Prediction of boiler water quality in the petrochemical industry based on BiLSTM-GRU model

doi:10.19965/j.cnki.iwt.2025-0062

Abstract

Abstract:

In order to effectively monitor the changes in boiler water quality during the petrochemical process and optimize the water treatment process, a BiLSTM-GRU model combining bidirectional long short-term memory(BiLSTM) and gated recurrent unit (GRU) was proposed. The model captured the dependency relationships in the time series of water quality parameters through BiLSTM, and combined GRU to improve training efficiency, achieving accurate prediction of key water quality indicators such as pH, dissolved oxygen, phosphate, and high-concentration silicate in boiler water. The results showed that the BiLSTM-GRU combined model had better mean absolute error (MAE) and root mean square error (RMSE) than both LSTM and LSTM-GRU models, demonstrating higher prediction accuracy. Validation using boiler water data from similar sites confirmed the strong robustness of the model, indicating its potential as an effective technical solution for online monitoring and precise control of industrial boiler water quality.

Key words: bidirectional long short-term memory network, gated recurrent unit, industrial boiler water, water quality prediction

摘要：

为有效掌握石化过程中锅炉水质变化情况并优化水处理过程，提出了一种结合双向长短期记忆网络（BiLSTM）与门控循环单元（GRU）的BiLSTM-GRU组合预测模型。该模型通过BiLSTM捕捉水质参数时间序列中的依赖关系，并结合GRU提升训练效率，实现对锅炉水质中pH、溶解氧、磷酸盐和高浓度硅酸盐等关键水质指标的精准预测。结果显示，BiLSTM-GRU组合模型的平均绝对误差（MAE）和均方根误差（RMSE）均优于LSTM和LSTM-GRU模型，表现出更高的预测精度。另外，采用同类型不同站点的锅炉水质数据进行验证，该模型也展现了良好的鲁棒性。这表明该模型可为工业锅炉水质的在线监测与精准控制提供有效技术支持。

关键词: 双向长短期记忆神经网络, 门控循环单元, 工业锅炉水, 水质预测

CLC Number:

X703
TP183

Yao CHEN, Zhengxi CHEN, Changwei LIU, Qingjie LIU, Yan WANG, Shengnan WANG. Prediction of boiler water quality in the petrochemical industry based on BiLSTM-GRU model[J]. Industrial Water Treatment, 2026, 46(1): 178-183.

陈瑶, 陈正茜, 刘昌伟, 刘庆洁, 王艳, 王胜楠. 基于BiLSTM-GRU模型的石化工业过程锅炉水质预测[J]. 工业水处理, 2026, 46(1): 178-183.

/ / Recommend / Download Citations

URL: https://www.iwt.cn/EN/10.19965/j.cnki.iwt.2025-0062

https://www.iwt.cn/EN/Y2026/V46/I1/178

Figures/Tables 9

Fig.1 Design process

Fig.2 Correlation coefficients among various parameters of boiler water

Fig.3 Feature construction

Fig.4 Cell structure of LSTM （a） and BiLSTM （b）

Fig.5 BiLSTM-GRU hybrid neural network structure

Fig.6 Comparison of prediction results of BiLSTM-GRU model for boiler water quality indicators

Table 1 Comparison of prediction errors of three models for boiler water quality parameters

参数	模型	MAE	RMSE
pH	LSTM	0.095 3	0.127 4
	LSTM-GRU	0.090 6	0.122 8
	BiLSTM-GRU	0.085 0	0.117 2
电导率	LSTM	4.901 8	7.360 8
	LSTM-GRU	4.837 8	7.215 6
	BiLSTM-GRU	4.872 7	7.091 1
磷酸盐	LSTM	0.095 3	0.127 4
	LSTM-GRU	0.090 6	0.122 8
	BiLSTM-GRU	0.085 0	0.117 2
高浓度硅酸盐	LSTM	29.838 3	30.468 8
	LSTM-GRU	26.158 0	29.981 1
	BiLSTM-GRU	21.553 3	24.800 3

Table 2 Prediction error results of all parameters in the generalization test station

参数	MAE	RMSE
pH	0.055 4	0.105 8
电导率	2.830 0	5.233 7
磷酸盐	0.629 8	1.574 2
高浓度硅酸盐	21.663 8	27.005 4

Fig.7 Comparison of prediction results of BiLSTM-GRU model for boiler water quality indicators in generalization test station

