Tracking the information about your manuscript
Communicate with the editorial office
Query manuscript payment status Editor LoginCollecting, editing, reviewing and other affairs offices
Managing manuscripts
Managing author information and external review Expert Information Peer ReviewOnline Review
Online Communication with the Editorial Department
Download Center
Online First
Discussion on the effect of carbon oxide loading on ion exchange purification of flue gas decarbonization amine solution
2026-09-01Research on safety and reliability guarantee of coal-fired power units in Hainan power grid under the new power system
2026-09-01Optimization study on windward sector layout of cooling towers based on unequal height filling under crosswind conditions
2026-09-01Study on steam flow and blast heating characteristics of the low pressure cylinder of a steam turbine under zero output condition
2026-09-01Ammonia absorption coupled with sulfur dioxide depolarized electrolysis for flue gas desulfurization and hydrogen production
2026-09-01Flexible peak-regulation dispatch of an oxy-fuel combustion integrated energy system considering dynamic oxygen-supply constraints
2026-09-01Cross-machine redundancy synchronization mechanism of thermal power virtual DCS controllers under Hypervisor architecture
2026-08-25Cross-condition few-shot fault diagnosis of boilers based on prototypical metric and mixed state-space model
2026-08-24Multi-objective optimization of boiler combustion based on BO-XGBoost and improved NSGA-II
2026-08-24Research on VLM semantic reasoning based alarm re-examination for thermal power plant safety monitoring
more..Journal introduction
Founded in: 1972
Headed by: China Huaneng Group Co., Ltd.
Sponsored by: Xi'an Thermal Power Research Institute Co., Ltd., Chinese Society for Electrical Engineering
Periodicity: Monthly
Standard Serial Number:CN 61-1111/TM, ISSN 1002-3364
TEL: (86-029) 82002270
E-mail: rlfdzzs@tpri.com.cn
Optimization of start-up and shutdown processes for coal-fired units based on LightGBM-Transformer model and NSGA-Ⅱ
Ding Pengkai;Chen Heng;Huang Zijian;Zhang Zhaoqing;Liu Wenyi;Pan Peiyuan;[Objective] With the high penetration of renewable energy in power systems, coal-fired units are required to provide peak-shaving services with higher frequency and larger depth, making start-up load ramp-up and shutdown load ramp-down more time-critical and cost-sensitive. This study aims to improve the responsiveness and economic performance of non-steady-state operating phases during start-stop transitions under deep peak-shaving conditions, while ensuring compliance with operational safety constraints. [Methods] Multi-month DCS data from a 660 MW coal-fired unit sampled at 5 s were analyzed. Non-steady segments in start-up and shutdown were firstly identified by combining sliding-window range thresholds with steady-state criteria, yielding transient datasets aligned with the underlying operational transitions. Candidate variables were then screened using the maximal information coefficient to retain features most strongly associated with output power, supplemented with a process progress descriptor to represent phase evolution and improve discrimination across different load intervals. To accommodate cross-month operating variability, start-stop events were organized into source and target domains. Within each target-domain segment, a time-stratified split was adopted, with the first 10% used for validation and the remaining 90% reserved for testing, thereby preserving temporal causality and avoiding leakage from future samples. Given the distinct dynamics of start-up load ramp-up and shutdown load ramp-down, two surrogate predictors were developed(LightGBM for tabular feature interactions and a Transformer for temporal dependency representation), and their hyperparameters were tuned via Bayesian optimization. After model selection, sourcedomain data were combined with the target-domain validation subset for final training. The trained surrogate models were embedded in an NSGA-II multi-objective optimization framework. Under safety constraints and ramp-rate limits, the optimization jointly improved load-ramping capability and reduced fuel consumption, producing executable set-point trajectories and load-change strategies suitable for operational guidance. [Results] The phasespecific surrogate models achieved stronger predictive performance than common baselines such as XGBoost and random forests across operating conditions, and maintained robust generalization in cross-month evaluation. When applied to representative non-steady segments, NSGA-Ⅱ produced feasible set-point trajectories that satisfied the imposed constraints throughout the transient. Relative to baseline operation, the optimized strategies reduced the average duration of non-steady start-stop segments by approximately 5.34% and lowered cumulative fuel consumption by approximately 3.53%, indicating concurrent gains in speed and economy. [Conclusion] A datadriven hybrid modeling and multi-objective evolutionary optimization approach is developed for coal-fired unit start-up and shutdown under deep peak-shaving conditions. By coupling phase-specific power surrogates with constraint-aware optimization, the method generates implementable set-point trajectories that jointly accelerate transient response and reduce fuel use within safe operating envelopes, providing practical support for flexible operation and online decision assistance.
