Adaptive Fuzzy Control System for Inverter Arc Welding Power Supplies

Abstract

This invention provides an adaptive fuzzy control system for inverter arc welding power supplies, belonging to the field of welding process control and equipment automation. Specifically, it includes a sensor module that collects working values ​​corresponding to key parameters during the welding process in real time; a data processing unit that converts the working values ​​into fuzzy sets; a fuzzy controller that performs fuzzy inference on the fuzzy sets based on a preset fuzzy rule base to output a fuzzy control signal; a data processing unit that defuzzifies the fuzzy control signal to convert it into a precise control signal; an inverter power supply module that receives the defuzzified precise control signal and adjusts the output power of the welding power supply; a feedback system that monitors real-time parameters of weld quality during the welding process and transmits these parameters to an adaptive module to achieve closed-loop control; and an adaptive module that dynamically adjusts the fuzzy rule base and membership function based on the welding quality information provided by the feedback system. This application's processing scheme improves the system's control accuracy and adaptability.

Description

Technical Field

[0001] This invention relates to the field of welding process control and equipment automation, specifically to an adaptive fuzzy control system for inverter arc welding power supplies. Background Technology

[0002] Welding technology is widely used in the field of metal structure manufacturing, especially in industries such as automobile manufacturing, shipbuilding, construction, and aerospace, where arc welding technology is widely applied. Inverter arc welding power supplies play a crucial role in the modern welding industry. Through inverter technology, they effectively convert direct current into adjustable alternating current, thereby achieving precise control of welding current and voltage to adapt to different materials and complex welding conditions.

[0003] Traditional welding power supply control systems mostly rely on the classic PID control strategy. While this strategy performs well with linear systems, its performance is relatively limited when faced with the complex nonlinearities and external disturbances encountered in actual welding processes. The characteristics of the electric arc during welding are affected by various factors, including material properties, ambient temperature variations, gas flow, and the shape and position of the workpiece. These factors lead to a nonlinear relationship between welding current and voltage. Traditional PID controllers struggle to handle these complexities, often requiring frequent adjustments to control parameters, increasing system complexity and instability. Furthermore, external disturbances during welding (such as changes in gas flow and workpiece shape) affect arc stability. PID controllers respond slowly to these disturbances, failing to adjust the welding power supply output in a timely manner, resulting in unstable weld quality.

[0004] In modern manufacturing, the diversification of welding materials and the increasing complexity of welding processes require control systems with strong adaptability. These systems must be able to automatically adjust control strategies based on different welding materials, processes, and environmental conditions to ensure consistent welding quality. Existing technological solutions lack sufficient adaptability to different welding conditions and cannot meet these requirements. Summary of the Invention

[0005] Therefore, in order to overcome the shortcomings of the prior art, the present invention provides an adaptive fuzzy control system for inverter arc welding power supply. This system has significant technical advantages and effects over the prior art in terms of control accuracy, adaptability, intelligence level, modular system design and wide application, and brings significant technical improvement to the welding technology and industrial control fields.

[0006] To achieve the above objectives, this invention provides an adaptive fuzzy control system for an inverter-type arc welding power supply, comprising a sensor module, a data processing unit, a fuzzy controller, an adaptive module, an inverter power supply module, and a feedback system. The sensor module collects real-time operating values ​​corresponding to key parameters during the welding process, including at least arc current, voltage, welding speed, and ambient temperature. The data processing unit converts the real-time collected operating values ​​into fuzzy sets. The fuzzy controller performs fuzzy inference operations on the fuzzy sets based on a preset fuzzy rule base and comprehensively judges various key parameters to output a fuzzy control signal. The processing unit defuzzifies the fuzzy control signal and converts it into a precise control signal. The inverter power supply module receives the defuzzified precise control signal and adjusts the output power of the welding power supply through pulse width modulation to ensure the stability of the welding arc. The feedback system monitors real-time parameters of the weld quality during the welding process, including at least the weld flatness and surface quality, and transmits these real-time parameters to the adaptive module to achieve closed-loop control. The adaptive module dynamically adjusts the shape of the fuzzy rule base and membership function based on the welding quality information provided by the feedback system to optimize control performance and enhance the system's adaptability and robustness.

