Partial Fault Detection of Air-conditioning System by Neural Network Algorithm using Data Preprocessing Method

데이터 전처리기법을 적용한 신경망 알고리즘의 냉방기 부분고장 검출

  • 한도영 (국민대학교 기계ㆍ자동차공학부) ;
  • 이한홍 (국민대학교 기계공학과 대학) ;
  • 윤태훈 (국민대학교 기계공학과 대학원)
  • Published : 2002.07.01

Abstract

The fault detection and diagnosis technology may be applied in order to decrease the energy consumption and the maintenance cost of the air-conditioning system. In this study, two different types of faults in the air-conditioning system, such as the condenser fouling and the evaporator fan slowdown, were considered. The neural network algorithm combined with data preprocessor was developed and applied to detect the faults of the real system. Test results show that this method is very effective to detect the faults in the air-conditioning system. Therefore, this developed method can be used for the development of the air-conditioner fault detection system.

Keywords

References

  1. Breuker, M. S. and Braun, J. E., 1998, Common faults and their impacts for rooftop air conditioners, HVAC&R Research, Vol. 4, No. 3, pp. 303-318
  2. Chen, B. and Braun, J. E., 2001, Simple rule-based methods for fault detection and diagnostics applied to packaged air conditioners, ASHRAE Transactions 2001, V.107, Pt.1
  3. McIntosh, I. B. D., Mitchell, J. W. and Beckman, W. A., 2000, Fault detection and diagnosis in chillers, ASHRAE Transactions 2000, V. 106, Pt. 2
  4. Frank, D. and Pletta, J. B., 1992, Neural network sensor fusion for security application, Intelligent Engineering Systems through Artificial Neural Networks, Vol. 2, pp. 745-750
  5. Ch'ng, C. G. and Yak, A. S., 1998, Neural networks for process diagnosis, ICARCV, pp. 494-498
  6. Han, D. and Lee, H., 1999, The development of multi heat pump, Ministry of Commerce, Industry and Energy
  7. Han, D. and Yoon, T., 2000, Partial fault response of multi-type air conditioner, Proceedings of the SAREK, pp. 319-323