Sequential Prediction Method of Single Parameter for Thermal System Based on MWSA

doi:10.16183/j.cnki.jsjtu.2021.300

Abstract

Abstract:

The changes in the status parameters of the thermal system reflect the operating status of the system in real time. The forecast results of the trend extraction and time series prediction of the current equipment status parameters can be used as a reference for the next operation management strategy and equipment maintenance, which can be used for the long-term system safe and stable operation. In this paper, a method which is described as MWSA, based on the midpoint and regression based empirical mode decomposition (MREMD), the wavelet threshold denoising (WTD) and midpoint and techniques, the singular value decomposition (SVD) and optimized parameter permutation entropy (PE), and an auto regressive integrated moving average model (ARIMA), is applied to the single-parameter time series prediction of thermal systems. First, the MREMD is used to decompose the monitored operating state parameters into a number of intrinsic mode functions (IMF) and residual components. Next, the components that do not meet the screening conditions are subjected to wavelet thresholding. The denoised components and the components that originally meet the filtering conditions are recomposed into new IMF components. Finally, the K-means clustering algorithm based on SVD and PE is used to classify the recomposed IMF components, the component with a lower entropy value is selected and reconstructed into a trend item, and ARIMA is used to predict. An actual case verifies that this method can effectively overcome the interference of high-frequency noise in the original parameter timing, and the prediction accuracy is higher than that of similar methods without noise reduction treatment.

Key words: sequential prediction, mode decomposition, threshold denoising, clustering algorithm, auto regressive model

CLC Number:

N37
TK 221

XIAO Pengfei, NI He, JIN Jiashan. Sequential Prediction Method of Single Parameter for Thermal System Based on MWSA[J]. Journal of Shanghai Jiao Tong University, 2023, 57(1): 36-44.

