Data cleaning method for the process of acid production with flue gas based on improved random forest

doi:10.1016/j.cjche.2022.12.013

中国化学工程学报 ›› 2023, Vol. 59 ›› Issue (7): 72-84.DOI: 10.1016/j.cjche.2022.12.013

• Full Length Article • 上一篇下一篇

Data cleaning method for the process of acid production with flue gas based on improved random forest

Xiaoli Li^1,2, Minghua Liu¹, Kang Wang¹, Zhiqiang Liu³, Guihai Li⁴

1. Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China;
2. Beijing Key Laboratory of Computational Intelligence and Intelligent System, Engineering Research Center of Digital Community, Ministry of Education, Beijing 100124, China;
3. Guixi Smelter, Jiangxi Copper Co., Ltd, Guixi 335400, China;
4. Beijing RTlink Technology Co., Ltd, Beijing 100102, China

收稿日期:2022-05-17 修回日期:2022-12-24 出版日期:2023-07-28 发布日期:2023-10-14
通讯作者: Minghua Liu,E-mail:liumh829@163.com
基金资助:
This study is supported by the National Natural Science Foundation of China (61873006) and Beijing Natural Science Foundation (4204087, 4212040).

Data cleaning method for the process of acid production with flue gas based on improved random forest

Xiaoli Li^1,2, Minghua Liu¹, Kang Wang¹, Zhiqiang Liu³, Guihai Li⁴

1. Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China;
2. Beijing Key Laboratory of Computational Intelligence and Intelligent System, Engineering Research Center of Digital Community, Ministry of Education, Beijing 100124, China;
3. Guixi Smelter, Jiangxi Copper Co., Ltd, Guixi 335400, China;
4. Beijing RTlink Technology Co., Ltd, Beijing 100102, China

Received:2022-05-17 Revised:2022-12-24 Online:2023-07-28 Published:2023-10-14
Contact: Minghua Liu,E-mail:liumh829@163.com
Supported by:
This study is supported by the National Natural Science Foundation of China (61873006) and Beijing Natural Science Foundation (4204087, 4212040).

摘要/Abstract

摘要： Acid production with flue gas is a complex nonlinear process with multiple variables and strong coupling. The operation data is an important basis for state monitoring, optimal control, and fault diagnosis. However, the operating environment of acid production with flue gas is complex and there is much equipment. The data obtained by the detection equipment is seriously polluted and prone to abnormal phenomena such as data loss and outliers. Therefore, to solve the problem of abnormal data in the process of acid production with flue gas, a data cleaning method based on improved random forest is proposed. Firstly, an outlier data recognition model based on isolation forest is designed to identify and eliminate the outliers in the dataset. Secondly, an improved random forest regression model is established. Genetic algorithm is used to optimize the hyperparameters of the random forest regression model. Then the optimal parameter combination is found in the search space and the trend of data is predicted. Finally, the improved random forest data cleaning method is used to compensate for the missing data after eliminating abnormal data and the data cleaning is realized. Results show that the proposed method can accurately eliminate and compensate for the abnormal data in the process of acid production with flue gas. The method improves the accuracy of compensation for missing data. With the data after cleaning, a more accurate model can be established, which is significant to the subsequent temperature control. The conversion rate of SO₂ can be further improved, thereby improving the yield of sulfuric acid and economic benefits.

关键词: Acid production, Data cleaning, Isolation forest, Random forest, Data compensation

Abstract: Acid production with flue gas is a complex nonlinear process with multiple variables and strong coupling. The operation data is an important basis for state monitoring, optimal control, and fault diagnosis. However, the operating environment of acid production with flue gas is complex and there is much equipment. The data obtained by the detection equipment is seriously polluted and prone to abnormal phenomena such as data loss and outliers. Therefore, to solve the problem of abnormal data in the process of acid production with flue gas, a data cleaning method based on improved random forest is proposed. Firstly, an outlier data recognition model based on isolation forest is designed to identify and eliminate the outliers in the dataset. Secondly, an improved random forest regression model is established. Genetic algorithm is used to optimize the hyperparameters of the random forest regression model. Then the optimal parameter combination is found in the search space and the trend of data is predicted. Finally, the improved random forest data cleaning method is used to compensate for the missing data after eliminating abnormal data and the data cleaning is realized. Results show that the proposed method can accurately eliminate and compensate for the abnormal data in the process of acid production with flue gas. The method improves the accuracy of compensation for missing data. With the data after cleaning, a more accurate model can be established, which is significant to the subsequent temperature control. The conversion rate of SO₂ can be further improved, thereby improving the yield of sulfuric acid and economic benefits.

