期刊名称：International Journal of Soft Computing & Engineering
电子版ISSN：2231-2307
出版年度：2012
卷号：2
期号：3
页码：114-120
出版社：International Journal of Soft Computing & Engineering
摘要：The fault detection and diagnosis (FDD) play an important role in the monitoring, repairing and maintaining of technical systems. In this paper, we presented a new method based on soft computing approach for FDD in a special type of HVAC system namely unitary system. In the proposed method, the feature vectors are extracted by applying wavelet transform to output signals of model. Then, a Takagi-Sugeno (T-S) fuzzy classifier detects and diagnoses the faults by use of extracted feature vectors, if the faults exist. The T-S fuzzy classifier needs to be trained. With inspiration from training formulation of support vector machine (SVM), the training process has been stated as an optimization problem. For solving the mentioned optimization problem, a reliable evolutionary algorithm namely differential evolution (DE) is used. One of the important types of faults in the unitary HVAC system is refrigerant leakage. FDD of refrigerant leakage is highlighted in the presented paper. The simulation has been done in MATLAB-Simulink and the efficacy of the proposed method is demonstrated based on the experimental results.
