Fault Diagnosis of Power Plant Induced Draft Fan Based on PNN-WNN-DS Information Fusion

doi:10.12096/j.2096-4528.pgt.22162

Abstract

Abstract:

Aiming at the problems of complex operating conditions of induced draft fan, harsh working environment, and easy failure of power plant induced draft fan, a fault diagnosis method of the improved dempster-shafer evidence theory was proposed. In this method, the probabilistic neural network (PNN) and wavelet neural network (WNN) were used for preliminary diagnosis, and the evidence bodies were formed according to the output of PNN and WNN. Then the improved D-S fusion method was used for fusion diagnosis. The improved D-S method distributes conflict information according to the trust degree of the evidence and the focal element, so that the support rate of the focal element with high trust degree is strengthened, and the focal element with low trust degree is weakened, which makes the fusion diagnosis result more reasonable. The simulation results show that the proposed method can effectively diagnose the vibration fault of induced draft fan, avoid misdiagnosis, improve the accuracy of diagnosis, and reasonably distribute conflicting information.

Key words: power plant induced draft fan, focal element, fault diagnosis, improved D-S evidential theory

CLC Number:

TM 621

Hang ZHANG, Chuanjie ZHOU, Lin ZHANG, Jietao CHEN, Chunmei XU, Daogang PENG. Fault Diagnosis of Power Plant Induced Draft Fan Based on PNN-WNN-DS Information Fusion[J]. Power Generation Technology, 2022, 43(6): 951-958.

Figures/Tables 8

Fig. 1 Structure diagram of fusion diagnosis based on improved D-S

Tab. 1 Training sample data

序号	故障模式	特征1	特征2	特征3	特征4	特征5	特征6	特征7	特征8	目标输出
1	模式1	3.28	0.17	1.81	0	-0.19	2.35	1.86	1.02	0 0 1
2	模式1	3.28	0.19	1.81	0	0.32	2.17	1.87	1.03	0 0 1
3	模式1	3.27	0.18	1.81	0	0.26	2.38	1.87	1.03	0 0 1
…	…	…	…	…	…	…	…	…	…	…
22	模式1	3.38	0.13	1.84	0	-0.43	4.27	1.87	1.02	0 0 1
23	模式2	1.67	0.07	1.29	0	-0.4	2.01	1.31	1.02	0 1 0
24	模式2	1.69	0.1	1.3	0	-0.21	2.2	1.34	1.03	0 1 0
25	模式2	1.68	0.04	1.3	0	0.38	2.01	1.31	1.01	0 1 0
…	…	…	…	…	…	…	…	…	…	…
45	模式2	1.67	0.09	1.29	0	0.14	1.25	1.33	1.03	0 1 0
46	模式3	1.85	1.13	1.36	0.1	-0.58	2.8	1.68	1.24	1 0 0
47	模式3	2	0.43	1.42	0.01	0.51	3.23	1.57	1.11	1 0 0
48	模式3	1.97	1.11	1.4	0.07	1.31	4.24	1.9	1.35	1 0 0
…	…	…	…	…	…	…	…	…	…	…
90	模式3	4.55	1.48	2.13	0.1	-2.22	7.79	2.26	1.06	1 0 0

Tab. 2 Test sample data

序号	故障模式	特征1	特征2	特征3	特征4	特征5	特征6	特征7	特征8	目标输出
1	模式1	3.37	0.17	1.84	0	-0.04	2.5	1.88	1.03	0 0 1
2	模式1	3.54	0.14	1.88	0	-0.4	2.51	1.91	1.02	0 0 1
3	模式1	3.6	0.25	1.9	0.01	0.17	1.63	1.96	1.03	0 0 1
4	模式2	1.66	0.08	1.29	0	-0.43	2.71	1.32	1.02	0 1 0
5	模式2	1.67	0.05	1.29	0	0.78	2.22	1.32	1.02	0 1 0
6	模式3	3.45	1.93	1.86	0.29	1.08	3.17	2.51	1.35	1 0 0
7	模式3	4.2	2.29	2.05	0.2	-0.02	3.72	2.68	1.31	1 0 0
8	模式3	4.18	2.19	2.05	0.37	-1.31	3.57	2.38	1.16	1 0 0
9	模式3	3.61	2.68	1.9	0.76	0.16	1.59	2.75	1.45	1 0 0
10	模式3	4.55	0.93	2.13	0.04	-1.48	5.62	2.29	1.07	1 0 0

