Monitoring of Joint Gap Formation in Laser Beam Butt Welding by Neural Network-Based Acoustic Emission Analysis (original) (raw)
Related papers
Journal of Intelligent Manufacturing, 2011
The need for the control of the depth of weld penetration has been and remains of a long term interest in the automated welding process. In this study, the relationship between the depth of weld penetration and the acoustic signal acquired during the laser welding process of high strength steels is investigated. The acoustic signals are first preprocessed by the spectral subtraction noise reduction method and analyzed both in the time domain and frequency domain. Based on this analysis, two algorithms are developed to acquire the acoustic signatures. The acquired acoustic signatures are then used to characterize the depth of weld penetration by using a neural network and a multiple regression analysis. The results show that the acoustic signatures can characterize and predict the depth of weld penetration well under different laser welding parameters.
Engineering Research Journal
Automatic inspection of welded gas pipelines is desirable because human inspectors are not always consistent evaluators. In addition, automatic inspection decreases the cost of inspection process and improves the inspection quality. In this paper, a proposed system named Phased Array for Characterizing Discontinuities "PACD" which combined 2D S-scan images into 3D images (volumetric scan) is more reliable and accurate and easily interpretation. The system is checking the shape, location, width and length of each discontinuity and the echo width and height of the A-scan. After that PACD inputs them to learned artificial neural network (ANN) to characterize 12-types of common welding discontinuities in gas pipelines welded by shield metal arc welding (SMAW). After verification the PACD system can be easily characterized welding discontinuities types with zero error.
Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 2009
It is a trend to use high-strength steels in the automobile industry because of their good formability, weldability, and high strength—volume ratio. In order to achieve quality control, it is necessary to monitor the welding process online. In this paper, acoustic signals generated during the laser welding process of high-strength steel DP980 were recorded and analysed. A microphone was used to acquire the acoustic signals. A spectral subtraction method was used to reduce the noise in the acoustic signals, and a Welch—Bartlett power spectrum density estimation method was used to analyse the frequency characteristics of the acoustic signals. The results indicate that good welds with full penetration (FP) could be clearly distinguished from bad welds with partial penetration (PP). An algorithm based on the different sound pressures between FP and PP was developed to identify the penetration state in the time domain. Another algorithm based on the different frequency characteristics fr...
Scientific Reports
Laser welding is a key technology for many industrial applications. However, its online quality monitoring is an open issue due to the highly complex nature of the process. this work aims at enriching existing approaches in this field. We propose a method for real-time detection of process instabilities that can lead to defects. Hard X-ray radiography is used for the ground truth observations of the sub-surface events that are critical for the quality. A deep artificial neural network is applied to reveal the unique signatures of those events in wavelet spectrograms from the laser back-reflection and acoustic emission signals. The autonomous classification of the revealed signatures is tested on reallife data, while the real-time performance is reached by means of parallel computing. The confidence of the quality classification ranges between 71% and 99%, with a temporal resolution down to 2 ms and a computation time per classification task as low as 2 ms. This approach is a new paradigm in the digitization of industrial processes and can be exploited to provide feedbacks in a closed-loop quality control system.
2011
It is a trend to use high-strength steels in the automobile industry because of their good formability, weldability, and high strength–volume ratio. In order to achieve quality control, it is necessary to monitor the welding process online. In this paper, acoustic signals generated during the laser welding process of high-strength steel DP980 were recorded and analysed. A microphone was used to acquire the acoustic signals. A spectral subtraction method was used to reduce the noise in the acoustic signals, and a Welch–Bartlett power spectrum density estimation method was used to analyse the frequency characteristics of the acoustic signals. The results indicate that good welds with full penetration (FP) could be clearly distinguished from bad welds with partial penetration (PP). An algorithm based on the different sound pressures between FP and PP was developed to identify the penetration state in the time domain. Another algorithm based on the different frequency characteristics fr...
Evaluate and control the weld quality, using acoustic data and artifical neural network modeling
Indian Journal of Scientific Research, 2014
The weld quality depends on many factors and parameters such as continuity of the weld, the weld penetration and the absence of defects in the weld. All these parameters have to be after the welding process (Off-line) examined. Since Welding sound signal is an important feedback, In this research it is used as a (On-line) Criterion to determine the weld quality. The purpose of this investigation is to evaluate and control the weld quality using acoustic parameters as input and Weld quality parameter as output in an artificial neural network. For this purpose, acoustic parameters welding process (The difference between the maximum and average sound intensity, The Average of Fast Fourier Transform-FFT coefficients and Standard deviation of FFT coefficients) as inputs and weld quality parameter (the percentage of weld quality) that is given by nondestructive testing and welding inspection, is considered as an output. The selection process for this study is The gas-shielded welding process (MIG), One of the most commonly used types of welding. Acoustic signals is recorded in the laboratory during the welding process. Acoustic parameters of the process is extracted by the signal processing. Weld quality parameter, also by Welding Inspection and Testing the quality of welded joints is determined. Finally, The relationship between acoustic parameters and weld quality parameter can be studied with the help of neural network modeling. After data analysis and prediction models, the results are presented.
