics. Song Yonglun of South China University of Technology designed a self-organizing fuzzy controller for weld seam tracking, and cooperated with CCD vision sensing and image information processing technology to obtain a simple fuzzy ratio under TIG welding and flat panel docking conditions. The controller has better online tracking effect and better adaptability to changes in weld trajectory. Its self-organizing part has three functions: performance measurement, control quantity correction, and control rule correction. Wang Gang of Tianjin University studied the fuzzy tracking control of robotic oscillating welding based on BP. Based on the establishment of robotic oscillating welding visual tracking system, a fuzzy control system for welding rectification was constructed, and a three-layer BP neural network was proposed to simulate. The mapping relationship between fuzzy control quantities is realized, which realizes the neural network fuzzy control of robot swing welding based on BP.
The current research on CCD visual tracking system is very active at home and abroad. It combines with arc welding robot technology and represents the main development direction of automation and intelligence of arc welding process. In the motion of arc welding robots, the moment of inertia of each axis is a time-varying, strongly coupled, nonlinear complex system. The main problems that should be solved by visual sensing are: (1) miniaturization and practical use of CCD; (2) rapid and effective processing of visual information. Wavelet transform technology has shown excellent performance in noise reduction and image processing, and it is a research hotspot in the field of signal processing in the world.
3 fuzzy control of arc welding power supply
The control of arc welding power supplies is gradually becoming intelligent, and electronically controlled arc welding inverters offer the possibility of intelligent control.
Zhao Judong, from Tianjin University, proposed an adaptive fuzzy control scheme to perform on-line real-time control of pulsed MIG/MAG welding droplet transfer. An adaptive fuzzy control system based on all-digital controlled IGBT inverter arc welding power source was established. The system has been designed and controlled by the 8098 microcontroller program.
Zhu Jinhong of Luoyang Institute of Technology designed a new type of resistance spot welding device with ADuC812 microprocessor as the core.
Zhu Liumei of Huazhong University of Science and Technology used the system identification technique in fuzzy set theory to establish a 8031 ​​single-chip microcomputer control system with a welding arc fuzzy model.
Li Hejun of Gansu University of Technology used fuzzy logic to design the wire feeding system of submerged arc welding. The welding process is stable within the welding current of 300~1000A, and the control system is robust.
Wang Yasheng of Xi'an Jiaotong University has developed an intelligent control system that uses 80C196KC digital single-chip microcomputer as the core device to automatically form the optimal matching of CO2 short-circuit transition welding voltage and current through software.
Japan's T.Mita et al. used fuzzy logic to automatically set the voltage of the CO2 welder.
Japan's Panosonic has introduced the second generation of intelligent IGBT arc welding inverters AAII-350, 500 for robots. Among them, 350A achieves constant weld width control and 500A achieves constant penetration control. AAII can automatically identify the change in wire elongation when the welding current changes, and use fuzzy logic to establish the relationship between the weld width or penetration and the appropriate wire feed speed and output voltage to stabilize the weld quality. In order to improve the response speed of the fuzzy control, a 16-bit microcontroller is used. Through the process experiment, the CO2 welding short-circuit transition process is optimized to combine 4 million current waveforms as a fuzzy knowledge base to ensure the best welding specifications and achieve high-speed, high-quality welding.
4 Fuzzy control of welding quality
Due to the physical and chemical metallurgical reaction in the welding process, accompanied by the intense heat and mass transfer process, real-time detection of welding quality is difficult under strong arcing. The welding process is a system with large lag, multiple input, multiple output and intrinsic nonlinearity. There is uncertainty between the parameters of the welding process. It is difficult to establish an accurate mathematical model of the object. The conventional PID control method is often not ideal. . The skilled operator can summarize some fuzzy linguistic variables such as large or small changes in welding quality based on the weld information obtained by visual inspection. The welding quality can be accurately controlled by manual adjustment of welding speed, arc and heat input. Within the scope. Fuzzy control is based on fuzzy linguistic variables. It has been proved that the fuzzy controller is a model-independent estimator suitable for this kind of occasion. Japanese scholar Shimakenji et al. used fuzzy logic for the fusion width control of pulsed MIG welding, and established a fuzzy expert system for arc welding robots. LangariG used adaptive algorithms to modify the eigenfunctions of fuzzy subsets, and used fuzzy subsets to describe control. Rules, the content of each rule changes dynamically with the welding process. According to the change of output quantity, the rules of the line are modified to establish the self-organizing fuzzy control system of the arc welding process. G.Startke uses fuzzy logic to optimize the welding process parameters of the arc welding robot on the PC microcomputer platform. The process parameters include the welding torch. Gesture, distance from the tip to the workpiece, welding speed, welding voltage and current, wire feed speed.
Chen Qiang of Tsinghua University systematically studied the fuzzy control of the arc welding process. For the penetration control of univariate MIG welding, CCD sensor is used to extract molten pool image information, and the welder's operation experience is summarized into fuzzy control rules. Experimental research shows that the fuzzy control method of MIG welding makes a very good change in the melting width caused by heat dissipation. Good response. A preliminary study was also carried out on the fuzzy control of multivariable CO2 welding. The control objectives were: weld pool width, cooling time, splash rate, and welding efficiency. The control amount is: arc current, short circuit current, and welding speed. The fuzzy control of multivariable systems is more complicated. The design must rely on the experience of experts, and through experiments, constantly adjust the fuzzy rules, in order to improve the fuzzy control system. Huang Shisheng et al. studied a controller used in GTAW welding quality control and using fuzzy neural network for fuzzy reasoning. Two BP networks were used to realize parameter self-adjusting fuzzy and integral hybrid control, and the input was “encodedâ€. It reduces the training time of the BP network and enhances the real-time control. Simulation tests and process tests show the rationality and effectiveness of the control method. South China University of Technology combines parameter self-tuning fuzzy control with PI control and successfully used for the fusion control of GTAW welding. Zhang Jun et al. used fuzzy technology for the interpretation of welding material quality information. In order to interpret the quality information of the arc welding process online, a welding current and arc voltage signal acquisition and analysis system was developed. Through the analysis of the feature information, the knowledge support is used to judge the performance indicators of the welding process. The fuzzy comprehensive evaluation method is used to obtain the quality analysis of comprehensive consideration of multi-performance indicators.
5 Conclusion
With the rapid introduction of electronic technology, computer technology, automatic control technology and information and software technology into the field of welding, automation and intelligentization of welding production has become an important direction for the development of welding technology in the 21st century. Using the latest computer vision theory to develop the welding robot vision sensing and control technology, the development of intelligent welding robots that can identify the target environment, accurately track the trajectory in real time and adjust the welding parameters has become one of the important development trends in the welding field. The traditional control method and the intelligent control method are combined, and the traditional PID control is performed by using the switch during the coarse adjustment, and the intelligent control is used in the case of nonlinear, multi-coupling and no accurate mathematical model. It has been proved theoretically that fuzzy control can approximate any nonlinear function with arbitrary precision, but it is limited by the current technical level. The binning of fuzzy variables and the number of fuzzy rules are limited. The determination of membership functions has not yet unified theoretical guidance. With certain human factors, the accuracy of fuzzy control needs to be improved. The trend of fuzzy logic and neural network and expert system fusion shows the strong vitality of fuzzy logic, neural network and expert system complement each other. The combination of fuzzy logic and expert system can make full use of expert system knowledge reasoning mechanism and knowledge extraction ability, and fuzzy control technology will become the core technology of the 21st century.
Bathroom Accessories Towel Rod Rack
KaiPing HuiPu Shower Metalwork Industrial CO,LTD , https://www.hp-shower.com