Tab. 3 Accuracy of PNN diagnosis under different spread values

s_pread	0.2	0.18	0.16	0.14	0.12	0.1	0.08
正确率/%	56	80	90	90	100	98	98

Tab. 4 Accuracy of WNN diagnosis under different iterations

迭代次数	100	150	200	300	500	1 000	2 000
正确率/%	92	98	98	98	98	94	96

Tab. 5 Diagnostic accuracy of different fusion methods

方法	式(1)	式(2)	式(3)	式(4)	式(11)
正确率/%	100	32	100	100	100

Tab. 6 Diagnostic results of test sample 1 under different methods

方法	m(A₁)	m(A₂)	m(A₃ )	$m (Θ)$	结果	结果评价
PNN	0.237 3	0.218 4	0.544 3	0	模式1	对
WNN	0.072 4	0.141 1	0.786 5	0	模式1	对
式(1)	0.036 1	0.064 7	0.899 2	0	模式1	对
式(2)	0.017 2	0.030 8	0.428 1	0.523 9	不定	错
式(3)	0.088 3	0.113 3	0.733 6	0.064 8	模式1	对
式(4)	0.098 3	0.125	0.776 7	0	模式1	对
式(11)	0.025 2	0.113 9	0.860 9	0	模式1	对

Tab. 6 Diagnostic results of test sample 1 under different methods

方法	m(A₁)	m(A₂)	m(A₃ )	$m (Θ)$	结果	结果评价
PNN	0.237 3	0.218 4	0.544 3	0	模式1	对
WNN	0.072 4	0.141 1	0.786 5	0	模式1	对
式(1)	0.036 1	0.064 7	0.899 2	0	模式1	对
式(2)	0.017 2	0.030 8	0.428 1	0.523 9	不定	错
式(3)	0.088 3	0.113 3	0.733 6	0.064 8	模式1	对
式(4)	0.098 3	0.125	0.776 7	0	模式1	对
式(11)	0.025 2	0.113 9	0.860 9	0	模式1	对

Tab. 7 Diagnostic results of test sample 43 under different methods

方法	m(A₁)	m(A₂)	m(A₃ )	$m (Θ$ )	结果	结果评价
PNN	0.262 9	0.303 2	0.433 9	0	Type1	对
WNN	0.169 3	0.465 4	0.365 3	0	Type2	错
式(1)	0.129 3	0.410 1	0.460 6	0	Type1	对
式(2)	0.044 5	0.141 1	0.158 5	0.655 9	不定	错
式(3)	0.146 7	0.322 9	0.347 5	0.182 8	Type1	对
式(4)	0.186 2	0.393 2	0.420 6	0	Type1	对
式(11)	0.081 4	0.414	0.504 7	0	Type1	对

Tab. 7 Diagnostic results of test sample 43 under different methods

方法	m(A₁)	m(A₂)	m(A₃ )	$m (Θ$ )	结果	结果评价
PNN	0.262 9	0.303 2	0.433 9	0	Type1	对
WNN	0.169 3	0.465 4	0.365 3	0	Type2	错
式(1)	0.129 3	0.410 1	0.460 6	0	Type1	对
式(2)	0.044 5	0.141 1	0.158 5	0.655 9	不定	错
式(3)	0.146 7	0.322 9	0.347 5	0.182 8	Type1	对
式(4)	0.186 2	0.393 2	0.420 6	0	Type1	对
式(11)	0.081 4	0.414	0.504 7	0	Type1	对

