Research On Several Factors Of CNC Grinding Precision Of Camshaft

Camshafts are very important parts of automobile engine and other internalcombustion engines. Their machining quality and efficiency directly influences thequality of car products. Many plants make the camshaft utilizing modular andordinary grinding wheel in China. The products made by this way have not enoughcontour precision, efficiency and flexibility. To improve contour precision, efficiency and flexibility, CNC camshaft machinetool with CBN grinding wheel was proposed in foreign country. It ensures the highefficiency, high contour precision and high flexibility. This technology was widelyutilized in foreign country. Many company and research center study the newgrinding method for camshaft in China. But this technology is not mature inChina .It can not get enough contour precision though this immature technology. Toimprove the precision of camshaft, this thesis studies some factors that can influencethe precision of camshaft. This thesis put stress on the following aspects: (1)This thesis studies the motion mathematic model of machining camshaft.The characteristics of constant linear velocity grinding(CLVG) method and constantangle velocity grinding(CAVG) method are analyzed. This thesis analyses theproblem and influence to the machining precision of the two methods. To solute theproblem of CAVG and CLVG, this thesis proposed a new machining method. Thismethod can improve machining precision through optimizing the input commander.The optimization of the input commander can reduce the velocity, acceleration andjerk of servo motor without reducing efficiency. It can reduce the frequencycomponent that system is difficult to follow, and reduce fluctuate of velocity ofgrinding point. III(2)To the optimization presented before, this thesis proposed the dynamic antsystem(DACS) algorithm for this problem. Dynamic ant system algorithm is animproved ant system algorithm. This thesis improved ant system algorithm in threeaspects, and obtained better performance than conventional ant system algorithm. (3)The mathematical model for transmission of camshaft grinding machine hasbeen studied. Basing on this model, we do simulation and experiments. We analyzedthe roots that cause current problem. The mathematical model for servo controlsystem was studied. Optimization for PID of servo control system basing on particleswarm optimization(PSO) was proposed. This algorithm has better performance thangenetic algorithm. The evolution speed of PSO is six times than genetic algorithm atleast (4)Artificial neural network(ANN)is used to Error prediction model for machinesystem. But its performance will become worse when the training sets are small.To overcome ANN's shortcoming, we proposed a new method for Error predictionmodel. This method bases on the support vector machine. The experiment shows thatthis method can find internal regulation of the data, and has better performance thanANN. This thesis studies the compensation of amplitude and phase to the servocontrol system.