On-line Estimation of Workpiece Height by Using Neural Networks and Hierarchical Adaptive Control of WEDM

Abstract

Wire breakage and unstable machining drastically reduce the machining efficiency and accuracy in wire electrical discharge machining (WEDM). When a stair-shaped workpiece is machined, poor electrolyte flow around the steps leads to wire rupture or unstable machining. This paper presents a WEDM adaptive control system that maintains optimal machining and improves the stability of machining at the stair section where workpiece thickness changes. A three-layer back propagation neural network is used to estimate the thickness of a workpiece. The developed adaptive control system is executed in the hierarchical structure of three control loops, using fuzzy control strategy. In the first control loop, the total sparking frequency is controlled within a safe level for wire rupture suppression. In the second control loop, the proportion of abnormal sparks is maintained at a pre-determined level for process control purposes. Based on the estimated thickness of a workpiece, adaptive parameter optimisation is carried out to determine the optimal machining settings and to provide the reference targets for the other two control loops. Experimental results demonstrate that the workpiece height can be estimated by using a feed-forward neural network. The developed adaptive control system results in faster machining and better machining stability than does the commonly used gap voltage control system.

Application of genetic algorithm-based fuzzy logic control in wire transport system of wire-EDM machine

“…Fuzzy control theory can easily capture the approximate and qualitative aspects of human knowledge and can perform reasoning as well as simulate human intelligence for controlling a complicated process. Fuzzy control strategy has been successfully applied to metal cutting, EDM and wire-EDM processes where mathematical models are unknown or ill-defined (Hsu and Fann, 1996;Boccadoro and Dauw, 1995;Lin et al, 2001a,b;Yan et al, 2001). However, the design of fuzzy control has to rely on a prior knowledge of a human expert and thus the controller performance is mainly dependent on the quality of his expertise.…”

Spark Analysis Based on the CNN-GRU Model for WEDM Process

“…It was found that the ANN-LWPA integration system has some advantages on reducing the value of fitness functions by comparison with the experimental regression model, ANN model, and conventional LWPA result. Furthermore, Yan et al successively developed a servo control system based on fuzzy rule-based control strategy and adjusting strategy, as well a hierarchical adaptive control system based on the estimation of workpiece height on-line by using ANN to reduce the wire breakage and improve the machining stability and speed compared to the commonly used gap voltage control system [ 33 , 34 ].…”

Workpiece height estimation in wire electrical discharge machining by using support vector regression

“…2 A feedforward neural network (NN) was used to estimate the workpiece height and distinguish the machining conditions in WEDM. 5 An NN model was also derived to estimate the workpiece height, and a ruled-based strategy was proposed to maintain an optimal and stable machining. 6 The NN model for estimating workpiece heights discussed above is only applied to stair-shaped workpieces and does not have universality.…”