Figures/Tables 11

Fig.1

Fig.2

Tab.1

Tab.2

Fig.3

Fig.4

Fig.5

Tab.3

Fig.6

Fig.7

Fig.8

References 23

[1]	倪何, 覃海波, 郑奕杨. 考虑给水泄漏的锅炉升负荷仿真及其可靠性[J]. 上海交通大学学报, 2021, 55(4): 444-454.
	NI He, QIN Haibo, ZHENG Yiyang. Simulation and performance reliability of boiler load raising process considering leakage of feed water[J]. Journal of Shanghai Jiao Tong University, 2021, 55(4): 444-454.
[2]	秦文学, 王嘉兴, 王继强, 等. 基于ARMA模型的燃煤机组主蒸汽压力控制策略[J]. 热力发电, 2020, 49(9): 127-132.
	QIN Wenxue, WANG Jiaxing, WANG Jiqiang, et al. Control strategy of main steam pressure of coal-fired unit based on ARMA model[J]. Thermal Power Generation, 2020, 49(9): 127-132.
[3]	朱少民, 夏虹, 吕新知, 等. 基于ARIMA和LSTM组合模型的核电厂主泵状态预测[J]. 核动力工程, 2022, 43(2): 246-253.
	ZHU Shaomin, XIA Hong, LYU Xinzhi, et al. Condition prediction of reactor coolant pump in nuclear power plants based on the combination of ARIMA and LSTM[J]. Nuclear Power Engineering, 2022, 43(2): 246-253.
[4]	RUIZ-AGUILAR J J, TURIAS I, GONZÁLEZ-ENRIQUE J, et al. A permutation entropy-based EMD-ANN forecasting ensemble approach for wind speed prediction[J]. Neural Computing and Applications, 2021, 33(7): 2369-2391. doi: 10.1007/s00521-020-05141-w URL
[5]	陈亮, 刘宏立, 郑倩, 等. 基于EEMD-SVD-PE的轨道波磨趋势项提取[J]. 哈尔滨工业大学学报, 2019, 51(5): 171-177.
	CHEN Liang, LIU Hongli, ZHENG Qian, et al. An EEMD-SVD-PE approach to extract the trend of track irregularity[J]. Journal of Harbin Institute of Technology, 2019, 51(5): 171-177.
[6]	YANG Z J, LING B W K, BINGHAM C. Joint empirical mode decomposition and sparse binary programming for underlying trend extraction[J]. IEEE Transactions on Instrumentation and Measurement, 2013, 62(10): 2673-2682. doi: 10.1109/TIM.2013.2265451 URL
[7]	梁兵, 汪同庆. 基于HHT的振动信号趋势项提取方法[J]. 电子测量技术, 2013, 36(2): 119-122.
	LIANG Bing, WANG Tongqing. Method of vibration signal trend extraction based on HHT[J]. Electronic Measurement Technology, 2013, 36(2): 119-122.
[8]	刘海江, 刘世高, 李敏. 换挡加速度信号的EMD和小波阈值降噪方法[J]. 噪声与振动控制, 2018, 38(2): 198-203.
	LIU Haijiang, LIU Shigao, LI Min. EMD and wavelet threshold denoising method of gear-shift acceleration signals[J]. Noise and Vibration Control, 2018, 38(2): 198-203.
[9]	黄礼敏. 海浪中非平稳非线性舰船运动在线预报研究[D]. 哈尔滨: 哈尔滨工程大学, 2016.
	HUANG Limin. On-line prediction of non-stationary and nonlinear ship motions at sea[D]. Harbin: Harbin Engineering University, 2016.
[10]	BEHERA A P, GAURISARIA M K, RAUTARAY S S, et al. Predicting future call volume using ARIMA models[C]//2021 5th International Conference on Intelligent Computing and Control Systems. Madurai, India: IEEE, 2021: 1351-1354.
[11]	仇琦, 杨兰, 丁旭, 等. 基于改进EMD-ARIMA的光伏发电系统短期功率预测[J]. 电力科学与工程, 2020, 36(8): 42-50. doi: 1672-0792(2020)08-0042-09
	QIU Qi, YANG Lan, DING Xu, et al. An improved short-term power prediction method of PV power generation system based on EMD-ARIMA model[J]. Electric Power Science and Engineering, 2020, 36(8): 42-50. doi: 1672-0792(2020)08-0042-09
[12]	尤保健. 基于六西格玛理论的轴流泵叶轮水力效率影响因子分析[J]. 水电能源科学, 2020, 38(2): 168-171.
	YOU Baojian. Influencing factors analysis of hydraulic efficiency of axial flow pump impeller based on six sigma theory[J]. Water Resources and Power, 2020, 38(2): 168-171.
