Feature Extraction Method for Wet Flue Gas Desulfurization Under Operating Conditions Based on Mutual Information and PCA Theory

doi:10.11930/j.issn.1004-9649.201910099

Abstract

Abstract: So far the principal component analysis is still the most commonly used method for feature extraction in the optimization research on wet flue gas desulfurization (WFGD) system. However, due to the nonlinearities and the high mutual coupling of the factors affecting the energy consumption of the WFGD system, there is no feasible way for the existing feature extraction methods to evaluate the defects of the non-linear relationship between the features. Therefore this paper proposes a feature extraction method based on mutual information and principal component analysis (MI-PCA), in which the covariance matrix in principal component analysis is replaced with a mutual information matrix between features. The eigenvector represents the direction of each principal component in the new principal component space, while the eigenvalue determines the principal component dimension as the evaluation criterion. Through the implementation of the proposed method the characteristics of the measured desulfurization data of a power plant is therefore extracted. From the experimental results this method has demonstrated better dimensionality reduction effect. With the aid of the support vector machine based on grid search method as the classifier, higher classification accuracy is observed assuming the same dimensions of the principal components. Finally, by using this method in the optimization of the operation mode of the slurry circulation pump, the average power consumption has been reduced by about 14.69%.

Key words: wet flue gas desulfurization, feature extraction, mutual information, principal component analysis, support vector machine

LIU Wenhui, XU Zunyi, ZHANG Xuran, ZHANG Haiyan. Feature Extraction Method for Wet Flue Gas Desulfurization Under Operating Conditions Based on Mutual Information and PCA Theory[J]. Electric Power, 2020, 53(8): 158-163.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.electricpower.com.cn/EN/10.11930/j.issn.1004-9649.201910099

