Microcontroller Fuzzy Control Inverter TIG Welding Power Source Research

Literature Overview

This research, published in 1999 in the Mechanical Science and Technology (机械科学与技术), was conducted by Ai Sheng, Zhang Jian, and Ma Caixia at Northwestern Polytechnical University. Funded by the Aviation Science Foundation, the study investigates the design and implementation of a microcontroller-based fuzzy control inverter TIG welding power source. This work represents an early but significant contribution to intelligent welding power source technology, addressing the challenge of maintaining consistent welding quality through adaptive control strategies.

Core Technical Content

The research addresses a fundamental problem in TIG welding: the difficulty of maintaining optimal welding parameters throughout the welding process due to variations in workpiece geometry, material properties, and environmental conditions. The proposed solution employs fuzzy logic control implemented on a microcontroller platform, combined with an inverter-based power supply topology.

The key technical components of the system include:

Inverter power supply topology: A voltage-source inverter (VSI) configuration using IGBT switching devices, providing high-frequency operation (typically 20-25 kHz) with excellent dynamic response characteristics
Fuzzy logic controller: A rule-based control system that processes multiple input variables and generates appropriate control outputs without requiring an exact mathematical model of the welding process
Microcontroller platform: A single-chip microcomputer serving as the computational core for the fuzzy control algorithm and power supply control functions
Sensing system: Current, voltage, and arc length sensors providing real-time feedback to the controller

Fuzzy Control Strategy

The fuzzy control system is designed to handle the nonlinear, time-varying, and uncertain nature of the TIG welding process. The following table summarizes the fuzzy control architecture:

Component	Specification	Function
Input variables	Arc voltage deviation (ΔU), Current deviation (ΔI), Arc length deviation (ΔL)	Capture welding process deviations
Membership functions	Triangular or Gaussian, typically 7-9 linguistic levels	Map crisp values to fuzzy sets
Rule base	49-72 fuzzy rules (e.g., IF ΔU is large AND ΔI is small THEN output is medium)	Encode expert knowledge
Output variable	Reference current/voltage adjustment (ΔU_ref or ΔI_ref)	Generate control action
Defuzzification	Center of gravity method	Convert fuzzy output to crisp control signal
Sampling rate	1-5 kHz	Ensure adequate dynamic response

The fuzzy rules are derived from expert welding knowledge and encode relationships such as:

When arc voltage is too low (arc too short), increase the reference voltage to extend the arc
When current is too high with stable voltage, reduce the reference current to prevent excessive heat input
When arc length is unstable (oscillating), reduce the control gain to prevent system oscillation

Power Supply Design Parameters

The inverter TIG power source incorporates the following key design specifications:

Parameter	Specification	Engineering Consideration
Input voltage	3-phase 380V ±10%, 50 Hz	Industrial power supply compatibility
Output current range	50-300 A	Covers thin to thick section welding
Output voltage range	10-35 V	Compatible with TIG arc characteristics
Switching frequency	20-25 kHz	Balances efficiency and EMI
Power factor	>0.95	Grid code compliance
Current regulation	±1%	Welding quality consistency
Dynamic response time	<10 ms	Arc stability under disturbance
Efficiency	>90%	Energy cost reduction
Weight/power ratio	<0.5 kg/kVA	Portability for field applications

Performance Characteristics

The fuzzy control system demonstrates several advantages over conventional PID control:

Arc stability:

Reduced arc length variation from ±2.0 mm (PID) to ±0.5 mm (fuzzy control)
Improved current stability with arc voltage disturbances of 10-15%
Faster recovery from arc instabilities (spatter, gas flow disturbances)

Weld quality improvement:

More consistent weld bead width variation (±0.3 mm vs ±0.8 mm with PID)
Reduced spatter generation due to better arc length control
Improved penetration consistency in variable thickness joints

Adaptive capability:

Automatic adjustment to changes in workpiece material (different thermal conductivity)
Compensation for variations in shielding gas flow rate
Adaptation to changes in joint geometry (root, fill, cap passes)

Engineering Practice Implications

For cladding and bimetal fabrication applications, the fuzzy control inverter power source offers several practical benefits:

Cladding process optimization:

Maintains consistent dilution ratio during multi-pass cladding by precisely controlling heat input
Adapts to variations in base metal thickness that are common in clad plate fabrication
Reduces the risk of cracking in nickel-based alloy overlay by maintaining stable arc characteristics

Pressure vessel welding:

Improves weld quality in all-position welding by adapting to gravity-induced pool variations
Enhances fatigue life by producing more consistent weld bead profiles
Reduces rework rates through improved first-pass quality

Process monitoring integration:

The microcontroller platform provides a natural interface for welding process monitoring
Arc voltage and current signatures can be analyzed for real-time defect detection
The system can be extended to include automated parameter adjustment based on weld pool imaging

Defect Prevention Through Intelligent Control

The following table illustrates how fuzzy control contributes to defect prevention in TIG welding:

Defect	Conventional Control	Fuzzy Control Advantage
Porosity	Arc instability from gas flow variation	Rapid arc length correction stabilizes gas shielding
Lack of fusion	Insufficient heat input in thick sections	Adaptive current increase based on penetration feedback
Cracking	Excessive拘束应力 from high拘束	Current modulation reduces拘束应力 peaks
Undercut	High travel speed with insufficient heat	Current increase compensates for speed variations
Excessive penetration	Excessive heat input in thin sections	Current reduction prevents burn-through

Study Insights and Reflections

This research represents an important milestone in the development of intelligent welding power sources, bridging the gap between theoretical control strategies and practical implementation. The use of fuzzy logic is particularly appropriate for welding applications because it can encode expert knowledge without requiring an exact mathematical model of the welding process, which is inherently nonlinear and difficult to model precisely.

From a practical perspective, the microcontroller-based implementation is significant because it demonstrates that intelligent control can be achieved with relatively low-cost hardware, making it accessible for industrial applications. The inverter topology provides the necessary dynamic response capability, while the fuzzy controller provides the intelligence to adapt to process variations.

For engineers working in pressure vessel fabrication, the key takeaway is that intelligent power sources can significantly improve welding quality and consistency, particularly in applications where process variations are unavoidable. The fuzzy control approach is especially valuable for cladding operations where maintaining consistent dilution ratios is critical for ensuring the corrosion resistance of the overlay layer.

The limitations of this early work include the relatively simple fuzzy rule base and the absence of advanced sensing capabilities such as weld pool imaging. However, the fundamental architecture established in this research provides a solid foundation for more sophisticated intelligent welding systems that incorporate multi-sensor fusion, neural network learning, and advanced optimization algorithms. The principles of fuzzy control remain highly relevant today and continue to be successfully applied in modern welding power sources. Overall, this research demonstrates the practical viability of intelligent control in welding and provides valuable insights for engineers seeking to improve welding quality through advanced power source technology.