Key words: Acid production, Data cleaning, Isolation forest, Random forest, Data compensation

Xiaoli Li, Minghua Liu, Kang Wang, Zhiqiang Liu, Guihai Li. Data cleaning method for the process of acid production with flue gas based on improved random forest[J]. 中国化学工程学报, 2023, 59(7): 72-84.

Xiaoli Li, Minghua Liu, Kang Wang, Zhiqiang Liu, Guihai Li. Data cleaning method for the process of acid production with flue gas based on improved random forest[J]. Chinese Journal of Chemical Engineering, 2023, 59(7): 72-84.

参考文献

[1] X. Li, S. Zhou, Process selection and adapTableility evaluation of sulphuric acid plant based on copper smelting flue gas, Sulphuric Acid Industry 2020 (07) (2020) 17-21+24
[2] Z. Pi, Z. Zhang, K. Wang, W. Chai, Treatment technology on wet plume of tail gas from copper concentrate drying, Nonferrous Metals Engineering Research 42 (03) (2021) 25-27.
[3] X.L. Li, S.Q. Shen, S.X. Yang, K. Wang, Y. Li, Analysis and multi-objective optimization of slag powder process, Appl. Soft Comput. 96 (2020) 106587.
[4] Z. Jiang, C. Xie, Y. Yao, H. Cheng, Summary of current status of sulphuric acid plant with smelting gas desulfurization process, Sulphuric Acid Industry (12) (2019) 29-32.
[5] K. Giri, T.K. Biswas, P. Sarkar, ECR-DBSCAN: an improved DBSCAN based on computational geometry, Mach. Learn. Appl. 6 (2021) 100148.
[6] X. Li, J. Gao, K. Wang, Multi-objective optimization of mill inlet and outlet temperature in slag powder production process, Control Theory and Applications 37 (02) (2020) 275-282.
[7] R. Zhao, B. Du, L.P. Zhang, A robust nonlinear hyperspectral anomaly detection approach, IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. 7 (4) (2014) 1227-1234.
[8] J. Zhai, Y. Wang, Research on network anomaly detection algorithm based on fuzzy clustering, Electronic Measurement Technology 42 (16) (2019) 172-176.
[9] T. Tian, M. Shi, X. Song, Y. Ma, X. Feng, Anomaly detecting method for time series based on sliding windows, Instrument Technique and Sensor (7) (2021) 112-116.
[10] H. Fei, G. Li, Abnormal data detection algorithm for wsn based on k-means clustering, Computer Engineering 41 (07) (2015) 124-128.
[11] H.G. Han, Z.F. Zhao, X.L. Wu, S.H. Yang, Z. He, N. Zhao, Data cleaning method for municipal wastewater treatment based on improved random forest, J. Beijing Univ. Technol. 47 (2021) (5)421-430.
[12] C. Zhen, T. Yin, Short term wind power prediction based on parameter optimization of the variational modal decomposition and support vector regression, International Core Journal of Engineering 7 (10) (2021) 421-430.
[13] L. Gao, Y. Yang, X. Ren, J. Yu, Q. Han, MPC strategy of air source heat pump temperature based on bp neural network, Control Engineering 28 (09) (2021) 1765-1772.
[14] L.J. Chai, Y.N. Cai, M. Gao, Y. Hao, J.K. Chen, L. Jiang, A forecasting aided state estimation for distribution network based on weighted average interpolation and cubature Kalman filter, Electr. Power Constr. 42 (2021) 1.
[15] Z. Lv, J. Zhao, Y. Liu, W. Wang, Missing data imputation based on maximal variance weight information coefficient for gas flow in steel industry, Control Theory and Applications 32 (05) (2015) 646-654.
