Fault Detection And Diagnosis Of Air-Conditioning System Using Convolutional Neural Network

Ramlan, Rohaizah (2019) Fault Detection And Diagnosis Of Air-Conditioning System Using Convolutional Neural Network. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (24 Pages) Fault Detection And Diagnosis Of Air-Conditioning System Using Convolutional Neural Network.pdf - Submitted Version Download (3MB)

Abstract

Fault diagnosis plays a vital role for large system in this modern industry. In general, faults are deviations from normal behavior of the system that indicates something is going wrong in the monitored system. Improving this system contributes great energy savings and avoid total system breakdown. One of the large systems that is commonly used in most of the building is heating, ventilation and air- conditioning (HY AC) system. Therefore, this project attempts to identify faulty as well as to monitor the HV AC system and diagnose fault within the chiller of the HV AC system. (The purpose of the proposed project is to ensure the satisfactory thermal comfort among users). This project uses convolutional neural network (CNN) for faults classification. The data is fed to CNN and undergoes training and testing processes by using spliting method and k-fold cross validation. The accuracy of CNN on air conditioning system fault data is tested and analysed, by varying the configurations. The results show that CNN is able to classify all faults with more than 97% accuracy

Actions (login required)

View Item