放电成型加工中的单晶硅材料去除率建模分析与实验研究

MODELING ANALYSIS AND EXPERIMENTAL RESEARCH OF THE MATERIAL REMOVAL RATE OF THE MONOCRYSTALLINE SILICON IN THE COURSE OF THE ELETRICAL DISCHARGE MACHINING

在半导体材料放电加工可行性的基础上,分析了影响材料去除率的几个主要因素,其中包括空载电压、峰值电流、脉冲宽度以及脉冲间隔。采用中心组合设计实验,考察峰值电流、脉冲宽度、脉冲间隔这3个因素对单晶Si放电加工的材料去除率的影响,建立了单晶Si放电加工的材料去除率的响应模型,进行响应面分析。方差分析结果表明模型具有很好的拟合程度和适应性。采用满意度函数(DFA)确定了单晶Si放电加工的最佳工艺参数,当峰值电流取18.5A、脉冲宽度取358.62μs、脉冲间隔取20μs时,满意度为0.912,此时材料去除率的最优值为76.26 mm^3/min。用所确定的最佳工艺参数在电火花成型机床上重复多次实验,测得P型单晶硅的平均MRR为73.86 mm^3/min。模型预测结果与最佳工艺参数下的实验结果平均相对误差为3.2%,验证实验表明该模型能实现相应的半导体材料放电加工过程的材料去除率预测。

Based on the feasibility of the discharge machining of the semiconductor material, it analyze several major factors which influences the material removal rate, including the open circuit voltage, peak current, pulse width and pulse interval. Using central composite design experiments, it inspects the influence of the three factors peak current, pulse width and pulse interval to the material removal rate of the monocrystalline silicon in the course of the electrical discharge machining and establishes a response model of the material removal rate of the monocrystalline silicon in the course of the electrical discharge machining and makes analysis to the response surface. The result of the variance analysis indicate that the model has good fitting degree and adaptability. It uses satisfactory rate function （DFA） to determine the optimum technology parameters of the monocrystalline silicon in the course of the electrical discharge When the value of the peak current is 18.5 A, the value of the pulse width is 358. 62μs, the value of the pulse interval is 20μs and the degree of satisfaction is 0. 912, then the optimal value of the material removal rate is 76.26 mm3/min It has measured the average MRR of the P type monocrystalline silicon is 73.86 mm3/min through repeated experiments many times on the electrical discharge shaping machine with the optimal processing parameters which are identified The average relative error of the model prediction results compare with experimental results under the optimum technological parameters is 3.4%. The verification test indicates that this model can achieve the prediction of the material removal rate of the corresponding, semiconductor material in the course of the discharge machining.