[0007] In one embodiment, the data processing unit performs fuzzification using at least one membership function, wherein the membership function is at least one of the following: triangular membership function, trapezoidal membership function, Gaussian function, and bell function.

[0008] In one embodiment, the fuzzy inference method used by the fuzzy controller is an inference method selected based on the characteristics of multiple input variables during the welding process. The inference method is at least one of the maximum-minimum synthesis method, the weighted average method, and the maximum membership method.

[0009] In one embodiment, the adaptive module employs an adaptive adjustment strategy, which is at least one of a neural network, a genetic algorithm or a fuzzy neural network, a rule base for dynamically optimizing a fuzzy controller, and membership function parameters.

[0010] In one embodiment, the feedback system includes a data fusion unit for fusing feedback data from multiple sensors to eliminate the impact of noise and uncertainty on the control effect. The data fusion unit uses Kalman filtering or Bayesian estimation for fusion.

[0011] In one embodiment, the inverter power module automatically adjusts the pulse width modulation parameters using a dynamic response adjustment mechanism based on the real-time changing arc characteristics during the welding process, in order to improve the dynamic response characteristics and control accuracy of the power supply.

[0012] In one embodiment, the fuzzy rule base includes "if-then" rules designed based on expert experience and experimental data, and the construction steps of the fuzzy rule base include fuzzy rule extraction, rule optimization, and redundant rule deletion.

[0013] In one embodiment, the system further includes a fault diagnosis unit that works in conjunction with the inverter power module. This fault diagnosis unit is used to monitor the operating status of the power module and perform fault prediction and self-repair to ensure the long-term stable operation of the welding system.

[0014] In one embodiment, the real-time parameters include at least one of welding speed, weld position, and weld angle.

[0015] In one embodiment, the adaptive module employs a multi-objective optimization algorithm to comprehensively consider welding quality, control energy consumption, and system response time, thereby achieving global optimization of the control strategy.

[0016] Compared with existing technologies, the advantages of this invention lie in combining the strengths of adaptive control technology and fuzzy logic control methods to develop an intelligent control system suitable for inverter arc welding power supplies. This system can effectively address complex nonlinear and dynamic changes during the welding process, improving welding quality and production efficiency. This concept not only provides new ideas for the development of welding technology but can also be extended to other industrial processes requiring high-precision control, possessing broad application prospects and research value. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the overall architecture of the adaptive fuzzy control system for an inverter arc welding power supply in an embodiment of the present invention;

[0019] Figure 2 This is a circuit block diagram of the inverter arc welding power supply in an embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of the workflow of the adaptive fuzzy control system in an embodiment of the present invention;

[0021] Figure 4 This is a structural diagram of the control fuzzy generator in an embodiment of the present invention. Detailed Implementation

[0022] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0023] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] It should be noted that the following description covers various aspects of embodiments within the scope of protection of this invention. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this application, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number and aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using other structures and / or functionalities besides one or more of the aspects set forth herein.

[0025] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0026] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0027] The application of adaptive fuzzy control in inverter arc welding power supplies can be traced back to the widespread use of inverter arc welding power supplies in the industrial welding field. Because inverter technology can efficiently convert input AC power into high-frequency DC power, thereby achieving precise control of welding current and voltage, this equipment has become the preferred choice for various welding processes due to its excellent energy conversion efficiency and arc control capabilities. By combining fuzzy control with adaptive adjustment, control parameters can be dynamically adjusted, enabling the system to automatically adapt to changes in welding conditions, thus maintaining arc stability and welding quality.

[0028] Significant progress has been made in adaptive fuzzy control technology. By combining modern intelligent algorithms, such as machine learning and deep learning, the system possesses self-learning and self-optimization capabilities, automatically adjusting control parameters and fuzzy rule bases to cope with uncertainties and nonlinear changes in the welding process. With the improvement of hardware computing power, the real-time performance of adaptive fuzzy control has been greatly improved, enabling dynamic adjustment of the arc during welding, thereby ensuring welding stability and high quality. Furthermore, adaptive fuzzy control is widely used in automated welding systems, especially in welding robot path tracking and arc length adjustment. Through fuzzy processing and adjustment of sensor data, precise control of the welding trajectory is achieved.

