A novel point cloud-based method for electrode wear simulation in electro-discharge machining of complex geometries

Electro-discharge machining (EDM) is widely used in mold and aerospace manufacturing, but inevitable electrode wear can affect the geometric accuracy of machined parts. Existing wear prediction methods mainly focus on hole drilling and milling, with no studies addressing multi-axis EDM for complex electrode shapes. This paper proposed a point cloud-based method for simulating and predicting electrode wear. First, the mechanism and patterns of electrode wear for complex-shaped electrodes were investigated through discharge craters and continuous-pulse experiments. Then, the point cloud method was used to discretize the workpiece and electrode models, followed by kinematic planning. Discharge points were determined based on the minimum discharge distance, and material removal and electrode wear were simulated according to wear patterns. To improve simulation efficiency, a KD-Tree search algorithm was employed to accelerate the identification of discharge points, and a parallel discharge strategy was proposed for high-density point clouds. The results show that in two-dimensional simulations, electrode axial length and corner feature errors are under 0.1 mm, and the KD-Tree algorithm, with a point cloud density of 400 points/mm², improves search efficiency by 78 times compared to sequential searching. The 3D simulation of blisk flow channel machining demonstrates high consistency with the actual electrode wear morphology. With a point cloud density of 125 points/mm³, the parallel discharge strategy increases simulation efficiency by more than 225 times compared to single-point material removal. The proposed method can efficiently and accurately predict the electrode shape for various electrical parameters, arbitrary electrode shapes, and any motion path in EDM.