References 17

1	富双进．电站风机故障预警系统的研究[D]．保定：华北电力大学，2015．
	FENG S J ．Research of power plant fan failure prognostic system[D]．Baoding：North China Electric Power University， 2015．
2	徐卫仙，王杰，乔立捷．发电厂锅炉引风机叶柄断裂原因分析[J]．发电与空调，2015，36(6)：45-47． doi:10.3969/J.ISSN.2095-3429.2015.06.013
	XU W X， WANG J， QIAO L J ．Cause analysis on fracture on blower fan petiole of power plant boiler[J]．Power Generation and Air Conditioning，2015，36(6)：45-47． doi:10.3969/J.ISSN.2095-3429.2015.06.013
3	危威．大型火电厂风机质量不平衡故障分析及处理[J]．发电技术，2020，41(2)：198-201． doi:10.12096/j.2096-4528.pgt.19111
	WEI W ．Analysis and treatment of mass unbalance fault of fan in large thermal power plant[J]．Power Generation Technology，2020，41(2)：198-201． doi:10.12096/j.2096-4528.pgt.19111
4	张晓斌，王旭辉．1 000 MW机组轴流式引风机振动故障诊断及治理[J]．锅炉制造，2021(6)：57-59．
	ZHANG X B， WANG X H ．Vibration diagnosis and treatment of axial flow fan for 1 000 MW generation unit[J]．Boiler Manufacturing，2021(6)：57-59．
5	唐贵基，孙军科，王晓龙，等．基于改进SSD-Teager时频分析的引风机转子故障诊断方法[J]．电力自动化设备，2022，42(3)：121-127．
	TANG G J， SUN J K， WANG X L， et al ．Diagnosis method for rotor fault of induced draft fan based on improved SSD-Teager timer-frequency anslysis[J]．Electric Power Automation Equipment，2022，42(3)：121-127．
6	周琦．基于机器学习的电站风机故障诊断研究[D]．保定：华北电力大学，2019．
	ZHOU Q ．Study on machine learning fault diagnosis[D]．Baoding：North China Electric Power University，2019．
7	PENG T， JIAN L ．Research on fault diagnosis method of axial flow induced draft fan of power plant based on machine learning[C]//2019 4th International Conference on System Reliability and Safety，Rome，Italy．IEEE，2019：325-330．
8	JAGTAP H P， BEWOOR A K， KUMAR R ．Failure analysis of induced draft fan used in a thermal power plant using coordinated condition monitoring approach：a case study[J]．Engineering Failure Analysis，2020：1-18．
9	韩崇昭，朱洪艳，段战胜，等．多源信息融合[M]．北京：清华大学出版社，2006．
	HAN C Z， ZHU H Y， DUAN Z S，et al ．Multisource information fusion[D].Beijing：Tsinghua University Press，2006.
10	YAGER R R ．On the dempster-shafer framework and new combination rules[J]．Information Sciences (S1007-7634)，1987，41(2)：93-137．
11	孙全，叶秀清，顾伟康．一种新的基于证据理论的合成公式[J]．电子学报，2000，28(8)： 117-119． doi:10.3321/j.issn:0372-2112.2000.08.036
	SUN Q， YE X Q， GU W K ．A new combination rules of evidence theory[J]．ACTA Electronica Sinica， 2000，28(8)：117-119． doi:10.3321/j.issn:0372-2112.2000.08.036
12	李弼程，王波，魏俊，等．一种有效的证据理论合成公式[J]．数据采集与处理，2002，17(1)：33-36． doi:10.3969/j.issn.1004-9037.2002.01.008
	LI B C， WANG B， WEI J，et al ．An efficient combination rule of evidence theory[J]．Journal of Data Acquisition & Processing，2002，17(1)：33-36． doi:10.3969/j.issn.1004-9037.2002.01.008
13	曹洁，郭雷雷．一种基于局部冲突分配的证据组合规则[J]．计算机应用研究，2013，30(7)：2033-2035． doi:10.3969/j.issn.1001-3695.2013.07.029
	CAO J， GUO L L ．Evidence combination rule based on local conflict distribution strategy[J]．Application Research of Computers，2013，30(7)：2033-2035． doi:10.3969/j.issn.1001-3695.2013.07.029
14	夏飞，马茜，张浩，等．改进D-S证据理论在电动汽车锂电池故障诊断中的应用[J]．智能系统学报，2017，12(4)： 526-537． doi:10.11992/tis.201605001
	XIA F， MA X， ZHANG H，et al ．Application of improved D-S evidence theory in fault diagnosis of lithium batteries in electric vehicles[J]．CAAI Transactions on Intelligent Systems，2017，12(4)： 526-537． doi:10.11992/tis.201605001
15	邓勇，施文康，朱振福．一种有效处理冲突证据的组合方法[J]．红外与毫米波学报，2004，23(1)：27-32． doi:10.3321/j.issn:1001-9014.2004.01.006
	DENG Y， SHI W K， ZHU Z F ．Efficient combination approach of conflict evidence[J]．Journal of Infrared Millimeter Waves，2004，23(1)：27-32． doi:10.3321/j.issn:1001-9014.2004.01.006
16	李艺林，张林，龙捷，等．小波包与PNN在抖动控制电路故障诊断中的应用[J]．导航与控制，2018，17(6)：79-85．
	LI Y L， ZHANG L， LONG J，et al ．Application of wavelet packet and PNN in the fault diagnosis of dither control circuit[J]．Navigation and Control，2018，17(6)：79-85．
17	梁红梅，肖志怀．基于SA-WNN模型的水电机组故障诊断研究法[J]．计算机测量与控制，2016，24(8)：24-27． doi:10.16526/j.cnki.11-4762/tp.2016.08.007
	LIANG H M， XIAO Z H ．Study for vibration fault diagnosis of hydro-turbine generating unit based on SA-WNN[J]．Computer Measurement & Control，2016，24(8)： 24-27． doi:10.16526/j.cnki.11-4762/tp.2016.08.007