[0029] like Figure 1 As shown, this invention proposes an adaptive fuzzy control system for inverter arc welding power supplies, aiming to achieve effective control of dynamic changes and complex working conditions during the welding process through a combination of fuzzy logic and adaptive adjustment. This adaptive fuzzy control system consists of six main modules: a sensor module, a data processing unit, a fuzzy controller, an adaptive module, an inverter power supply module, and a feedback system. These modules are interconnected through data transmission and feedback mechanisms to achieve closed-loop control of the welding process.

[0030] The sensor module is used to collect the working values ​​of key parameters in the welding process in real time. The key parameters include at least arc current, voltage, welding speed and ambient temperature.

[0031] The data processing unit converts the real-time acquired working values ​​into fuzzy sets. It also defuzzifies the fuzzy control signals, converting them into precise control signals.

[0032] The fuzzy controller performs fuzzy inference operations on the fuzzy set based on a preset fuzzy rule base, and makes comprehensive judgments on a variety of key parameters to output a fuzzy control signal.

[0033] The inverter power module receives the defuzzified precise control signal and adjusts the output power of the welding power source through pulse width modulation to ensure the stability of the welding arc.

[0034] The feedback system monitors real-time parameters of weld quality during the welding process. These real-time parameters include at least the weld flatness and surface quality. The system then transmits these real-time parameters to the adaptive module to achieve closed-loop control.

[0035] The adaptive module dynamically adjusts the shape of the fuzzy rule base and membership function based on the welding quality information provided by the feedback system to optimize control performance and enhance the system's adaptability and robustness.

[0036] The workflow of this adaptive fuzzy control system is as follows:

[0037] The sensor module collects the working values ​​of key parameters in the welding process in real time. The key parameters include at least arc current, voltage, welding speed and ambient temperature.

[0038] The data processing unit converts the real-time collected working values ​​into fuzzy sets;

[0039] The fuzzy controller performs fuzzy inference operations on the fuzzy set based on a preset fuzzy rule base, and makes comprehensive judgments on a variety of key parameters to output a fuzzy control signal;

[0040] The data processing unit defuzzifies the fuzzy control signal and converts it into a precise control signal;

[0041] The inverter power module receives the defuzzified precise control signal and adjusts the output power of the welding power source through pulse width modulation to ensure the stability of the welding arc.

[0042] The feedback system monitors the real-time parameters of weld quality during the welding process. The real-time parameters include at least the flatness and surface quality of the weld. The system then transmits these real-time parameters to the adaptive module to achieve closed-loop control.

[0043] The adaptive module dynamically adjusts the shape of the fuzzy rule base and membership function based on the welding quality information provided by the feedback system to optimize control performance and enhance the system's adaptability and robustness.

[0044] This embodiment integrates an adaptive fuzzy control system for inverter arc welding power supplies, covering aspects such as system design, control strategy, modular structure, dynamic adjustment, and self-learning mechanism. Optimizations have been made in the following areas:

[0045] 1. Application of Adaptive Fuzzy Control Strategy: By introducing adaptive fuzzy control, precise control of the nonlinear system during welding was achieved. Adaptive fuzzy control combines the advantages of fuzzy logic in handling nonlinear and uncertain problems with the online optimization characteristics of adaptive control, thereby enhancing the system's adaptability under dynamically changing welding conditions.

[0046] 2. Real-time self-learning and dynamic adjustment mechanism: A self-learning algorithm is used to train and update the control system online in real time, enabling adaptive adjustment of control parameters. Through machine learning methods, the system can continuously "learn" the characteristics of the welding arc state during the welding process and adjust the control strategy according to these characteristics to gradually improve the accuracy and response speed of the control system.

[0047] 3. Modular System Architecture Design: The modular design of the system ensures that each submodule has independent functions and clear interfaces, guaranteeing the system's flexibility and scalability. The system consists of multiple functional units, including sensor modules, data processing units, fuzzy controllers, adaptive modules, inverter power supply modules, and feedback systems. These modules work collaboratively through standard data interfaces, forming a tightly coupled closed-loop control system. This modular architecture not only improves the system's maintainability but also facilitates independent optimization of each module, thereby enhancing the overall system performance.

[0048] 4. Enhanced System Anti-interference Capability: Interference in the arc welding process mainly originates from the characteristics of welding materials, environmental noise, and arc instability. Fuzzy processing and adaptive adjustment enhance the system's anti-interference performance. The fuzzy controller uses membership functions to fuzzify the effects of noise and interference on the input signal, thereby reducing the accuracy requirements of the system. The adaptive module adjusts the fuzzy rules in real time based on changes in feedback data, enabling the system to maintain precise control of the welding process even in the presence of significant noise or interference.