[16] Z.H. Pang, G.P. Liu, D.H. Zhou, D.H. Sun, Data-driven control with input design-based data dropout compensation for networked nonlinear systems, IEEE Trans. Control Syst. Technol. 25 (2) (2017) 628-636.
[17] J. Gu, Z. Dong, C. Zhang, X. Du, M. Guizani, Dynamic stress measurement with sensor data compensation, Electronics 8 (8) 859.
[18] C.S. Zhu, S.H. Li, Random forest regression model based on improved fruit fly optimization algorithm and its application in wind speed forecasting, J. Lanzhou Univ. Technol. 47 (2021) (4)83-90.
[19] L.J. Feng, C.H. Zhao, B. Huang, Adversarial smoothing tri-regression for robust semi-supervised industrial soft sensor, J. Process. Control 108 (2021) 86-97.
[20] W.K. Yu, C.H. Zhao, Robust monitoring and fault isolation of nonlinear industrial processes using denoising autoencoder and elastic net, IEEE Trans. Control Syst. Technol. 28 (3) (2020) 1083-1091.
[21] Y.X. Zhang, Y.N. Dong, K. Wu, T. Chen, Hyperspectral anomaly detection with otsu-based isolation forest, IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens. 14 (2021) 9079-9088.
[22] F.T. Liu, K.M. Ting, Z.H. Zhou, Isolation forest, In: 2008 eighth IEEE international conference on Data Mining, Pisa, Italy, 2008.
[23] Y.L. Ao, H.Q. Li, L.P. Zhu, S. Ali, Z.G. Yang, The linear random forest algorithm and its advantages in machine learning assisted logging regression modeling, J. Petroleum Sci. Eng. 174 (2019) 776-789.
[24] Z. Guo, B. Yu, M.Y. Hao, W.S. Wang, Y. Jiang, F. Zong, A novel hybrid method for flight departure delay prediction using Random Forest Regression and Maximal Information Coefficient, Aerosp. Sci. Technol. 116 (2021) 106822.
[25] J.L. Speiser, M.E. Miller, J. Tooze, E. Ip, A comparison of random forest variable selection methods for classification prediction modeling, Expert Syst Appl 134 (2019) 93-101.
[26] L. Breiman, Random forests, Mach. Learn.45 (1) (2001) 5-32.
[27] S. Katoch, S.S. Chauhan, V. Kumar, A review on genetic algorithm: past, present, and future, Multimed. Tools Appl.80 (5) (2021) 8091-8126.
[28] H. Liu, S.Y. Shi, P. Yang, J.M. Yang, An improved genetic algorithm approach on mechanism kinematic structure enumeration with intelligent manufacturing, J. Intell. Robotic Syst.89 (3-4) (2018) 343-350.
[29] Y. Sun, J.R. Li, X.L. Fu, H.F. Wang, H.H.Li, Application research based on improved genetic algorithm in cloud task scheduling, J. Intell. Fuzzy Syst. 38 (1) (2020) 239-246.
[30] C.C. Ding, L. Chen, B.R. Zhong, Exploration of intelligent computing based on improved hybrid genetic algorithm, Clust. Comput.22 (4) (2019) 9037-9045.

Data cleaning method for the process of acid production with flue gas based on improved random forest

Data cleaning method for the process of acid production with flue gas based on improved random forest

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	Xinhai Zhou, Dawei Zhou, Xinhui Bao, Yang Zhang, Jie Zhou, Fengxue Xin, Wenming Zhang, Xiujuan Qian, Weiliang Dong, Min Jiang, Katrin Ochsenreither. Production of palmitoleic acid by oleaginous yeast Scheffersomyces segobiensis DSM 27193 using systematic dissolved oxygen regulation strategy[J]. 中国化学工程学报, 2023, 53(1): 324-331.
[2]	Hanxiao Du, Lixin Xie, Jie Liu, Shichang Xu. Concentration of mixed acid by electrodialysis for the intensification of absorption process in acrylic acid production[J]. 中国化学工程学报, 2021, 36(8): 10-18.