Abstract: To excavate enterprise experience knowledge deeply, and improve data driven conceptual design of coal mining machine.Based on the preliminary study, this paper optimized the ?-SVR reasoning model,increased design productivity, machine surface height and weight in the output parameters, reduce the error rate from 5% to 2.5%.and developed user-friendly reasoning model for shearer conceptual design by making full use of the mixed advantages of C# and MATLAB.C# runs the m-file without MATLAB environment by calling the dynamic link library using .NET interfaces. This gave full play to its data interaction advantages,improved the execution efficiency, and reduced the development difficulty with MATLAB powerful data computing capacity.The model can help designers get the technical parameters such as the cutting power by inputting mining height, cutting depth and other primitive parameters.The operation is simple and quick, and the result is accurate and reliable.