Sensor Optimization for Cutting Status Monitoring in Single Manufacturing Unit
Cutting status monitoring;Adaptive evolutionary algorithm(AEA);Sensor distribution The effectiveness of status monitoring in cutting process is contingent on the effectiveness of the sensor location optimization. A novel adaptive evolutionary algorithm (AEA) based on genetic algorithm (GA) to achieve an optimal sensor configuration for monitoring the cutting status closely related to the workpiece quality is proposed. Based on an understanding of the cutting principle, the minimal failure probability at minimum cost is the optimal target and the detectability and the false alarm rate are defined as the constraint conditions. The feasible region of solution is presented in decimal coding . The crossover operator and population size are adapted relative to the iterations estimated by a Markov model,which improve the convergence. The results of a case study have been found to agree well with the optimized results by AEA .The optimal scheme of AEA yields the minimal sensing stations while simultaneously attaining 99.99％ system detectability and renders nearly 60％ decrease in sensor cost in comparison to the scheme of saturated sensing.
Kang He Minping Jia Zhuanzhe Zhao Rong Wang
School of Mechanical Engineering, Southeast University, Nanjing 211189, PR China
国际会议
厦门
英文
636-639
2012-06-05（万方平台首次上网日期，不代表论文的发表时间）