Day-ahead electricity price forecasting for electricity spot market based on hybrid LSTM-Transformer model
Zhu Runze;Wang Dejun;Zhang Zongxing;Zhang You;[Objective] The large-scale integration of renewable energy into the power grid has significantly exacerbated the risks of volume and price volatility in the electricity spot market, thereby creating an urgent need to improve electricity price forecasting capability. To address this challenge, this study constructs a hybrid LSTMTransformer model based on actual operational data from Shandong Province, China. The primary aim is to achieve accurate day-ahead electricity price forecasting, thereby providing robust support for market operation optimization and risk management and control. [Methods] First, in the process of multi-dimensional feature fusion, the operational mechanisms of key influencing factors on electricity prices are systematically analyzed. The Pearson correlation coefficient is employed to quantify the linear correlation strength between these factors and electricity prices. This approach effectively resolves the problems inherent in traditional methods, such as unclear forecasting mechanisms and insufficient accuracy, which typically arise from relying on single features or empirical judgments. Second, the Transformer model is optimized by integrating LSTM, self-attention, and multi-layer attention mechanisms. This integration establishes a hierarchical framework that combines the strengths of sequential feature extraction and global dependency modeling. As a result, the model effectively enhances the ability to capture the complex long-and short-term dependencies embedded in electricity price sequences. Furthermore, a strategy of “deep mining of time-domain features coupled with frequency-domain loss constraints” is adopted. By fusing timedomain and frequency-domain losses, the model improves its fitting accuracy for both price fluctuations and numerical values. Meanwhile, a similar-day model is introduced to construct feature samples that closely reflect real-world conditions, which further enhances the forecasting accuracy. [Results] Through comprehensive case study analyses, the results demonstrate that the proposed model operates stably even under market conditions characterized by complex influencing factors and severe price fluctuations. The model significantly improves the forecasting accuracy of day-ahead electricity prices compared to conventional approaches. Quantitative evaluations show that the model achieves superior performance in tracking price trends and capturing sudden price spikes, which are common in high-renewable penetration scenarios. [Conclusion] The constructed LSTM-Transformer hybrid model, combined with multi-dimensional feature fusion and the time-frequency domain loss constraint strategy, can effectively address the challenges of volume and price fluctuations brought about by the highproportion integration of renewable energy. It provides a high-precision and high-robustness solution for day-ahead electricity price forecasting in the electricity spot market, thereby contributing to more stable and efficient market operations.
