Application of Principal Component Analysis combined with Random Forest in the Judgment of the Failure Degree of Chiller
碩士 === 國立臺北科技大學 === 能源與冷凍空調工程系 === 107 === With the global economy booming, various types of high-rise buildings are built up. Nowadays, the air conditioning system in each building is definitely one of the essential equipment. The air conditioning system provides our comfortable living environment...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2019
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/5z6xxb |
| Summary: | 碩士 === 國立臺北科技大學 === 能源與冷凍空調工程系 === 107 === With the global economy booming, various types of high-rise buildings are built up. Nowadays, the air conditioning system in each building is definitely one of the essential equipment. The air conditioning system provides our comfortable living environment and improves work efficiency. The air conditioning equipment will gradually become old as t air conditioners increases operation hour during operation. When the system or equipment is in an abnormal operating state, its energy consumption value will be higher than the normal value. Poorly maintained, degraded and poorly controlled equipment can cause about 15% to 30% of energy wastage in the building. Therefore, in this thesis Random Forest method and Principal Component Analysis combined with Random Forest method were used to analyze the degree of faults of seven kinds of Centrifugal Chiller common failures. The fault database for the application is the ASHRAE research project (1043-RP). The results show that the accuracy of the Random Forest method is about 30% higher than that of the Principal Component Analysis combined with the Random Forest method.
|
|---|