[0049] In one embodiment, the data processing unit performs fuzzification using at least one membership function, which is at least one of triangular membership function, trapezoidal membership function, Gaussian function, and bell function.

[0050] In one embodiment, the fuzzy inference method used by the fuzzy controller is selected based on the characteristics of multiple input variables during the welding process. The inference method is at least one of the maximum-minimum synthesis method, the weighted average method, and the maximum membership method.

[0051] In one embodiment, the adaptive module employs an adaptive adjustment strategy, which is at least one of a neural network, a genetic algorithm or a fuzzy neural network, a rule base for dynamically optimizing the fuzzy controller, and membership function parameters.

[0052] In one embodiment, the feedback system includes a data fusion unit for fusing feedback data from multiple sensors to eliminate the impact of noise and uncertainty on the control effect. The data fusion unit uses Kalman filtering or Bayesian estimation for fusion.

[0053] In one embodiment, the inverter power module automatically adjusts the pulse width modulation parameters using a dynamic response adjustment mechanism based on the real-time changing arc characteristics during the welding process, in order to improve the dynamic response characteristics and control accuracy of the power supply.

[0054] In one embodiment, the fuzzy rule base includes "if-then" rules designed based on expert experience and experimental data, and the construction steps of the fuzzy rule base include fuzzy rule extraction, rule optimization, and redundant rule deletion.

[0055] In one embodiment, the system further includes a fault diagnosis unit that works in conjunction with the inverter power module. This fault diagnosis unit is used to monitor the operating status of the power module and perform fault prediction and self-repair to ensure the long-term stable operation of the welding system.

[0056] In one embodiment, the real-time parameters include at least one of welding speed, weld position, and weld angle.

[0057] In one embodiment, the adaptive module employs a multi-objective optimization algorithm to comprehensively consider welding quality, control energy consumption, and system response time, thereby achieving global optimization of the control strategy.

[0058] Example 1

[0059] The sensor module collects real-time operating values ​​of key parameters during the welding process. The sensor module is designed using a multi-sensor fusion strategy to achieve real-time monitoring of various key parameters during welding. Specifically, the module includes current sensors, voltage sensors, temperature sensors, and speed sensors. By acquiring high-precision data such as arc current, voltage, welding speed, and ambient temperature, the system can respond quickly to changes in welding conditions. To improve data acquisition accuracy, the sensor sampling frequency is set to above 1kHz, enabling the capture of rapidly changing parameter information during welding. The results of multi-sensor fusion are preprocessed using a data fusion algorithm to reduce noise and enhance the system's anti-interference capability.

[0060] The data processing unit is responsible for fuzzifying and defuzzifying the working values ​​collected by the sensor module, and is one of the core components of the system's fuzzy logic control. The fuzzification process, based on a predefined membership function, converts precise input data into a fuzzy set. The membership function is designed using triangular, trapezoidal, or Gaussian membership functions to accommodate data fluctuations under different operating conditions. The defuzzification step converts the fuzzy controller's output signal into a specific numerical control signal, making the control signal's numerical changes smooth and adjustable, thereby ensuring the response characteristics of the welding power supply and the quality of the weld.

[0061] The fuzzy controller performs fuzzy inference operations on fuzzy sets based on a preset fuzzy rule base, and comprehensively judges various key parameters to output fuzzy control signals. The fuzzy controller realizes logical reasoning and decision-making in the welding process through the preset fuzzy rule base. The rule base consists of several "if-then" rules, which can cover the control requirements of various working conditions in the welding process. These rules are based on the fuzzy values ​​of input variables (such as current and voltage) to generate suitable control signals. This is generally implemented through membership functions, which can take the form of triangular membership functions, trapezoidal membership functions, or Gaussian functions.

[0062] The triangular membership function is one of the most common fuzzy membership functions, suitable for most applications. Its basic form is:

[0063]

[0064] Here, a, b, c, and d are the four parameters that define the membership function, representing the reference points on the left and right sides of the triangle, and the location of the peak, respectively.

