CLADDING TECHNOLOGY SHANXI CO., LTDFuzzy-Integral Hybrid Control of TIG Bead Width Parameter Self-Adjustment
Literature Overview
The paper by Huang Shisheng and He Jianfeng from the South China University of Technology, published in the journal "Control Theory and Applications" in 1995, addresses a fundamental challenge in GTAW (Gas Tungsten Arc Welding) overlay and cladding operations: the real-time automatic control of weld bead width. In cladding applications, maintaining a consistent bead width is critical because it directly affects the overlap ratio between successive passes, the dilution rate of the overlay material into the base metal, and ultimately the metallurgical integrity of the bonded interface. This work represents an early and pioneering application of hybrid control strategies combining fuzzy logic with integral control to achieve automatic parameter adjustment during TIG welding.
Core Technical Concept
The central idea is that bead width in TIG welding is a nonlinear, time-varying process variable influenced by multiple interacting parameters including welding current, travel speed, arc voltage, electrode stick-out, and shielding gas flow rate. Traditional PID control alone struggles with such nonlinear dynamics, while pure fuzzy control lacks the steady-state precision needed for tight dimensional tolerances. The proposed hybrid scheme uses a fuzzy controller to handle the nonlinear dynamic response during transients and an integral controller to eliminate steady-state error, thereby achieving both fast transient response and high accuracy.
The control objective is to maintain the weld bead width within a specified target range by adjusting the welding current in real time. The system architecture typically involves an optical or image-based sensor to measure bead width, a signal conditioning and processing unit, and a control algorithm that outputs a corrected welding current setpoint to the power source.
Control Strategy Analysis
The fuzzy logic component of the hybrid controller maps input variables such as error (difference between measured and target bead width) and the rate of change of error into control actions using a rule base. The fuzzy rules are designed based on expert knowledge of welding process behavior. For instance, if the bead width is significantly below the target and the error is increasing, the fuzzy controller commands a rapid increase in welding current. If the bead width is approaching the target, the fuzzy controller reduces the correction magnitude to prevent overshoot.
The integral component supplements the fuzzy controller by accumulating the error over time and providing a corrective signal that ensures zero steady-state error. This is particularly important in cladding operations where the bead width must remain constant over long weld lengths, and any drift in the control output would lead to unacceptable variation in overlay thickness.
| Control Parameter | Typical Range | Effect on Bead Width |
|---|---|---|
| Welding Current (I) | 80-300 A | Increasing current increases bead width |
| Travel Speed (v) | 200-800 mm/min | Increasing speed decreases bead width |
| Arc Voltage (V) | 12-22 V | Increasing voltage increases bead width |
| Electrode Stick-out | 8-15 mm | Increasing stick-out increases bead width |
| Shielding Gas Flow | 8-15 L/min | Primary effect on arc stability |
Engineering Relevance to Cladding Operations
In the context of weld overlay cladding for pressure vessels and heat exchangers, bead width consistency is directly linked to several critical quality requirements. According to NB/T 47014 and ASME IX, the qualification of overlay welding procedures requires demonstration of consistent mechanical properties and bond strength across the entire overlay layer. Variations in bead width lead to variations in dilution, which in turn affect the chemical composition of the overlay weld metal and can compromise corrosion resistance.
For example, in 304 stainless steel overlay on carbon steel, the dilution rate must typically be controlled below 30% to maintain adequate chromium content for corrosion resistance. If bead width varies by more than 10-15% across a single pass, the dilution rate at the edges of the bead may exceed acceptable limits, creating localized regions susceptible to intergranular corrosion or pitting.
The hybrid fuzzy-integral control approach offers a practical solution for automated cladding systems where manual operator skill cannot be consistently replicated. This is especially relevant for large-diameter pressure vessels and storage tanks where overlay welding may involve hundreds of passes over large surface areas.
Key Insights and Reflections
The 1995 publication of this work is noteworthy for its forward-looking application of intelligent control theory to a traditional manufacturing process. At that time, fuzzy control was still an emerging field, and its application to welding was largely theoretical. The practical implementation challenges, including sensor reliability, signal noise filtering, and the computational requirements for real-time fuzzy inference, were not fully addressed.
However, the conceptual framework remains highly relevant. Modern welding control systems increasingly incorporate adaptive and intelligent control strategies, and the hybrid approach of combining rule-based reasoning with precision feedback control is a pattern that continues to evolve in automated welding technology. For cladding engineers, the key takeaway is that bead width control is not merely a cosmetic concern but a fundamental metallurgical requirement that directly impacts the functional performance of the overlay layer.
The limitations of this early work include the absence of experimental validation on actual cladding applications, the lack of consideration for multi-layer multi-pass scenarios, and the simplification of the welding process model. Nevertheless, as a foundational contribution to intelligent welding control, this paper provides valuable theoretical grounding for engineers seeking to understand the principles behind automated parameter adjustment in GTAW-based overlay processes.