[1] Du ZhiminJin XinqiaoYang XuebinFan Bo,. Fault isolation of air handling unit based on neural network [J]. Journal of Southeast University (English Edition), 2010, 26 (2): 355-358. [doi:10.3969/j.issn.1003-7985.2010.02.047]
Copy
Fault isolation of air handling unit based on neural network(
)
基于神经网络的空调箱故障诊断
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 26
- Issue:
- 2010 2
- Page:
- 355-358
- Research Field:
- Energy and Power Engineering
- Publishing date:
- 2010-06-30
Info
- Title:
- Fault isolation of air handling unit based on neural network
- 基于神经网络的空调箱故障诊断
- Author(s):
- Du ZhiminJin XinqiaoYang XuebinFan Bo
- Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200240, China
- PACS:
- TU83
- DOI:
- 10.3969/j.issn.1003-7985.2010.02.047
- Abstract:
- Aiming at various faults in an air conditioning system, the fault characteristics are analyzed.The influence of the faults on the energy consumption and thermal comfort of the system are also discussed.The simulation results show that the measurement faults of the supply air temperature can lead to the increase in energy consumption.According to the fault characteristics, a data-driven method based on a neural network is presented to detect and diagnose the faults of air handling units.First, the historical data are selected to train the neural network so that it can recognize and predict the operation of the system.Then, the faults can be diagnosed by calculating the relative errors denoting the difference between the measuring values and the prediction outputs.Finally, the fault diagnosis strategy using the neural network is validated by using a simulator based on the TRNSYS platform.The results show that the neural network can diagnose different faults of the temperature, the flow rate and the pressure sensors in the air conditioning system.
- 针对空调系统中的不同故障, 分析了空调箱的故障特性, 并讨论了不同故障对空调系统能耗及热舒适性的影响.仿真试验结果表明, 送风温度的测量故障会导致系统能耗的增加.根据故障特性, 提出了一种基于神经网络的数据处理方法, 用以检测和诊断空调箱中的传感器故障.该方法首先选取历史数据对神经网络进行训练, 实现对系统运行状态的识别和预测.然后, 通过比较测量值与预测值, 计算出相对误差, 实现对故障的诊断.最后, 利用基于TRNSYS的仿真器, 对神经网络的故障诊断策略进行了验证.结果表明, 神经网络可以有效诊断空调系统中的温度、流量和压力传感器故障.
References:
[1] Usoro P B, Schick I C, Negahdaripour S.An innovation-based methodology for HVAC system fault detection [J].Journal of Dynamic Systems, Measurement and Control, 1985, 107(44): 284-289.
[2] Hyvarnen J, Karki S.Building optimization and fault diagnosis source book, IEA ANNEX 25[R].Paris: International Energy Agency, 1995.
[3] Dexter A L, Pakanen J.Demonstrating automated fault detection and diagnosis methods in real buildings[R].Espoo, Finland: International Energy Agency, 2001.
[4] Katipamula S, Brambley M R, Luskay L.Automated proactive techniques for commissioning air-handling units [J].Journal of Solar Energy Engineering, 2002, 125(3): 282-291
[5] Salsbury T I, Diamond R C.Fault detection in HVAC systems using model-based feedforward control [J].Energy and Buildings, 2001, 33(1): 403-415.
[6] Sreadharan P, Haves P.Comparison of chiller models for use in model-based fault detection[R], Austin, TX, USA:Texas A& M University, 2001.
[7] Norford L K, Wright J A, Buswell R A, et al.Demonstration of fault detection and diagnosis methods for air-handling units(ASHRAE 1020-RP)[J].International Journal of HVAC& R Research, 2002, 8(1): 41-71.
[8] Comstock M C, Braun J E.Development of analysis tools for the evaluation of fault detection and diagnostics in chillers [R].West Lafayette, IN, USA: Purdue University, 1999.
[9] Reddy T A, Andersen K K.An evaluation of classical steady-state off-line linear parameter estimation methods applied to chiller performance data [J].International Journal of HVAC& R Research, 2002, 8(1): 101-124.
[10] Jiang Y, Li J, Yang X.Fault direction space method for on-line fault detection [J].ASHRAE Transactions, 1995, 101(2): 219-228.
[11] Wang S W, Wang J B.Law-based sensor fault diagnosis and validation for building air-conditioning systems[J].International Journal of HVAC& R Research, 1999, 5(4): 353-380.
[12] Wang S W, Chen Y M.Fault-tolerant control for outdoor ventilation air flow rate in building based on neural network [J].Building and Environment, 2002, 37(7): 691-704.
[13] Wang S W, Xiao F.Sensor fault detection and diagnosis of air-handling units using condition-based adaptive statistical method [J].International Journal of HVAC& R Research, 2006, 12(1): 127-150.
[14] Chen Y M, Hao X L, Zhang G Q, et al.Flow meter fault isolation in building central chilling systems using wavelet analysis [J].Energy Conversion and Management, 2006, 47(13/14):1700-1710.
Memo
- Memo:
-
Biography: Du Zhimin(1977—), male, lecturer, duzhimin@sjtu.edu.cn.
Citation: Du Zhimin, Jin Xinqiao, Yang Xuebin, et al.Fault isolation of air handling unit based on neural network[J].Journal of Southeast University(English Edition), 2010, 26(2):355-358.