Optimization of operation strategy for single-tower double-loop desulfurization system based on improved NSGA-Ⅱ algorithm
Zhao Zheng;Jiang Guiwen;Tang Xinyu;He Peng;[Objective] Driven by stringent environmental regulations, coal-fired power plants urgently require highly efficient and intelligent emission control technologies. The widely utilized single-tower dual-loop wet flue gas desulfurization(WFGD) system often faces significant challenges during practical operation, notably severe fluctuations in sulfur dioxide(SO2) emissions at the system outlet and high desulfurization costs. These issues primarily stem from the system's complex nonlinear dynamics, large inertia, and time-delay characteristics. To address these operational bottlenecks, this study proposes a novel hybrid optimization strategy integrating an advanced deep-learning prediction model with a multi-objective optimization algorithm. The primary goal is to achieve a win-win operational paradigm, ensuring strict environmental compliance while simultaneously minimizing economic expenditures. [Methods] The proposed strategy is implemented through a systematic twostage framework. First, a predictive model is constructed by fusing the Transformer architecture with a bidirectional long short-term memory(BiLSTM) network. The Transformer effectively captures global long-range dependencies of operational data, while the BiLSTM extracts complex bidirectional temporal features. To overcome the limitations of empirical hyperparameter tuning, an improved black-winged kite algorithm(BKA) is introduced. This heuristic algorithm systematically optimizes the model's key hyperparameters, significantly enhancing the forecasting accuracy of the outlet SO2 concentration. Second, a comprehensive multi-objective optimization model is established, taking into account both the predicted outlet SO2 concentration and the system's overall desulfurization costs as conflicting objectives. To solve this complex trade-off problem, an improved non-dominated sorting genetic algorithm II(NSGA-Ⅱ) is applied to optimize operational parameters and derive precise setpoints for key controllable variables. [Results] Extensive experimental results demonstrate that the improved BKAoptimized Transformer-BiLSTM prediction model achieves superior performance. It significantly outperforms baseline models, including standard BiLSTM, CNN-BiLSTM, and the unoptimized Transformer-BiLSTM, across critical evaluation metrics such as root mean square error(RMSE), mean absolute error(MAE), and mean absolute percentage error(MAPE). Furthermore, by utilizing the improved NSGA-Ⅱ algorithm, the optimization system successfully outputs a set of well-distributed Pareto optimal solutions. These solutions effectively balance environmental protection and economic efficiency, providing power plant operators with a diverse range of flexible control strategies tailored to varying load conditions. [Conclusion] The proposed hybrid optimization strategy proves highly effective in the intelligent management of the single-tower dual-loop WFGD system. By integrating accurate deep learning predictions with robust multi-objective optimization, the strategy successfully mitigates outlet SO2 fluctuations and substantially reduces desulfurization costs. This research rigorously validates the framework's ability to achieve a balance between environmental protection and economic efficiency, offering a practical technical reference for upgrading industrial emission control systems.
Carbon dioxide emission prediction model for coal-fired power plants based on CNN-BiLSTM-Attention
He Haifei;Zhou Liangsong;Chen Junpeng;Zhao Zihang;Chen Erqiang;Zhi Changshuang;Miao Shihong;[Objective] Coal-fired power plants play a pivotal role in China's power system, and accurate forecasting of their CO2 emissions is crucial for achieving energy-saving and emission-reduction targets. To address this challenge, this paper proposes a short-term CO2 emission prediction model for coal-fired power plants based on CNN-BiLSTM-Attention. [Methods] First, feature selection is performed by analyzing the correlation between CO2 emission rates and various input variables, including unit load, SO2 emission rate, NOx emission rate, and environmental factors, thereby reducing redundant information that may impair model training. Specifically, the Pearson correlation is used to quantify the linear relationship between each candidate feature and the CO2 emission rate, and features with a correlation coefficient below 0.2 are excluded. The raw operational data are then normalized to the range [0,1] and segmented into samples using a sliding time window to capture temporal dynamics. Subsequently, the selected features are fed into the proposed CNN-BiLSTM-Attention model. In this hybrid architecture, the CNN layer extracts local patterns from the input sequence, the BiLSTM layer captures both forward and backward temporal dependencies, and the attention mechanism assigns adaptive weights to the most informative time steps. To further enhance performance, an improved Kepler optimization algorithm(IKOA) is introduced to optimize the model's hyperparameters. The model is validated using real-world operational data from coal-fired power plant units. In addition to root mean square error(RMSE), the mean