MIG Weld Geometry Monitoring Using Analytic Hierarchy Process

Literature Overview

The paper by Ma Guohong, Hong Lei, Li Jian, Yu Lesheng, He Yinshui, and Xiao Wenbo, published in the Journal of Nanchang University (Engineering Science) in 2021, presents a method for monitoring MIG weld geometry using the Analytic Hierarchy Process (AHP). Supported by the National Natural Science Foundation of China (51665037), the Jiangxi Provincial Key Laboratory of Lightweight High-Strength Structural Materials (2017BCD40003), and the Ministry of Education Key Laboratory of Nondestructive Testing Technology at Nanchang Hangkong University (EW201980090), this research addresses an important challenge in welding quality control: the systematic evaluation and monitoring of weld geometry characteristics.

Core Methodology and Technical Framework

The Analytic Hierarchy Process is a multi-criteria decision-making method that decomposes complex problems into hierarchical structures of criteria, sub-criteria, and alternatives. The authors applied this methodology to weld geometry monitoring by establishing a hierarchical evaluation framework that considers multiple geometric parameters simultaneously.

The hierarchical structure developed in this research includes:

Hierarchy Level	Criteria	Sub-criteria	Weight
Goal	Weld Geometry Quality	—	1.0
Criteria	Penetration Quality	Root penetration, Back penetration	0.35
Criteria	Surface Geometry	Bead width, Bead height, Surface uniformity	0.30
Criteria	Profile Characteristics	Reinforcement height, Groove profile conformity	0.20
Criteria	Dimensional Accuracy	Size consistency, Position accuracy	0.15

Each sub-criterion is evaluated using a combination of visual inspection, dimensional measurement, and non-destructive testing data. The AHP framework enables the integration of these diverse data sources into a single comprehensive quality index that can be monitored in real-time during the welding process.

Implementation and Data Acquisition

The monitoring system described in the paper acquires weld geometry data through multiple sensing modalities:

Sensing Method	Parameters Measured	Frequency	Accuracy
Optical imaging	Bead width, height, surface profile	30 frames/sec	±0.1 mm
Laser displacement	Reinforcement height, groove profile	1 kHz	±0.01 mm
Arc voltage sensing	Arc length, weld pool width	10 kHz	±0.05 mm
Current sensing	Heat input, penetration estimate	10 kHz	±0.5 A

The multi-sensor approach provides comprehensive geometric information that feeds into the AHP evaluation framework. The system processes this data in real-time and generates a quality index that indicates whether the weld geometry meets specification requirements.

Results and Validation

The authors validated the AHP-based monitoring method through extensive experimental welding trials on carbon steel and stainless steel specimens. The results demonstrated that the method accurately predicts weld geometry quality with a correlation coefficient exceeding 0.92 when compared to post-weld dimensional measurements.

Key findings from the validation studies include:

Weld Parameter	AHP Prediction Error	Acceptable Range
Penetration depth	±0.3 mm	Within ±0.5 mm
Bead width	±0.2 mm	Within ±0.3 mm
Reinforcement height	±0.15 mm	Within ±0.2 mm
Surface uniformity	±0.1 mm	Within ±0.15 mm

The system successfully identified weld geometry deviations in over 95 percent of test cases, demonstrating its effectiveness as a real-time monitoring tool.

Engineering Practice Application

For welding engineers, this AHP-based monitoring method offers a systematic approach to weld geometry quality control. In cladding and weld overlay applications, where the geometry of the overlay layer directly affects the functional performance of the bimetallic component, such monitoring is particularly valuable. The method can be adapted to monitor overlay layer thickness, dilution rate indicators, and interface geometry in real-time.

The hierarchical evaluation framework can be customized for specific applications by adjusting the criteria weights and evaluation thresholds. For example, in hydrogenation reactor fabrication where overlay layer uniformity is critical, the weight assigned to surface uniformity would be increased relative to other criteria.

Integration with Quality Management Systems

The AHP-based monitoring method integrates naturally with existing quality management systems including PDCA (Plan-Do-Check-Act) cycles and FMEA (Failure Mode and Effects Analysis) frameworks. The real-time quality index generated by the system provides the "Check" element in PDCA, enabling immediate corrective action when weld geometry deviates from specifications. The hierarchical criteria structure also aligns with FMEA's systematic approach to identifying and prioritizing potential failure modes.

Study Insights and Reflections

This paper presents a sophisticated approach to weld geometry monitoring that leverages multi-criteria decision-making methodology. The authors' application of AHP to welding quality control demonstrates the value of structured analytical approaches in engineering quality management. For engineers in the cladding and bimetallic products field, this methodology provides a framework for systematically evaluating weld quality across multiple dimensions simultaneously. The real-time monitoring capability enabled by this approach represents a significant advancement over traditional post-weld inspection methods, enabling proactive quality control rather than reactive defect detection. This shift toward real-time, integrated quality monitoring is essential for achieving the high reliability requirements of modern pressure vessel and cladding applications.