[0065] Trapezoidal membership functions are an extension of triangular membership functions, featuring flat regions at both ends to accommodate a wider range of inputs. Their basic form is:

[0066]

[0067] Among them, a, b, c, and d are the four key points of the trapezoid, which define the left and right slopes and the flat part in the middle of the trapezoid.

[0068] The Gaussian membership function can better simulate the fuzziness of continuous changes. The formula for the Gaussian membership function is as follows:

[0069]

[0070] Where c is the center point of the Gaussian function, and σ is the standard deviation of the voltage, used to control the width of the fuzzing.

[0071] Fuzzy rules are derived using an inference engine to obtain fuzzy output results. A commonly used inference method is based on min-max operations. The fuzzy inference result for each rule can be expressed as:

[0072]

[0073] in, and It is the membership degree of the input variable in rule j.

[0074] Different reasoning methods may have different ways of activating and combining rules. The specific reasoning process will select the most appropriate method according to the control requirements and application scenarios, so that the system has good adaptability and robustness when dealing with complex welding states.

[0075] The adaptive module dynamically adjusts the parameters of the fuzzy rule base and membership function based on the welding quality information provided by the feedback system to optimize control performance and enhance the system's adaptability and robustness. The adaptive module employs a parameter optimization method based on machine learning algorithms to achieve dynamic adjustment of the fuzzy rule base and membership function. The adaptive algorithm can use evolutionary computation methods such as neural networks or genetic algorithms to adjust the parameters of the fuzzy controller based on real-time data from the feedback system, ensuring the system maintains optimal performance when welding conditions change. The core of adaptive adjustment lies in adjusting the control strategy in real time based on welding quality parameters (such as weld smoothness, penetration depth, and spatter amount) provided by the feedback system.

[0076] The adaptive module typically calculates the system error and adjusts the parameters of the fuzzy control system based on the error. The formula for the error can be expressed as:

[0077] e(t) = y desired (t)-y(t)

[0078] Among them, y desired y(t) is the desired output and y(t) is the actual output. These error values ​​are used as inputs to the fuzzy control rules.

[0079] Adaptive parameter updates adjust the parameters of the fuzzy controller (such as the shape of the membership function or the rule weights) to minimize the error, as expressed in the formula:

[0080]

[0081] Where, θ i These are the parameters of the fuzzy controller, where E(t) is the error function and α is the learning rate. This is the gradient of the error function relative to the control parameters. By using gradient descent, the controller parameters (such as the parameters of the membership function) are adjusted based on the real-time error to optimize the control performance.

[0082] This adaptability gives the system strong robustness and anti-interference ability, enabling it to maintain high-quality welding results even under complex and nonlinear welding conditions.

[0083] The inverter power supply module, acting as the control signal execution unit, receives the defuzzified precise control signal and adjusts the inverter's output power through pulse width modulation (PWM). The module's design must consider the response speed and stability of the control signal to achieve precise control of the welding arc. The PWM control strategy adjusts the power supply's output current or voltage by varying the pulse width, effectively compensating for and regulating parameter fluctuations during the welding process, thereby optimizing weld formation and quality. A complete circuit diagram of the inverter-type arc welding power supply, including the rectifier circuit, DC-Link capacitor, inverter, fuzzy control, and drive circuit, is shown in the circuit block diagram below. Figure 2 .

[0084] The main function of the feedback system is to monitor the real-time parameters of weld quality during the welding process and transmit this information back to the adaptive module. Feedback data can include key indicators such as weld formation and surface quality, used to evaluate the current control effectiveness and adjust the control strategy. Through a closed-loop control mechanism, the feedback system forms a continuous cycle of data acquisition, processing, and feedback, enabling the system to self-optimize and adjust based on the actual welding conditions.

[0085] As attached Figure 3 As shown, the operation flow of the adaptive fuzzy control system in this embodiment includes the following steps:

[0086] 1. Data Acquisition and Fuzzy Processing: After system startup, the sensor module acquires welding current, voltage, speed, and temperature data in real time. The data processing unit fuzzifies the acquired precise data, converting it into corresponding fuzzy sets for use in fuzzy inference.

[0087] 2. Fuzzy Control Inference and Adaptive Adjustment: The fuzzy controller performs inference based on fuzzy data and a preset rule base, outputting a fuzzy control signal. The inference process comprehensively considers the interaction relationships of multiple input variables to generate a fuzzy output that meets welding requirements. The adaptive module dynamically adjusts the rule base and membership function based on real-time data provided by the feedback system to optimize the control effect.

