The effects of motion model, maching complex and numerical control system upon the crankpin roundness error in non-circular grinding were analyzed. According to the measured machining errors, the additional impulses as the displacement correction of grinding carriage were given to numerical control system for reducing the errors, which was an effective compensation strategy apt for non-circular grinding. However the strong nonlinearity of non-circular grinding system made it hard to describe the machining errors of crankpin precisely by a certain mathematic model. Considering both the error compensation experience and its development trend, a new compensation method based on fuzzy reasoning and self-learning for crankpin non-circular grinding was proposed. The experimental results of grinding show that the roundness error can be reduced into the expectation in only a few grinding circles by this method, which demonstrates its high efficiency and applicability.

crankshaft,  , non-circular grinding,  , error compensation,  , fuzzy reasoning

分析了曲轴随动磨削中，来自运动模型、工艺系统以及数控系统的误差对连杆颈圆度的影响。根据加工误差值，向数控系统提供附加脉冲修正量来调整砂轮架位移，是适应随动磨削特点的误差补偿策略。然而随动磨削系统具有很强的非线性，很难用数学模型来精确描述连杆颈加工误差的规律。因此结合以往补偿经验及其发展趋势，提出了具有自学习功能的曲轴随动磨削加工误差模糊推理补偿方法。磨削实验结果表明，采用该方法较好地解决了曲轴非圆磨削过程中的加工误差补偿，具有较高的补偿效率和适应性。

曲轴, 随动磨削, 误差补偿, 模糊推理