https://www.electricpower.com.cn/EN/Y2020/V53/I8/158

References

[1] 国家环境保护部. 火电厂大气污染物排放标准: GB13223—2011 [S]. 北京: 中国标准出版社, 2011.
[2] 崔仕文, 铁治欣, 丁成富, 等. 基于偏最小二乘支持向量机的烟气湿法脱硫效率预测模型[J]. 热力发电, 2017, 46(4): 81-87
CUI Shiwen, TIE Zhixin, DING Chengfu, et al. Prediction model for flue gas wet desulfurization efficiency based on partial least squares support vector machine[J]. Thermal Power Generation, 2017, 46(4): 81-87
[3] 李兴华, 何育东. 燃煤火电机组SO₂超低排放改造方案研究[J]. 中国电力, 2015, 48(10): 148-151
LI Xinghua, HE Yudong. Study on modification of ultra-low SO₂ emission in coal-fired power plants[J]. Electric Power, 2015, 48(10): 148-151
[4] 郭彦鹏, 潘丹萍, 杨林军. 湿法烟气脱硫中石膏雨的形成及其控制措施[J]. 中国电力, 2014, 47(3): 152-154, 159
GUO Yanpeng, PAN Danping, YANG Linjun. Formation and control of gypsum rain in wet flue gas desulfurization[J]. Electric Power, 2014, 47(3): 152-154, 159
[5] HUANG C J, TAN G B, CHEN Y Q. Application of wet flue gas desulfurization control system for 2×200 MW units [C]//International Conference on Artificial Intelligence, IEEE. Dengleng, 2011: 318-324.
[6] 刘黎伟. 石灰石-石膏湿法脱硫系统运行优化研究[J]. 电力科技与环保, 2019, 35(1): 35-36
LIU Liwei. Optimal operation research of limestone-gypsum wet flue gas desulfurization[J]. Electric Power Technology and Environmental Protection, 2019, 35(1): 35-36
[7] 段玖祥, 李小龙, 蔡培, 等. 超低排放燃煤电厂WFGD系统优化运行探讨[J]. 环境工程, 2017, 35(12): 73-76
DUAN Jiuxiang, LI Xiaolong, CAI Pei, et al. Optimization operation of WFGD systems in ultra-low emission coal-fired power plants[J]. Environmental Engineering, 2017, 35(12): 73-76
[8] 杨用龙, 苏秋凤, 张杨, 等. 双塔双循环脱硫系统优化与经济性运行研究[J]. 中国电力, 2018, 51(4): 136-142
YANG Yonglong, SU Qiufeng, ZHANG Yang, et al. System optimization and economical operation of the series absorption tower[J]. Electric Power, 2018, 51(4): 136-142
[9] 刘敏, 周然, 郑川江, 等. 火电厂脱硫系统运行优化控制模型研究[J]. 热能动力工程, 2017, 32(6): 95-99
LIU Min, ZHOU Ran, ZHENG Chuanjiang, et al. Study on operation optimization and control model for desulfurization system of thermal power plants[J]. Journal of Engineering for Thermal Energy and Power, 2017, 32(6): 95-99
[10] 许丹, 沈凯, 张亚平, 等. 基于模糊理论的湿法脱硫故障诊断和优化模型研究[J]. 环境工程, 2018, 36(2): 92-97
XU Dan, SHEN Kai, ZHANG Yaping, et al. Study of fault diagnosis and operation optimization models of WFGD based on fuzzy theory[J]. Environmental Engineering, 2018, 36(2): 92-97
[11] YANG Z K, LIU C Y, SONG X L, et al. Application of RBF neural network PID in wet flue gas desulfurization of thermal power plant [C]// International Conference on Machine Learning & Cybernetics, IEEE. Jeju, 2017: 301–306.
[12] 朱竹军. 基于专家控制的炉外湿法脱硫自动控制研究及应用[J]. 自动化与仪表, 2019, 34(1): 24-27, 32
ZHU Zhujun. Research and application of wet desulfurization automatic control based on expert control[J]. Automation & Instrumentation, 2019, 34(1): 24-27, 32
[13] JOLLIFFE I T. Principal component analysis [M]. New York: Springer-Verlag, 1986: 111–137.
[14] 张东平, 孙克勤, 潘效军, 等. 湿法烟气脱硫吸收塔仿真模型开发及应用[J]. 环境科学与技术, 2011, 34(3): 189-192
ZHANG Dongping, SUN Keqin, PAN Xiaojun, et al. A simulation model of wet scrubbers for flue gas desulfurization[J]. Environmental Science & Technology, 2011, 34(3): 189-192
[15] 田丽玲, 张丽萍. 一种数据驱动的湿法烟气脱硫系统的故障诊断方法[J]. 福州大学学报(自然科学版), 2012, 40(1): 82-86
TIAN Liling, ZHANG Liping. A data-driven fault diagnosis of wet flue gas desulfurization system[J]. Journal of Fuzhou University (Natural Science Edition), 2012, 40(1): 82-86
[16] BATTITI R. Using mutual information for selecting features in supervised neural net learning[J]. Neural Networks IEEE Transactions on, 1994, 5(4): 537-550.
[17] AMJADY N, DARAEEPOUR A. Day-ahead price forecasting of electricity markets by combination of mutual information technique and neural network [C]// Power and Energy Society General Meeting—Conversion and Delivery of Electrical Energy in the 21st Century, IEEE. Pittsburgh, PA, 2008: 1–7.
[18] KATO Y, SAITO H, EJIMA T, et al. An application of SVM: alphanumeric character recognition [C]//International 1989 Joint Conference on Neural Networks, IEEE. Washington DC, 1989: 576.
[19] 纪昌明, 周婷, 向腾飞, 等. 基于网格搜索和交叉验证的支持向量机在梯级水电系统隐随机调度中的应用[J]. 电力自动化设备, 2014, 34(3): 125-131
JI Changming, ZHOU Ting, XIANG Tengfei, et al. Application of support vector machine based on grid search and cross validation in implicit stochastic dispatch of cascaded hydropower stations[J]. Electric Power Automation Equipment, 2014, 34(3): 125-131
[20] YANG M Y, LI W Q, ZHANG H Y, et al. Parameters optimization improvement of SVM on load forecasting [C]//International Conference on Intelligent Human—Machine Systems and Cybernetics, IEEE. Hangzhou, 2016: 257–260.
[21] SUN Y Y, WANG Y M, GUO L L, et al. The comparison of optimizing SVM by GA and grid search [C]//International Conference on Electronic Measurement & Instruments, IEEE. Yangzhou, 2017: 354–360.