References 7

[1]	刘浩威，陈霖，李巨峰，等. 基于数据驱动的污水处理系统水质预测研究进展［J］. 工业水处理，2025，45（4）：1-9.
	LIU Haowei， CHEN Lin， LI Jufeng，et al. Research progress on water quality prediction in wastewater treatment systems based on data-driven approaches［J］. Industrial Water Treatment，2025，45（4）：1-9.
[2]	吴涛，颜辉武，唐桂刚. 三峡库区水质数据时间序列分析预测研究［J］. 武汉大学学报（信息科学版），2006，31（6）：500-502.
	WU Tao， YAN Huiwu， TANG Guigang. Prediction on time series analysis of water quality in Yangtze gorges reservoir area［J］. Geomatics and Information Science of Wuhan University，2006，31（6）：500-502.
[3]	郭杨，丁振军. 基于ARIMA模型的辽河流域水质主要指标预测研究［J］. 绿色科技，2021，23（24）：34-36.
	GUO Yang， DING Zhenjun. Prediction research on main indexes of water quality in Liaohe River Basin based on ARIMA model［J］. Journal of Green Science and Technology，2021，23（24）：34-36.
[4]	胡衍坤，王宁，刘枢，等. 时间序列模型和LSTM模型在水质预测中的应用研究［J］. 小型微型计算机系统，2021，42（8）：1569-1573.
	HU Yankun， WANG Ning， LIU Shu，et al. Research on application of time series model and LSTM model in water quality prediction［J］. Journal of Chinese Computer Systems，2021，42（8）：1569-1573.
[5]	杨潞霞，王梦冉，林兴亮，等. 基于CNN-LSTM的工业出水水质预测模型［J］. 净水技术，2024，43（5）：143-149.
	YANG Luxia， WANG Mengran， LIN Xingliang，et al. Prediction model of industrial effluent quality based on CNN-LSTM［J］. Water Purification Technology，2024，43（5）：143-149.
[6]	WANG Yuanyuan， ZHOU Jian， CHEN Kejia，et al. Water quality prediction method based on LSTM neural network［C］//2017 12th International Conference on Intelligent Systems and Knowledge Engineering（ISKE），Nanjing：IEEE，2017：1-5. doi:10.1109/iske.2017.8258814
[7]	石庆兰，束金阳，李道亮，等. 基于BiLSTM-GRU融合网络的稻虾养殖溶解氧含量预测［J］. 农业机械学报，2023，54（10）：364-370.
	SHI Qinglan， SHU Jinyang， LI Daoliang，et al. Dissolved oxygen prediction in rice and shrimp culture based on BiLSTM-GRU fusion neural networks［J］. Transactions of the Chinese Society for Agricultural Machinery，2023，54（10）：364-370.

[1]	Yongjie WAN, Ying ZHANG, Peng TONG, Xing TIAN, Hanhua HE, Zhenguo CHEN, Xiaojun WANG. Prediction of concentrated drainage water quality in cooling tower by interpretable machine learning model [J]. Industrial Water Treatment, 2026, 46(1): 156-162.
[2]	Chu ZHANG, Yida GE, Yanni LI, Zhengbo LI, Tian PENG. Prediction of wastewater quality by FAOA optimized outlier-robust ensemble deep RVFL [J]. Industrial Water Treatment, 2025, 45(9): 71-77.
[3]	Lin CHEN, Haowei LIU, Qinghong WANG, Guangming FENG, Yali ZHAN, Qiang WANG, Chunmao CHEN. Research on the prediction model of effluent quality in petrochemical industries wastewater treatment plants based on optimized machine learning algorithms [J]. Industrial Water Treatment, 2025, 45(7): 81-93.
[4]	Luxia YANG, Zhiyu WANG, Shuaijie SHEN, Yongjie MA, Yizheng FU. Research on sewage water quality prediction based on the PSO-LSTM-SATN model [J]. Industrial Water Treatment, 2025, 45(6): 159-166.
[5]	Lin CHEN, Xin YAN, Zhihe TANG, Zhaokuan RAN, Binlian LI, Hui LUAN, Chunmao CHEN. Water quality prediction model of air flotation unit based on optimization support vector regression machine [J]. Industrial Water Treatment, 2025, 45(5): 157-165.
[6]	Haowei LIU, Lin CHEN, Jufeng LI, Xin YAN, Zhaokuan RAN, Hui LUAN, Chunmao CHEN. Research progress on water quality prediction in wastewater treatment systems based on data-driven approaches [J]. Industrial Water Treatment, 2025, 45(4): 1-9.
[7]	Yifan YAO, Yushu JING, Liyan WANG, Changqing LIU. Prediction method of total nitrogen in wastewater treatment plant based on integrated model [J]. Industrial Water Treatment, 2023, 43(9): 187-194.
[8]	Zhu Zhiping, Li Maodong, Jin Peng, Yuan Hantao, Zhou Zhichao, Huang Xianjun. Study on the corrosion inhibition characteristic of HEDP in industrial boiler water [J]. INDUSTRIAL WATER TREATMENT, 2014, 34(10): 69-72.
[9]	Li Maodong, Zeng Bin, Yang Lin, Zhao Junming, Du Yuhui, Zhang Hui, Zhu Zhiping. Determination of HEDP in boiler water for industrial boilers [J]. INDUSTRIAL WATER TREATMENT, 2013, 33(1): 75-77.

Please choose a citation manager

Content to export