特徴抽出と専門家知識を利用したビル空調システム異常診断

抄録

In a large scale system like building air-conditioning system, measured time-series data is observed from many kinds of sensors. It is difficult to detect the fault by the administrators because only the limited experts can diagnose the unusual system. For this reasons, the new method is required, which can detect the fault from the measured data using a computer automatically.
This paper proposes how to extract feature pattern from measured time-series data using a learned neural network, and how to detect the fault with a knowledge expression model. The learned neural network can extract a feature pattern by calculating the weight of the neuron in output layer. The knowledge expression model can be constructed with the result of case analysis and the criteria. The result of the fault can be detected by the condition of the model, which is represented by the feature pattern. Through practical experiments, it is confirmed that the proposed method is effective to detect the fault of air conditioning system.