模拟电路故障诊断L1估计及其神经网络解法
A NEURAL-BASED NONLINEAR L1 OPTIMIZATION ALGORITHM FOR DIAGNOSIS OF NETWORKS
-
摘要: 基于精确罚函数法,提出了新的求解L1范致问题最优解的神经网络方法,它避免了Kennedy和Chua(1988)网络罚因子较大时性态变坏问题。对Bandler(1982)提出的模拟电路故障诊断L1范数法进行了改进,将线性约束L1问题转化为非线性约束L1问题,并用新的神经网络方法求解,计算量小。模拟实验表明,所提神经网络方法和改进的模拟电路故障诊断L1范数方法是可行的。
-
关键词:
- 故障诊断; L1范数; 神经优化
Abstract: Based on exact penalty function, a new neural-networks for solving the L1-norm optimization problem is proposed. In comparison with Kennedy and Chua s networks(1988), it has better properties. Based on Bandler s fault location method(1982), a new nonlinearly constrained L1 norm problem is developed. It can be solved with less computing time through only one optimization processing. The proposed neural networks can be used to solve the analog diagnosis L1 problem. The validity of the neural networks and the fault location L1 method are illustrated by extensive computer simulations. -
Bandler J W, et al. A linear programming approach to fault location in analog circuits. Proc. IEEE Int. Symp. CAS Chicago, IL: 1981, 256-261.[2]Bandler J W, et al. Fault isolation in linear analog circuits using the L1 norm. Proc. IEEE Int. Symp. CAS, Rome, Italy: 1982, 1140-1143.[3]Cichocki A, Unbehauen R. Neural networks for optimization and signal processing. John wiley, 1993, Chaps. 5, 7.[4]Pietrzykowski T. An exact potential method for constrained maxima. SIAM J. Num. Anal. 1969, 6(2): 299-304.[5]Kennedy M P, Chua L O. Neural networks for nonlinear programming. IEEE Trans. on CAS, 1988, 35(5): 554-562.[6]Charalambous C. On condition for optimality of the nonlinear L1 problem. SIAM J. Num. Anal. 1979,17(1): 123-135. -
计量
- 文章访问数: 2257
- HTML全文浏览量: 127
- PDF下载量: 371
- 被引次数: 0
下载:
