Fault diagnosis of chillers using central loss conditional generative adversarial network

doi:10.11949/0438-1157.20220568

Abstract

Abstract:

Aiming at the problem that the fault data generated by the chiller is insufficient and the quantity of normal data and fault data in the historical data set is unbalanced, which leads to the decline of the fault diagnosis accuracy. In this paper, a fault diagnosis method based on central loss conditional generative adversarial network (CLCGAN) and support vector machine (SVM) were proposed. Firstly, CLCGAN generates new fault data from a small amount of real fault data. Then, the generated fault data was mixed with the initial data set to balance the amount of normal data and fault data. Finally, the SVM model was constructed by using the balanced data set for fault diagnosis. When GAN generates chiller fault data, the dynamic center loss term was constructed and added into the objective function. Through the dynamic center loss term, the intra-class distance of various fault data generated by the chiller was reduced, so that the overlapping degree of each fault generated data was reduced and the reliability of generated data was increased. Before generating fault data, configure fault labels and input them into CLCGAN to guide the data generation process. In this way, the generated fault data can be evenly distributed among different fault types. The proposed method was validated on ASHRAE 1043-RP project data set, and the results show that the proposed method has higher fault diagnosis accuracy than other fault diagnosis methods that solve the problem of data imbalance.

Key words: chiller, fault diagnosis, generative adversarial networks, neural networks, algorithm, center loss, integrate

摘要：

针对冷水机组产生的故障数据不足，数据集中正常数据和故障数据数量不平衡，进而导致故障诊断精度下降的问题，提出一种基于中心损失的条件生成式对抗网络(central loss conditional generative adversarial network，CLCGAN)和支持向量机(support vector machine，SVM)的故障诊断方法。首先，CLCGAN利用少量真实故障数据生成新的故障数据；然后，将生成的故障数据与初始数据集混合，使正常数据与故障数据的数量达到平衡；最后，利用平衡数据集构建SVM模型进行故障诊断。在GAN生成冷水机组故障数据时，构建动态中心损失项并加入到目标函数中，利用动态的中心损失减少冷水机组生成的各种故障数据的类内距离，从而降低各个故障生成数据之间的重叠程度，增加生成数据的可靠性。在生成故障数据之前配置相应的故障标签，并输入到CLCGAN中指导数据生成过程，使生成的故障数据可以均衡地分布于各个故障类别。在ASHRAE 1043-RP数据集上对所提方法进行了验证，结果表明，相较于其他解决数据不平衡问题的故障诊断方法，所提方法具有更高的故障诊断准确率。

关键词: 冷水机组, 故障诊断, 生成式对抗网络, 神经网络, 算法, 中心损失, 集成

CLC Number:

TP 277

Xuejin GAO, Kun CHENG, Huayun HAN, Huihui Gao, Yongsheng QI. Fault diagnosis of chillers using central loss conditional generative adversarial network[J]. CIESC Journal, 2022, 73(9): 3950-3962.

高学金, 程琨, 韩华云, 高慧慧, 齐咏生. 基于中心损失的条件生成式对抗网络的冷水机组故障诊断[J]. 化工学报, 2022, 73(9): 3950-3962.

Figures/Tables 12

References 31

1	Afshari A, Friedrich L. A proposal to introduce tradable energy savings certificates in the emirate of Abu Dhabi[J]. Renewable and Sustainable Energy Reviews, 2016, 55: 1342-1351.
2	Katipamula S, Brambley M. Methods for fault detection, diagnostics, and prognostics for building systems—a review (Ⅰ)[J]. HVAC&R Research, 2005, 11(1): 3-25.
3	谢伟. 基于KPCA-LSSVM的冷水机组故障诊断研究[D]. 杭州: 杭州电子科技大学, 2019.
	Xie W. Research on fault diagnosis of chillers based on KPCA-LSSVM[D]. Hangzhou: Hangzhou Dianzi University, 2019.
4	余绍斌. 基于KECA-ELM的冷水机组故障检测与诊断研究[D]. 杭州: 杭州电子科技大学, 2019.
	Yu S B. Fault detection and diagnosis in chillers based on KECA-ELM[D]. Hangzhou: Hangzhou Dianzi University, 2019.
5	Yan K, Shen W, Mulumba T, et al. ARX model based fault detection and diagnosis for chillers using support vector machines[J]. Energy and Buildings, 2014, 81: 287-295.
6	Zhang S, Zhu X, Anduv B, et al. Fault detection and diagnosis for the screw chillers using multi-region XGBoost model[J]. Science and Technology for the Built Environment, 2021, 27(5): 608-623.
7	Xia Y D, Ding Q, Jiang A P, et al. Incipient fault diagnosis for centrifugal chillers using kernel entropy component analysis and voting based extreme learning machine[J]. Korean Journal of Chemical Engineering, 2022, 39(3): 504-514.
8	王占伟, 王林, 梁坤峰, 等. 基于融合的贝叶斯网络的冷水机组故障诊断[J]. 化工学报, 2018, 69(7): 3167-3173.
	Wang Z W, Wang L, Liang K F, et al. Chiller fault diagnosis based on fusional Bayesian network[J]. CIESC Journal, 2018, 69(7): 3167-3173.
9	Wang Y L, Wang Z W, He S W, et al. A practical chiller fault diagnosis method based on discrete Bayesian network[J]. International Journal of Refrigeration, 2019, 102: 159-167.
10	He S W, Wang Z W, Wang Z W, et al. Fault detection and diagnosis of chiller using Bayesian network classifier with probabilistic boundary[J]. Applied Thermal Engineering, 2016, 107: 37-47.
11	Han H, Xu L, Cui X Y, et al. Novel chiller fault diagnosis using deep neural network (DNN) with simulated annealing (SA)[J]. International Journal of Refrigeration, 2021, 121: 269-278.
12	刘旭婷, 李益国, 孙栓柱, 等. 基于稀疏局部嵌入深度卷积网络的冷水机组故障诊断方法[J]. 化工学报, 2018, 69(12): 5155-5163.
	Liu X T, Li Y G, Sun S Z, et al. Fault diagnosis of chillers using sparsely local embedding deep convolutional neural network[J]. CIESC Journal, 2018, 69(12): 5155-5163.
13	Ren Z X, Han H, Cui X Y, et al. Application of PSO-LSSVM and hybrid programming to fault diagnosis of refrigeration systems[J]. Science and Technology for the Built Environment, 2021, 27(5): 592-607.
14	Han H, Zhang Z, Cui X Y, et al. Ensemble learning with member optimization for fault diagnosis of a building energy system[J]. Energy and Buildings, 2020, 226: 110351.
15	齐咏生, 张海利, 王林, 等. 基于MSPCA-KECA的冷水机组故障监测及诊断[J]. 化工学报, 2017, 68(4): 1499-1508.
	Qi Y S, Zhang H L, Wang L, et al. Fault detection and diagnosis for chillers using MSPCA-KECA[J]. CIESC Journal, 2017, 68(4): 1499-1508.
16	Han H, Cui X Y, Fan Y Q, et al. Least squares support vector machine (LS-SVM)-based chiller fault diagnosis using fault indicative features[J]. Applied Thermal Engineering, 2019, 154: 540-547.
17	Zhao Y, Xiao F, Wang S W. An intelligent chiller fault detection and diagnosis methodology using Bayesian belief network[J]. Energy and Buildings, 2013, 57: 278-288.
18	Li C D, Shen C X, Zhang H Y, et al. A novel temporal convolutional network via enhancing feature extraction for the chiller fault diagnosis[J]. Journal of Building Engineering, 2021, 42: 103014.
19	Gao Y, Han H, Ren Z X, et al. Comprehensive study on sensitive parameters for chiller fault diagnosis[J]. Energy and Buildings, 2021, 251: 111318.
20	Lu H L, Cui X Y, Han H, et al. A feature importance ranking based fault diagnosis method for variable-speed screw chiller[J].Science and Technology for the Built Environment, 2022, 28(2): 137-151.
21	Zhong C W, Yan K, Dai Y T, et al. Energy efficiency solutions for buildings: automated fault diagnosis of air handling units using generative adversarial networks[J]. Energies, 2019, 12(3): 527.
22	范雨强, 崔晓钰, 韩华, 等. 不平衡数据技术在冷水机组故障诊断中的应用[J]. 工程热物理学报, 2019, 40(6): 1219-1228.
	Fan Y Q, Cui X Y, Han H, et al. Chiller fault diagnosis with the technology of imbalanced data[J]. Journal of Engineering Thermophysics, 2019, 40(6): 1219-1228.
23	Gao J Q, Han H, Ren Z X, et al. Fault diagnosis for building chillers based on data self-production and deep convolutional neural network[J]. Journal of Building Engineering, 2021, 34: 102043.
24	Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144.
25	Yan K, Chong A, Mo Y C. Generative adversarial network for fault detection diagnosis of chillers[J]. Building and Environment, 2020, 172: 106698.
26	Yan K, Ma L L, Dai Y T, et al. Cost-sensitive and sequential feature selection for chiller fault detection and diagnosis[J]. International Journal of Refrigeration, 2018, 86: 401-409.
27	Yan K, Su J Y, Huang J, et al. Chiller fault diagnosis based on VAE-enabled generative adversarial networks[J]. IEEE Transactions on Automation Science and Engineering, 2022, 19(1): 387-395.
28	Mirza M, Osindero S. Conditional generative adversarial nets[J/OL]. [2022-04-21]. .
29	Wen Y D, Zhang K P, Li Z F, et al. A discriminative feature learning approach for deep face recognition[C]//Computer Vision-ECCV 2016. Cham: Springer, 2016: 499-515.
30	Noshad M, Zeng Y, Hero A O. Scalable mutual information estimation using dependence graphs[C]// 2019 IEEE International Conference on Acoustics, Speech and Signal Processing. Brighton, UK: IEEE, 2019: 2962-2966.
31	Comstock M C, Braun J E. Development of analysis tools for the evaluation of fault detection and diagnostics in chillers[R]. West Lafa yette, IN: Purdue University, 1999.

组成部分	结构	节点数	激活函数
生成器	输入层	129	ReLU
	隐藏层1	257	ReLU
	隐藏层2	129	ReLU
	输出层	64	ReLU
判别器	输入层	65	ReLU
	隐藏层1	256	ReLU
	隐藏层2	64	ReLU
	隐藏层3	32	ReLU
	输出层1	7	ReLU
	输出层2	1	Sigmoid

组成部分	结构	节点数	激活函数
生成器	输入层	129	ReLU
	隐藏层1	257	ReLU
	隐藏层2	129	ReLU
	输出层	64	ReLU
判别器	输入层	65	ReLU
	隐藏层1	256	ReLU
	隐藏层2	64	ReLU
	隐藏层3	32	ReLU
	输出层1	7	ReLU
	输出层2	1	Sigmoid

故障类型	生成方法	正常状态-0	故障等级-1	故障等级-2	故障等级-3	故障等级-4
冷凝剂中有非冷凝物	按体积充氮	0	1.00%	1.70%	2.40%	5.70%
蒸发器水流量下降	按百分比减少水流量	13.6 kg/s	-10%	-20%	-30%	-40%
冷凝器水流量下降	按百分比减少水流量	17.0 kg/s	-10%	-20%	-30%	-40%
冷凝器结垢	在冷凝器中堵塞管道	164根管	12%	20%	30%	45%
润滑油过多	按百分比过量充油	10 kg	1.00%	1.70%	2.40%	5.70%
制冷剂过量	按百分比过量充注制冷剂	136 kg	10%	20%	30%	40%
制冷剂泄漏	按百分比排放制冷剂	136 kg	-10%	-20%	-30%	-40%

故障类型	生成方法	正常状态-0	故障等级-1	故障等级-2	故障等级-3	故障等级-4
冷凝剂中有非冷凝物	按体积充氮	0	1.00%	1.70%	2.40%	5.70%
蒸发器水流量下降	按百分比减少水流量	13.6 kg/s	-10%	-20%	-30%	-40%
冷凝器水流量下降	按百分比减少水流量	17.0 kg/s	-10%	-20%	-30%	-40%
冷凝器结垢	在冷凝器中堵塞管道	164根管	12%	20%	30%	45%
润滑油过多	按百分比过量充油	10 kg	1.00%	1.70%	2.40%	5.70%
制冷剂过量	按百分比过量充注制冷剂	136 kg	10%	20%	30%	40%
制冷剂泄漏	按百分比排放制冷剂	136 kg	-10%	-20%	-30%	-40%

	真实标签
预测标签		0	1
	0	N_TP	N_FP
	1	N_FN	N_TN