WAAM过程中基于机器学习预测工艺相关参数的可行性研究 - 使用316LSi填充材料

Aug 5, 2026

Upload by Quinn

Wire arc additive manufacturing (WAAM) geometry depends on the quality of deposited layers, while variations in process parameters and heat flow alter geometric precision versus intended dimensions. In situ geometry monitoring integrated for backward control is therefore critical. This study evaluates the influence of voltage, welding current, travel speed, and wire feed rate on bead geometry outputs—weld bead width (BW) and height (BH). Machine learning models (linear regression, random forest) predict bead dimensions, and k-nearest neighbors supports classification based on weld parameters. The resulting predictions are cross-validated against actual measurements, using a dataset generated from 316LSi filler experiments for training.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Abstract

1 Introduction

WAAM process advantages and energy sources

Role of 316L stainless steel and related findings

Challenges in surface morphology and geometric deviation

Monitoring and influence of heat input

Welding in the World [https://doi.org/10.1007/s40194-024-01855-w](https://doi.org/10.1007/s40194-024-01855-w) Feasibility study on machine learning methods for prediction of process‑related parameters during WAAM process using SS‑316L filler material Sharath P. Subadra1,2 · Eduard Mayer1,2 · Philipp Wachtel1,2 · Shahram Sheikhi1,2 Received: 15 August 2024 / Accepted: 15 October 2024 © The Author(s) 2024 Abstract The geometry of objects by means of wire arc additive manufacturing technology (WAAM) is a function of the quality of the deposited layers. The process parameters variation and heat flow affect the geometric precision of the parts, when compared to the actual dimensions. Therefore, in situ geometry monitoring which is integrated in such a way to enable a backward control model is essential in the WAAM process. In this article, an attempt is made to study the effect of four input variables, namely voltage (U), welding current (I), travel speed and wire feed rate on the output function in the form of two geometrical characteristics of a single weld bead. These output functions which are determinant of the weld quality are width of weld bead (BW) and height of weld bead (BH) . A machine learning approach is utilised to predict the bead dimensions based on the input parameters and to predict the parameters by assigning suitable scores. For predicting thebead dimensions, two models, namely linear regression and random forest, shall be utilised, whereas for the purpose of classification based on weld parameters, k-nearest neighbours model shall be employed. Through this work, a wide dataset of parameters in the form of input variable and output in the form bead dimensions are generated for 316LSi filler material which shall be used as a training data for a machine learning algorithm. Subsequently, the predicted parameters shall be cross-checked with actual parameters. Keywords Wire arc additive manufacturing · 316L steel · Parameters · Weld bead geometry · Machine learning · Predictions 1 Introduction Wire arc additive manufacturing (WAAM) has emerged asan important additive manufacturing (AM) technology to manufacture large-dimensional components [1] . This AM Recommended for publication by Commission XII-Arc Welding Processes and Production Systems. * Sharath P. Subadra[sharath.peethambaransubadra@haw-hamburg.de](sharath.peethambaransubadra@haw-hamburg.de) * Shahram Sheikhi [shahram.sheikhi@haw-hamburg.de](shahram.sheikhi@haw-hamburg.de) 1 Institute of Materials Science and Joining Technology, University of Applied Science Hamburg, Berliner Tor 13, 20099 Hamburg, Germany 2 Forschungs-Und Trasnferzentrum 3I, University of Applied Science Hamburg, Berliner Tor 13, 20099 Hamburg, Germany process has some advantages when compared to other processes including high deposition speed, high material usage efficiency, and inexpensive production and device investment costs [2] . Various arc-based welding processes may serve as the energy source for WAAM. These can be a conventional gas metal arc welding (GMAW) [3], cold metal transfer (CMT) which is a special type of GMAW [4], gas tungsten arc welding (GTAW) [5] and plasma arc welding (PAW) [6]. Due to the poor surface finish, WAAM is usually referred to as being a near-net-shape technique. Therefore, WAAM parts can be used in their as-built conditions as a pedestrian bridge [7] and excavator arm [8] . The application arena can be further widened by employing post-processing in the form of subtractive manufacturing technologies [9] . The choice of material is an important factor, where 316L stainless steel was seen to perform better due to its weldability and corrosion resistance. The material has a wide range of application in the nuclear sector, marine engineering and biomedical implants. Recent studies have focussed on analysing the metallurgical characteristics and mechanical properties of 316L thin-walled of thick-walled components from WAAM [10

FAQs

Why is in situ geometry monitoring important in WAAM?

WAAM part geometry is governed by the quality of deposited layers, and changes in process parameters and heat flow affect geometric precision. Monitoring enables backward control to maintain the intended dimensions.

Which input variables are studied for predicting weld bead geometry?

The study uses four inputs: voltage (U), welding current (I), travel speed, and wire feed rate.

Which machine learning models are used for the bead geometry and parameter-related tasks?

Linear regression and random forest are used to predict bead dimensions, while k-nearest neighbors is employed for classification based on weld parameters.

Aug 5, 2026

Upload by Quinn