【摘要】 矿用机械由于工作环境特殊，故障率远远高于其他机械。发动机是矿用机械的关键部件，对其进行超前管理和预知维修有助于延长设备的工作时间，提高工作效率和经济效益。采用Levy飞行策略改进的麻雀搜索算法(SSA)优化支持向量机(SVM)并建立预测模型，评价指标为平均相对误差和平方相关系数。通过与标准的SSA-SVM算法对比，仿真结果表明，改进的算法对发动机磨损状态的预测能力更优秀。更多还原

【Abstract】 Because of the special working environment, the failure rate of mining machinery is much higher than that of other machineries. The engine is the key part of miningmachinery, the advance management and predictive maintenance help to extend the working time of equipment, improve work efficiency and economic benefits. The sparrow search algorithm(SSA) modified by Levy flight tactics was used to optimize support vector machine(SVM) and built a prediction model. The evaluation indexes are mean relative error and square correlation coefficient. Compared with the normative SSA-SVM algorithm, the simulation results show that the improved algorithm has excellent ability to predict the wear state of the engine.更多还原