| As an important method of thermal processing, resistance spot welding has many advantages, such as a good quality, high efficiency, low cost, easy to realize automation etc. And therefore in the automotive, aerospace, rail passenger cars, electric components, industrial production has been widely used. About 30% of the world's total annual welding volume is done by the pressure welding method. With the increasing automation of manufacturing technology to improve the production process stability and reliability of spot weld quality also put forward higher requirements, therefore, how to precise control to monitor the resistance spot welding process, in order to meet the need of modern production has become a hotspot research of resistance spot welding quality control.Especially with the rapid development of computer technology, spot welding quality control has been constantly changing from the single parameter to the multi-parameter, a monitor mode to the multi-monitor mode, and the traditional decision-making methods to artificial intelligence decision-making. On the basis of the co-control strategies, the algorithm of heat regulation is optimized for resistance spot welding by use of fuzzy control and single-chip technology. In the nucleation stage, the use of high-quality welding heat in the process of nuclear fusion have shown two characteristics-dynamic current lag angle characteristic and dynamic current characteristic, shaped the enhanced given DRC and AM current parameters which has the feature of first up and later down, make both of them as real-time control basis. then the twice fuzzy decision-making method is discussed about multi-information fusion technology according to a lot of information, such as"current lag angle incremental deviation en","heat percentage adjustment incrementâ–³mn in present cycle"and"heat percentage adjustment incrementâ–³mn+1 in next cycle","welding current deviationâ–³In in present cycle". In the nucleation stage, use current that is an important factor to affect the quality of welding, enrich the quality control elements, so that make the adjustment scale and direction more reasonable.The double-characteristic fusion control algorithm is development and tested on the IDRC spot welding intelligent controller, ATmega168 microcontroller is used as the main control unit, which integrates large capacity memory and a rich powerful hardware interface circuit, programmable strong, so provides a reliable hardware support for the new algorithm. Using the system software in this monitor can realize simple regular method, constant current control method and one characteristic cooperative control method. In addition, the control system can realize to storage and retrieval the dynamic data about heat percentage, welding current and current lag angle for each nugget through the LCD panel. This provides a lot of useful data for analysis of the results.Software is the soul of control system, and the control algorithm is the core of software design. Therefore, this paper focuses on the optimal design of fuzzy controllers FLC1 and FLC2. Structure of fuzzy controller with the twice inference function and dual input / single output feature is selected, variables of FLC1 and FLC2 input / output is identified. In a reasoning, used incremental dynamic current angle deviation en and heat wave amplitudeâ–³mn of this cycle as input variables, the former directly reflects the deviation between actual DRC curve and given DRC curve. The latter provides useful reference information for fuzzy reasoning. Both have a causal relationship in nature, thus increased decision-making accuracy for output variablesâ–³mn+1. The first input of fuzzy controller FLC2 selected the output variableâ–³mn+1, it directly reflects the results of the current lag angle ware this cycle working on the next incremental frequency deviation of the output. Second input variablesâ–³In is the deviation between the given current at this cycle and the actual welding current, it reflects the deviation between the level of welding current this cycle and the ideal welding current. Precisely, it has a direct characterization of the energy level of thermal process, as a basis using fuzzy control rules can more timely, objective and effective correct the final outputâ–³m*n+1. In this paper, the structure of FLC1 and FLC2 are two-dimensional fuzzy controller, the fundamental aim is to reduce the difficulty of fuzzy inference. In software design features full use of subroutines, reasonable call, not only simplifies the reasoning process, but also greatly increased the rate of the program, so reflects the software design ingenuity.In order to avoid the blindness of heat regulation, after production of the outputâ–³m*n+1 of the second decision-making, pushed it again to compare with the previous outputâ–³mn, when the both had a wide margin in the reverse operating condition, to suppress the oscillating behavior of thermal regulation, halve the operation or inhibit the operation of the control strategy in order to ensure the convergence of heat regulation, improve the dynamic quality of thermal process, reflect the intelligent of the software better.The results show this multi-information fusion control mechanism has greater advantages and more desirable results than single constant current control and IDRC control method with one characteristic. Even in the case of fluctuations of welding heat or current, particularly in the presence of shunt situation, this algorithm also can guarantee the consistency of nugget quality and welding time, and can achieve a good quality of the dynamic adjustment. At the same time it provides a new idea for other steel spot welding control, and has important value in theory and application.
|
PDF Full Text Request