Machine Learning-Based NOx Prediction and Combustion Optimization for a Coal-Fired Boiler: The Role of Feature Engineering
- 1Southeast University , , ,
- 2Chinese Institute of Coal Science , ,
- 3Nanjing University , , ,
Abstract
The choice between traditional machine learning and deep learning for industrial time series prediction remains unsettled, particularly when feature engineering is accounted for. Using 10,000 operational samples from a 660 MW coal-fired boiler, we systematically compare support vector machine (SVM), LSTM, XGBoost, Random Forest, and MLP for NOx emission prediction. Using a 139-dimensional representation comprising the current and first-to-fourth-order lagged values of 27 operational variables, together with four autoregressive NOx lag terms, SVM achieves RMSE = 10.124 mg/Nm3 and R2 = 0.934, outperforming LSTM (20.203, 0.738), XGBoost (12.405, 0.901), Random Forest (12.475, 0.900), and MLP (32.236, 0.332). A fair re-evaluation─training LSTM, Transformer, TCN, and BiLSTM on raw 20-step sequences without manual lag engineering─confirms that deep architectures remain less competitive than SVM with explicit feature engineering (best DL model: fair LSTM, RMSE = 18.541, R2 = 0.779). The HAC-adjusted Diebold–Mariano tests (all one-sided p < 8.1 × 10–7), bootstrap confidence intervals, and Friedman–Nemenyi analysis provide convergent statistical evidence of SVM’s superiority. SHAP analysis identifies O2 as the dominant predictor (SHAP = 10.987), and time-series cross-validation confirms SVM’s robustness (mean RMSE = 6.062, std = 1.804). Single-objective differential evolution achieves a 21.16% NOx reduction at 600 MW load. A data-driven efficiency surrogate replaces hand-crafted formulas in multiobjective NSGA-II optimization, producing a scenario-dependent NOx-efficiency Pareto front whose sensitivity to coal calorific value is explicitly evaluated. These findings demonstrate that on industrial data sets of approximately 10,000 samples, domain-informed feature engineering can enable traditional methods to match or exceed deep learning at lower computational cost.
Related
- Sugarcane bagasse-MnFe2O4 magnetic biochar composite for ammonium removal: isotherms, kinetics, RSM optimization and machine learning prediction
- Machine-learning prediction of freezing temperature in Pb-contaminated soils remediated with biochar
- FastNet: Improving the Physical Consistency of Machine Learning Weather Prediction Models through Loss Function Design
- AIFS-DOP: End-to-End Medium-Range Weather Prediction from Observations Alone with Machine Learning
- A novel deep learning prediction method for PM2.5 integrating improved MSTL decomposition and FATA optimization algorithm
BibTeX
@article{li2026machine,
title = {{Machine Learning-Based NOx Prediction and Combustion Optimization for a Coal-Fired Boiler: The Role of Feature Engineering}},
author = {Peixuan Li and Dining Ma and Shicong Jiang and Ruiqi Zhou},
journal = {ACS Omega},
year = {2026},
month = oct,
doi = {10.1021/acsomega.6c09413},
url = {https://pubs.acs.org/acsodf/article/11/41/64143/5444325/Machine-Learning-Based-NOx-Prediction-and},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Machine Learning-Based NOx Prediction and Combustion Optimization for a Coal-Fired Boiler: The Role of Feature Engineering
AU - Peixuan Li
AU - Dining Ma
AU - Shicong Jiang
AU - Ruiqi Zhou
PY - 2026
DA - 2026/10/05
JO - ACS Omega
DO - 10.1021/acsomega.6c09413
UR - https://pubs.acs.org/acsodf/article/11/41/64143/5444325/Machine-Learning-Based-NOx-Prediction-and
AB - The choice between traditional machine learning and deep learning for industrial time series prediction remains unsettled, particularly when feature engineering is accounted for. Using 10,000 operational samples from a 660 MW coal-fired boiler, we systematically compare support vector machine (SVM), LSTM, XGBoost, Random Forest, and MLP for NOx emission prediction. Using a 139-dimensional representation comprising the current and first-to-fourth-order lagged values of 27 operational variables, together with four autoregressive NOx lag terms, SVM achieves RMSE = 10.124 mg/Nm3 and R2 = 0.934, outperforming LSTM (20.203, 0.738), XGBoost (12.405, 0.901), Random Forest (12.475, 0.900), and MLP (32.236, 0.332). A fair re-evaluation─training LSTM, Transformer, TCN, and BiLSTM on raw 20-step sequences without manual lag engineering─confirms that deep architectures remain less competitive than SVM with explicit feature engineering (best DL model: fair LSTM, RMSE = 18.541, R2 = 0.779). The HAC-adjusted Diebold–Mariano tests (all one-sided p < 8.1 × 10–7), bootstrap confidence intervals, and Friedman–Nemenyi analysis provide convergent statistical evidence of SVM’s superiority. SHAP analysis identifies O2 as the dominant predictor (SHAP = 10.987), and time-series cross-validation confirms SVM’s robustness (mean RMSE = 6.062, std = 1.804). Single-objective differential evolution achieves a 21.16% NOx reduction at 600 MW load. A data-driven efficiency surrogate replaces hand-crafted formulas in multiobjective NSGA-II optimization, producing a scenario-dependent NOx-efficiency Pareto front whose sensitivity to coal calorific value is explicitly evaluated. These findings demonstrate that on industrial data sets of approximately 10,000 samples, domain-informed feature engineering can enable traditional methods to match or exceed deep learning at lower computational cost.
SN - 2470-1343
ER -