[13]	王彬蓉, 王维博, 周超, 等. 基于EMD自适应重构的心音信号特征筛选及分类[J]. 航天医学与医学工程, 2020, 33(6): 533-541.
	WANG Binrong, WANG Weibo, ZHOU Chao, et al. Feature selection and classification of heart sound based on EMD adaptive reconstruction[J]. Space Medicine & Medical Engineering, 2020, 33(6): 533-541.
[14]	崔公哲, 张朝霞, 杨玲珍, 等. 一种改进的小波阈值去噪算法[J]. 现代电子技术, 2019, 42(19): 50-53.
	CUI Gongzhe, ZHANG Zhaoxia, YANG Lingzhen, et al. An improved wavelet threshold denoising algorithm[J]. Modern Electronics Technique, 2019, 42(19): 50-53.
[15]	CHANG F X, HONG W X, ZHANG T, et al. Research on wavelet denoising for pulse signal based on improved wavelet thresholding[C]//2010 First International Conference on Pervasive Computing, Signal Processing and Applications. New York, USA: IEEE, 2010: 564-567.
[16]	徐晨, 赵瑞珍, 甘小冰. 小波分析应用算法[M]. 北京: 科学出版社, 2016.
	XU Chen, ZHAO Ruizhen, GAN Xiaobing. Application algorithm of wavelet analysis[M]. Beijing: Science Press, 2016.
[17]	谢忠玉, 张立. 相空间重构参数选择方法的研究[J]. 中国科技信息, 2009(16): 42-43.
	XIE Zhongyu, ZHANG Li. Selection of embedding parameters in phase space reconstruction[J]. China Science and Technology Information, 2009(16): 42-43.
[18]	杨恒岳, 刘青荣, 阮应君. 基于k-means聚类算法的分布式能源系统典型日冷热负荷选取[J]. 热力发电, 2021, 50(3): 84-90.
	YANG Hengyue, LIU Qingrong, RUAN Yingjun. Selection of typical daily cooling and heating load of CCHP system based on k-means clustering algorithm[J]. Thermal Power Generation, 2021, 50(3): 84-90.
[19]	田行宇, 李传金. PP检验对异方差时间序列的伪检验[J]. 统计与决策, 2018, 34(17): 74-76.
	TIAN Xingyu, LI Chuanjin. Spurious tests to he-teroscedastic time series with PP test[J]. Statistics & Decision, 2018, 34(17): 74-76.
[20]	李国春, 王恩龙, 王丽梅, 等. 基于AIC准则判断锂电池最优模型[J]. 汽车工程学报, 2019, 9(5): 352-358.
	LI Guochun, WANG Enlong, WANG Limei, et al. Evaluating optimal models of lithium battery based on AIC[J]. Chinese Journal of Automotive Engineering, 2019, 9(5): 352-358.
[21]	李颖若, 韩婷婷, 汪君霞, 等. ARIMA时间序列分析模型在臭氧浓度中长期预报中的应用[J]. 环境科学, 2021, 42(7): 3118-3126.
	LI Yingruo, HAN Tingting, WANG Junxia, et al. Application of ARIMA model for mid-and long-term forecasting of ozone concentration[J]. Environmental Science, 2021, 42(7): 3118-3126. doi: 10.1021/es8002603 URL
[22]	饶国强, 冯辅周, 司爱威, 等. 排列熵算法参数的优化确定方法研究[J]. 振动与冲击, 2014, 33(1): 188-193.
	RAO Guoqiang, FENG Fuzhou, SI Aiwei, et al. Method for optimal determination of parameters in permutation entropy algorithm[J]. Journal of Vibration and Shock, 2014, 33(1): 188-193.
[23]	郑奕扬, 倪何, 金家善. 基于MSOP的蒸汽动力系统单参数运行稳定性评估方法[J]. 上海交通大学学报, 2021, 55(11): 1438-1444.
	ZHENG Yiyang, NI He, JIN Jiashan. An operation stability assessment method of a single-parameter in steam power system based on MSOP[J]. Journal of Shanghai Jiao Tong University, 2021, 55(11): 1438-1444.

IMF分量	相关系数	均方根误差
I₁	0.046 9	7.772 7
I₂	0.105 5	7.739 0
I₃	0.106 1	7.742 4
I₄	0.526 2	6.679 1
I₅	0.426 7	7.153 2
I₆	0.599 1	6.797 9
I₇	0.593 4	6.494 4

方法	IMF分量	信噪比	均方根误差
复合阈值函数	I₁	39.185 4	0.004 1
	I₂	36.503 3	0.013 8
	I₃	56.600 6	0.001 6
硬阈值函数	I₁	14.015 4	0.073 5
	I₂	38.981 8	0.010 3
	I₃	83.197 9	7.604 1×10^-5
软阈值函数	I₁	12.118 4	0.090 7
	I₂	32.954 7	0.020 7
	I₃	74.748 9	0.000 2

奇异值分量	τ_opt	m_opt	PE
P1	23	3	2.467 2
P2	12	2	0.975 2
P3	25	2	1.263 9
P4	18	2	1.212 7
P5	12	3	3.123 8
P6	9	3	2.972 9
P7	6	3	2.955 8
P8	6	4	4.709 8