Novelty detection using auto-associative neural network.

[摘要] The primary objective of novelty detection is to examine if a system significantly deviates from the initial baseline condition of the system. In reality, the system is often subject to changing environmental and operation conditions affecting its dynamic characteristics. Such variations include changes in loading, boundary conditions, temperature, and humidity. Most damage diagnosis techniques, however, generally neglect the effects of these changing ambient conditions. Here, a novelty detection technique is developed explicitly taking into account these natural variations of the system in order to minimize false positive indications of true system changes. Auto-associative neural networks are employed to discriminate system changes of interest such as structural deterioration and damage from the natural variations of the system.

[发布机构] Technical Information Center Oak Ridge Tennessee

[学科分类] 工程和技术（综合）

[关键词] Diagnosis;Humidity;Neural networks;Boundary conditions;Detection