[0088] 3. Defuzzification and Inverter Power Supply Control: The defuzzification process converts the fuzzy signal output by the fuzzy controller into specific numerical control commands and sends them to the inverter power supply module to adjust the output current or voltage of the power supply so that the state of the welding arc meets the desired welding process parameters.

[0089] 4. Feedback Monitoring and Closed-Loop Adjustment: The feedback system monitors the actual weld quality during the welding process and transmits the monitoring data to the adaptive module. The adaptive module continuously optimizes the control strategy based on the feedback data, forming a data-driven closed-loop control to ensure the stability of the welding process and the weld quality.

[0090] The adaptive fuzzy control system in this embodiment is applicable to various welding scenarios and working conditions. Specific application areas include, but are not limited to, the following:

[0091] The system dynamically adjusts welding parameters based on real-time collected parameter data to adapt to rapid changes in welding conditions. When the system detects changes in weld quality, it automatically adds new control rules or modifies existing ones, reducing the complexity of the rule base and improving control efficiency.

[0092] Various workpiece materials and qualities meet the requirements; for workpieces of different materials and thicknesses, the parameters of the welding power supply are automatically adjusted according to the material characteristics to ensure that the weld formation and quality meet the requirements.

[0093] Complex welding trajectories and multi-axis linkage welding; suitable for complex trajectory welding scenarios on robotic welding or CNC machine tools. By integrating with a motion control system, this system can precisely control parameters such as speed and direction of the welding path to achieve high-quality automatic welding.

[0094] As attached Figure 4 As shown, the control fuzzy controller structure diagram of this embodiment includes the following parts:

[0095] Fuzzy processing: Mapping precise input signals (such as current, voltage, temperature, etc.) to the membership degrees of fuzzy sets for subsequent fuzzy reasoning and decision-making processes, such as "low", "medium", "high", etc.

[0096] Fuzzy reasoning: By applying a preset fuzzy rule base, reasoning is performed on fuzzy sets, such as "if the voltage is high and the current is low, then the output decreases".

[0097] Defuzzification: Converting the results of fuzzy inference into specific control signals, such as output current or voltage adjustment values.

[0098] This embodiment has been optimized in the following aspects:

[0099] 1. Improved precision and stability of welding control: This embodiment effectively solves this problem by utilizing adaptive fuzzy control technology. The fuzzy controller can handle complex nonlinear input-output relationships and generate control signals through fuzzy rule reasoning. Combined with the adaptive module, the system can adjust the fuzzy rules and membership functions in real time, enabling the control strategy to automatically optimize as the arc state changes, thereby achieving more precise control.

[0100] 2. Strong dynamic adaptability, accommodating diverse welding conditions: Traditional welding control methods typically require manual adjustment of control parameters to cope with changing welding conditions. This process is cumbersome and prone to errors. This invention, by introducing an adaptive module, achieves automatic optimization and real-time adjustment of control parameters. The system can dynamically adjust the control strategy based on voltage and current fluctuations during welding, maintaining excellent control performance even when welding materials, ambient temperature, or welding speed change. This dynamic adaptability enhances the system's practicality in complex welding tasks.

[0101] 3. Significantly improves the system's anti-interference capability and robustness; the arc welding process is easily affected by external factors, such as environmental noise and changes in workpiece material. Existing linear control methods often exhibit unstable control performance when faced with such interference. This invention combines fuzzy control and adaptive algorithms to give the system stronger anti-interference capabilities. The fuzzy controller can fuzzify the input variables, mitigating control deviations caused by external disturbances, while the adaptive module can quickly adjust control parameters, further enhancing the system's robustness and stability.

[0102] 4. Reduced manual intervention and enhanced system intelligence: Compared to existing technologies that require frequent manual adjustments and experience-based parameter settings, the control system of this invention possesses a high degree of intelligence. Through a self-learning mechanism, the adaptive module can automatically optimize the control strategy based on feedback data during the welding process, reducing the need for operator intervention. This not only improves the system's autonomy and intelligence but also reduces reliance on operator skill levels, making the system easier to operate and maintain.