absolute error(MAE) and the coefficient of determination(R2) are also computed to comprehensively evaluate prediction performance. Comparative experiments show that among LSTM, BiLSTM, CNN-BiLSTM, and CNN-BiLSTM-Attention, the last model achieves the lowest RMSE of 24.329 2 t/h, compared to 26.261 0 t/h, 26.039 3 t/h, and 25.429 3 t/h for the other three models, respectively. Moreover, when optimizing the CNN-BiLSTM-Attention model with four metaheuristic algorithms(genetic algorithm(GA), whale optimization algorithm(WOA), Kepler optimization algorithm(KOA), and IKOA), the IKOA yields the best fitness value of 21.584 1 t/h, slightly outperforming WOA(21.594 6 t/h), KOA(22.136 0 t/h), and GA(22.807 2 t/h). Results demonstrate that the proposed model exhibits strong transferability across units of similar capacity and effectively captures CO2 emission trends over time, although prediction errors increase during seasonal transitions due to variations in operating conditions and coal quality.[Results] The proposed model consistently outperforms baseline models in both prediction accuracy and stability.[Conclusion] The CNN-BiLSTM-Attention model enables effective CO2 emission forecasting for coalfired power plants even without direct CO2 measurements, showing significant applicability and potential for broader deployment. This capability is particularly valuable for plants lacking continuous emission monitoring systems(CEMS), offering a cost-effective alternative for CO2 estimation. However, long-term or cross-seasonal predictions require further consideration of external factors such as coal quality. This study is primarily validated on limited units from a single plant; thus, its generalizability across different regions, coal sources, and operational strategies warrants further investigation. Future work will incorporate direct coal-quality indicators through multisource data fusion or online updating mechanisms to enhance long-term predictive performance and robustness.
Short-term power load forecasting via a similar-day clustering-TCN-Informer hybrid model
Wu Tao;Hong Mang;Liu Duanyang;Cao Qiang;Wang Ya;Hu Yin;Wang Hao;Yu Wei;Yang Yongjun;Luo Rui;Chen Wu;[Objective] Existing short-term power load forecasting models struggle to simultaneously learn the inherent differences in electricity consumption patterns across diverse scenarios and effectively capture long-range temporal dependencies, so their forecasting accuracy is significantly restricted in high-volatility situations such as extreme weather events, peak load periods, or sudden changes in user behavior. To address this limitation, this paper proposes a novel hybrid forecasting framework named “similar-day clustering-temporal convolutional network(TCN)-Informer”. [Methods] First, the fuzzy C-means(FCM) clustering algorithm is employed to categorize historical load samples into three distinct similar-day clusters(working days, weekends, and holidays) based on historical load data and calendar features(e.g., weekday/weekend flags, holiday labels). This preprocessing step effectively mitigates cross-pattern noise interference caused by heterogeneous consumption behaviors, laying a solid foundation for subsequent feature extraction. Second, for each clustered dataset, a causal dilated TCN is utilized to extract local temporal features and capture short-to medium-term load fluctuations, as its dilated convolution structure enables efficient expansion of the receptive field without increasing computational complexity while avoiding future information leakage. Subsequently, the Informer model equipped with a probabilistic sparse attention mechanism is introduced to capture long-range dependencies in load time series, which significantly reduces the computational burden compared to traditional attention mechanisms and enhances the model's ability to learn long-term temporal correlations. Moreover, exogenous variables closely related to power load, including meteorological factors(temperature, humidity, solar irradiation) and real-time electricity prices, are integrated into the forecasting framework to improve the comprehensiveness of input information. To optimize the model's hyperparameters, Bayesian optimization is adopted to efficiently search the global hyperparameter space, outperforming traditional grid search in both search efficiency and optimization performance. [Results] Experimental validations for 24-hour load forecasting are conducted using 30-minute resolution data from the AEMO-NSW power market, covering a one-year period with over 17 000 samples. The results demonstrate that the proposed model achieves remarkable performance: its root mean square error(RMSE) and mean absolute error(MAE) reduce by 16.59% and 14.65% compared with those of the classic LSTM model, and by 12.33% and 11.60% in contrast to the advanced TimesNet model, with the mean absolute percentage error(MAPE) as low as 3.02%. Particularly in high-volatility scenarios, the model maintains superior stability and accuracy. [Conclusion] These findings fully verify the prominent advantages of the proposed method in forecasting precision and robustness, which can provide reliable decision-making support for power grid dispatching, electricity spot trading, and energy resource optimization in practical power systems.
Contact Us
