HVAC Fault Diagnosis System Using Rough Set Theory and Support Vector Machine
ABSTRACT Preventive maintenance plays a very important role in the modern Heating, Ventilation and Air Conditioning (HVAC) systems for guaranteeing the thermal comfort, energy saving and reliability. The fault diagnosis on HVAC system is a difficult problem due to the complex structure of the HVAC and the presence of multi-excite sources. As the HVAC system fault information has inaccurate and uncertainty characteristic, A new kind of fault diagnosis system based on Rough Set Theory (RST) and Support Vector Machine (SVM) is presented in this paper. The hybrid model is integrated the advantages of RST effectively dealing with the uncertainty information and SVMpsilas greater generalization performance. The HVAC diagnosis experiment demonstrated that the solution can reduce the cost and raise the efficiency of diagnosis, and verified the feasibility of engineering application. As a result, the presented hybrid fault diagnosis method can help to maintain the health of the HVAC systems, reduce energy consumption and maintenance cost.
[Show abstract] [Hide abstract]
ABSTRACT: This paper presents a novel strategy for predicting the performance of open vertical refrigerated display cabinets which is based on a modified two-fluid (MTF) model and an adaptive support vector machine (ASVM) algorithm. A MTF model (physical model) was built for open vertical refrigerated display cabinets, and then an ASVM algorithm (machine learning algorithm) was built. To verify the quantity of air leakage from the cabinet during operation, an important performance factor of display cabinets, an MTF model was built. After the training and validation data sets were constructed from the output of the MTF model, the problem was solved using an ASVM algorithm. The defrosting water quantity and total energy consumption / total display area (TEC/TDA), achieved from the experiments by using the predicted combination of the controlled parameters, were found to be reduced by 39.2% and 19.3%, respectively, from the experimental results of the original display cabinet.International Journal of Refrigeration 11/2010; 33(7-33):1413-1424. DOI:10.1016/j.ijrefrig.2010.04.006 · 1.70 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: In order to predict blended coal's property accurately, a new kind of hybrid prediction model based on principal component analysis (PCA) and weighted support vector machine (WSVM) was established. PCA was used to transform the high-dimensional and correlative influencing factors data to low-dimensional principal component subspace. These new features are then used as the inputs of WSVM to solve the load forecasting problem. The theoretical analysis and the simulation results show that PCA can efficiently extract the nonlinear feature of initial data. PCA-WSVM has powerful learning ability, good generalization ability and low dependency on sample data compared single SVR and PCA-SVM. It also indicates that the integration of PCA and WSVM forecast cooling load effectively and can be used in building cooling load prediction.