[0103] 5. Modular design, easy to expand and maintain: This invention adopts a modular system design, dividing the control system into independent parts such as sensor modules, data processing units, fuzzy controllers, adaptive modules, inverter power supply modules, and feedback systems. Compared with existing integrated control system designs, this modular structure allows each module to be upgraded and maintained independently, enhancing the system's flexibility and scalability. Users can adjust or replace specific modules according to specific industrial application needs, facilitating system upgrades and technology updates.

[0104] 6. Wide applicability, suitable for various industrial applications: Existing welding control technologies are mostly designed for specific welding processes or equipment, with limited applicability. The adaptive fuzzy control system of this invention has wide applicability; besides being applied to inverter arc welding power supplies, it can also be extended to other nonlinear industrial control processes, such as cutting, heat treatment, and the control of power electronic equipment. Its versatility makes the system have broader application prospects in the field of industrial automation.

[0105] This embodiment has significant technical advantages and effects over existing technologies in terms of control precision, adaptability, intelligence level, modular system design and wide application, bringing significant technological advancements to the fields of welding technology and industrial control.

[0106] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An adaptive fuzzy control system for an inverter arc welding power supply, characterized in that, It includes a sensor module, a data processing unit, a fuzzy controller, an adaptive module, an inverter power supply module, and a feedback system. The sensor module collects working values ​​corresponding to key parameters during the welding process in real time. The key parameters include at least arc current, voltage, welding speed and ambient temperature. The data processing unit converts the real-time collected working values ​​into fuzzy sets; The fuzzy controller performs fuzzy inference operations on the fuzzy set based on a preset fuzzy rule base, and makes comprehensive judgments on various key parameters to output a fuzzy control signal; The data processing unit defuzzifies the fuzzy control signal and converts it into a precise control signal; The inverter power module receives the defuzzified precise control signal and adjusts the output power of the welding power source through pulse width modulation to ensure the stability of the welding arc. The feedback system monitors real-time parameters of weld quality during the welding process. These real-time parameters include at least the weld flatness and surface quality. The system then transmits these real-time parameters to the adaptive module to achieve closed-loop control. The adaptive module dynamically adjusts the parameters of the fuzzy rule base and membership function based on the welding quality information provided by the feedback system to optimize control performance and enhance the system's adaptability and robustness. The fuzzy inference method used by the fuzzy controller is selected based on the characteristics of multiple input variables in the welding process. The inference method is at least one of the maximum-minimum synthesis method, the weighted average method, and the maximum membership method. The feedback system includes a data fusion unit, which is used to fuse feedback data from multiple sensors to eliminate the impact of noise and uncertainty on the control effect. The data fusion unit uses Kalman filtering or Bayesian estimation for fusion. The fuzzy rule base includes "if-then" rules designed based on expert experience and experimental data. The construction steps of the fuzzy rule base include fuzzy rule extraction, rule optimization, and redundant rule deletion.

2. The adaptive fuzzy control system according to claim 1, characterized in that, The data processing unit performs fuzzification using at least one membership function, which is at least one of the following: triangular membership function, trapezoidal membership function, Gaussian function, and bell function.

3. The adaptive fuzzy control system according to claim 1, characterized in that, The adaptive module employs an adaptive adjustment strategy, which is at least one of a neural network, a genetic algorithm or a fuzzy neural network, a rule base for dynamically optimizing the fuzzy controller, and membership function parameters.

4. The adaptive fuzzy control system according to claim 1, characterized in that, The inverter power module automatically adjusts the pulse width modulation parameters using a dynamic response adjustment mechanism based on the real-time changes in arc characteristics during the welding process, thereby improving the dynamic response characteristics and control accuracy of the power supply.

5. The adaptive fuzzy control system according to claim 1, characterized in that, Also includes: A fault diagnosis unit that works in conjunction with the inverter power module is used to monitor the operating status of the power module and perform fault prediction and self-repair to ensure the long-term stable operation of the welding system.

6. The adaptive fuzzy control system according to claim 1, characterized in that, The real-time parameters include at least one of welding speed, weld position, and weld angle.

7. The adaptive fuzzy control system according to claim 1, characterized in that, The adaptive module employs a multi-objective optimization algorithm that comprehensively considers welding quality, control energy consumption, and system response time to achieve global optimization of the control strategy.

Images