[1]	Mingliang BAI, Dongxue ZHANG, Jinfu LIU, Jiao LIU, Daren YU. Anomaly Detection of Gas Turbine Hot Components Based on Deep Autoencoder and Support Vector Data Description [J]. Power Generation Technology, 2021, 42(4): 422-430.
[2]	Shangnian CHEN, Luping LI, Shihai ZHANG, Minnan OUYANG, Ang FAN, Xiankui WEN. Research Progress of Vibration Fault Diagnosis Technology for Steam Turbine Generator Sets [J]. Power Generation Technology, 2021, 42(4): 489-499.
[3]	Yaguang LI, Meng LI. Anomaly Detection of Wind Turbines Based on Deep Small-World Neural Network [J]. Power Generation Technology, 2021, 42(3): 313-321.
[4]	Changxing YANG, Luping LI. Research Progress on Blade Damage Fault Detection Technology of Wind Turbine [J]. Power Generation Technology, 2020, 41(6): 599-607.
[5]	Wei WEI. Analysis and Treatment of Mass Unbalance Fault of Fan in Large Thermal Power Plant [J]. Power Generation Technology, 2020, 41(2): 198-201.
[6]	Yun LUO,Xuelin CHEN,Ruidong LI,Yongjian SU,Yiwei XU,Junkai CHAO,Pengzhu LI,Haibin REN. Prediction Model and Application of Turbine Regulating Stage Pressure Under Variable Conditions [J]. Power Generation Technology, 2019, 40(2): 161-167.
[7]	Guogang YUAN,Zhushi RAO. Review of Non-electrical Quantity Monitoring and Fault Diagnosis for Power Transformer on Vibration Analysis Method [J]. Power Generation Technology, 2019, 40(2): 134-140.
[8]	Jianguo WANG,Yutong LIN,Ye TIAN,Peng DU,Peiyan ZHANG,Hongwei XIN,Yingjie WU. Fault Diagnosis of Rolling Bearing of Wind Turbine Based on VMD and Different Envelope Order Structure [J]. Power Generation Technology, 2018, 39(1): 63-69.
[9]	XING Tao, LI Zhijun, CAO Lingyan, DING Lihua. Transformer Fault Diagnosis Based on Adaptive Immune Algorithm [J]. Power Generation Technology, 2017, 38(6): 42-45.
[10]	KONG De-tong, JIA Si-yuan, WANG Tian-pin, LIU Qing-chao. Wind Turbines Fault Diagnosis Method Researched by Vibration Analysis [J]. Power Generation Technology, 2017, 38(1): 54-58.