Acoustic emission method for defect detection and identification in carbon steel welded joints
Journal of Constructional Steel Research, 2017
Detecting welding defects in industrial equipment (welded joints and built-up structures) is a key aspect in evaluating the probability of failure in different situations. Acoustic Emission (AE) is an effective non-destructive detecting technique, and can be a promising application for welding defect detection. This work presents a systematic experimental investigation on using AE technique for detecting and classifying different weld defects in carbon steel joint material. Four certified carbon steel samples were used in this study. A defect free control sample was used as the reference and three samples with induced defects, namely slag, porosity and crack. A pencil lead break (PLB) test was used to generate simulated AE sources on one side of the joint whereas the AE sensor was mounted on the other side to capture AE signals. A total of four experimental arrangements were used to investigate the effect of propagating distance (sensor to source distance) on the ability of AE to detect and identify defects in welds. For each of these arrangements, AE features such as peak amplitude, rise time, decay time, duration, and count numbers along with statistical features such as AE energy, root mean square (RMS) were extracted and analysed. Also, frequency analysis using FFT and wavelet transform were investigated for each weld test specimen for all arrangements. The results show that AE energy, peak amplitude and RMS 2 value can be used to automatically detect and identify the presence of a defect in carbon steel welds. It is concluded that AE has a considerable potential in use in welding inspection to assess the overall structural health and identify defects that can significantly reduce the strength and reliability of welded material and consequently reduce the risk of component's failure.
Materials and Design, 2020
• A quality inspection method for ultrasonic composite welding by combining AI technologies with process signals is proposed. • The failure load and weld quality level are predicted simultaneously by artificial neural network and random forest models. • The weld quality can be assessed more comprehensively by considering both the failure load and the weld quality level. Ultrasonic welding is a joining technology suitable for carbon-fiber-reinforced thermoplastic (CFRTP) components because of its high throughput, and ease of automation. An effective online weld-quality inspection technology can promote the industrial application of ultrasonic composite welding. Literature focused on the quality inspection of ultrasonic composite welding is scarce. To address this, the present study proposes an online weld-quality inspection method for ultrasonic composite welding by combining artificial intelligence (AI) technologies with welding process signatures. The failure load in a tensile-shear test and the weld quality level (i.e., under weld, normal weld, and over weld) are predicted simultaneously using artificial neural network (ANN) and random forest (RF) models. Eight features consisting of the duration and energy at each welding stage are extracted from the process signatures as independent variables. The results indicate that both the ANN and RF models exhibit high prediction accuracies. The weld quality can be assessed comprehensively and reasonably by considering both the failure load and weld quality level. The findings of this study demonstrate the feasibility of online weld-quality inspection for ultrasonic composite welding.
Science and Technology of Welding and Joining, 2009
The present paper describes the application of neural networks to obtain a model for estimating the stability of gas metal arc welding (GMAW) process. A neural network has been developed to obtain and model the relationships between the acoustic emission (AE) signal parameters and the stability of GMAW process. Statistical and temporal parameters of AE signals have been used as input of the neural networks; a multilayer feedforward neural network has been used, trained with back propagation method, and using Levenberg Marquardt's algorithm for different network architectures. Different welding conditions have been studied to analyse the incidence of the parameters of the process in acoustic signals. The AE signals have been processed by using the wavelet transform, and have been characterised statistically. Experimental results are provided to illustrate the proposed approach. Finally a statistical analysis for the validation of the experimental results obtained is presented. As a main result of the study, the effectiveness of the application of the artificial neural networks for modelling stability analysis in welding processes has been demonstrated. The regression analysis demonstrates the validity of neural networks to predict the stability of welding process using the statistical characterisation of the signal parameters of AE that have been calculated.
Real-time monitoring of welding process using air-coupled ultrasonics and acoustic emission
The International Journal of Advanced Manufacturing Technology, 2018
The automated weld quality assurance can improve efficiency and productivity. This paper presents the development of real-time weld quality assurance approach for gas tungsten arc welding (GTAW) using acoustic emission (AE) and air-coupled ultrasonic testing (UT). The major weld defect of interest in this paper is burn through, that is, melting through the base metal during welding that creates a hole/gap. The in situ monitoring system evaluates the changes in weld size leading to burn through by changing the weld heat input. Different categories of burn through are defined that include melting of the back of the plate without any molten metal exiting to formation of a hole in the plate. It is demonstrated that complete air-coupled UT cannot be used simultaneously with welding due to the influence of the magnetic field that develops in the weld torch during welding, which weakens the ultrasonic signal. Consequently, a rolling UT transmitter is combined with air-coupled UT receiver to increase the signal/noise value. Wave dispersion is detected due to the different levels of burn through. While UT method provides quantitative information about the weld state, any localized surface discontinuity causes sudden surges in the AE energy indicating nonuniform welding qualitatively. It is concluded that passive and active nondestructive evaluation methods should be combined to monitor weld quality real time for qualitative and quantitative assessment. Keywords Real-time. Gas tungsten arc welding (GTAW). Non-contact AE. Real-time ultrasonics. Burn through