作为化工生产过程中的关键设备，换热器的换热效率在实现节能减排目标中发挥着至关重要的作用。然而换热器运行过程中易因结垢导致换热效率下降，影响工艺稳定性和运行成本，甚至引发设备故障，因此其运行状态监测与维护管理已成为化学工程领域的关键问题。在换热器运行状态监测中，污垢因子用于量化换热器的结垢程度，而热负荷则反映其运行效率与能源消耗情况。然而，传统的定期维护与人工监测方式难以满足现代工业对高效、智能化管理的需求。针对这一问题，本文基于集成学习方法，分别构建污垢因子预测和热负荷预测模型，实现换热器运行状态的动态监测。通过提高预测精度和增强模型可解释性，优化设备维护策略，从而提升换热器的运行效率，促进化工生产的可持续发展。

针对现有污垢因子预测方法参数量大且模型不稳定的问题，提出了一种基于Stacking集成学习的污垢因子预测模型。首先，利用K-Nearest Neighbors（KNN）、Random Forest、Bagging、Extreme Gradient Boosting（XGBoost）、Light Gradient Boosting Machine（LightGBM）和Gaussian Process Regression（GPR）构建多个污垢因子预测子模型，并评估其性能。随后，选择表现最优的Random Forest、XGBoost和LightGBM进行集成，并采用线性回归层融合其预测结果，从而构建最终的Stacking模型。实验结果表明，该模型在污垢因子预测方面表现优异，具体性能指标为MSE = 1.26×10⁻⁴，MAE = 4.38×10⁻³，R² = 0.99683，RAE% = 2.39，MAPE% = 15.94，不仅优于现有的Multi-Layer Perceptron（MLP）方法，在性能上也超越了GPR。本研究首次建立了基于集成学习的换热器污垢因子预测基准，并进一步验证了集成学习方法在复杂工程应用场景中的优势与适用性。

针对现有热负荷预测方法在处理复杂非线性数据时精度有限且缺乏可解释性的问题，本文提出了一种新方法——InsightBoost，以提升预测性能和可解释性。InsightBoost由SupertBoost模型与SHAP解释性分析模块组成。SupertBoost集成了六种基础学习器（Random Forest、XGBoost、LightGBM、Gradient Boosting、Adaptive Boosting（AdaBoost）、Categorical Boosting（CatBoost）），并采用线性回归作为元模型融合各基础模型的预测结果，同时利用GridSearchCV进行超参数优化，以提高模型的泛化能力。SHAP值则用于量化各特征对预测结果的影响，提升模型透明度与可解释性。实验结果表明，InsightBoost在预测精度和可解释性方面均优于传统的单一机器学习算法，其关键性能指标为MSE = 1.04×10⁻²，MAE = 7.92×10⁻²，R² = 0.9908，RMSE = 0.1022。该方法有效克服了现有白箱模型适用性受限与机器学习黑箱模型透明性不足的问题，为换热器的优化运行与智能化决策提供了有力支持。

As a critical component in chemical production processes, heat exchangers play a vital role in achieving energy conservation and emission reduction goals through their thermal efficiency. However, during operation, fouling often leads to a decline in heat transfer performance, which can compromise process stability, increase operational costs, and even cause equipment failure. Consequently, the monitoring and maintenance of heat exchanger operating conditions has become a key issue in the field of chemical engineering. In operational monitoring, the fouling factor is commonly used to quantify the extent of fouling, while thermal load reflects the operational efficiency and energy consumption of the heat exchanger. Traditional approaches based on periodic maintenance and manual monitoring, however, fall short of meeting the demands of modern industry for efficient and intelligent management. To address this challenge, this study proposes an ensemble learning-based approach to develop predictive models for both the fouling factor and thermal load, enabling dynamic monitoring of heat exchanger performance. By improving prediction accuracy and enhancing model interpretability, the proposed method aims to optimize maintenance strategies, thereby enhancing heat exchanger efficiency and promoting the sustainable development of chemical production processes.

To tackle the issues of large parameter sizes and instability in existing fouling factor prediction models, a Stacking ensemble learning-based model is proposed. Initially, multiple base models—K-Nearest Neighbors (KNN), Random Forest, Bagging, Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Gaussian Process Regression (GPR)—are constructed and evaluated. The top-performing models (Random Forest, XGBoost, and LightGBM) are selected for integration, and their outputs are combined using a linear regression meta-model to form the final Stacking model. Experimental results demonstrate the model’s superior performance in fouling factor prediction, with metrics of MSE = 1.26×10⁻⁴, MAE = 4.38×10⁻³, R² = 0.99683, RAE% = 2.39, and MAPE% = 15.94. The proposed model outperforms both the traditional Multi-Layer Perceptron (MLP) and GPR models, establishing a benchmark for fouling factor prediction in heat exchangers using ensemble learning, and validating its applicability in complex engineering scenarios.

To address the limitations of existing thermal load prediction methods in handling complex nonlinear data and lacking interpretability, a novel approach—InsightBoost—is proposed to enhance both predictive performance and model transparency. InsightBoost integrates the SupertBoost model with a SHAP-based interpretability module. SupertBoost combines six base learners (Random Forest, XGBoost, LightGBM, Gradient Boosting, Adaptive Boosting (AdaBoost), and Categorical Boosting (CatBoost)) and employs linear regression as a meta-model to aggregate predictions. Hyperparameters are optimized via GridSearchCV to improve generalization. SHAP values are used to quantify the contribution of each feature to the prediction outcome, thereby enhancing model interpretability. Experimental results show that InsightBoost outperforms traditional standalone machine learning algorithms in both accuracy and explainability, with performance metrics of MSE = 1.04×10⁻², MAE = 7.92×10⁻², R² = 0.9908, and RMSE = 0.1022. This approach effectively bridges the gap between the limited applicability of white-box models and the opacity of black-box machine learning models, offering robust support for optimized heat exchanger operation and intelligent decision-making.

[1] 中华人民共和国国务院. 2030年前碳达峰行动方案[R]. 2021年10月.

[2] Gupta S K, Gupta S, Gupta T, et al. A review on recent advances and applications of nanofluids in plate heat exchanger[J]. Materials Today: Proceedings, 2020, 44(10): 166-180.

[3] Kapustenko P, Klemeš J J, Arsenyeva O. Plate heat exchangers fouling mitigation effects in heating of water solutions: A review[J]. Renewable and Sustainable Energy Reviews, 2023, 179: 113283.

[4] Inamdar H V, Groll E A, Weibel J A, et al. Air-side fouling of finned heat exchangers: Part 1, review and proposed test protocol[J]. International Journal of Refrigeration, 2023, 151: 77-86.

[5] Berce J, Zupančič M, Može M, et al. Infrared thermography observations of crystallization fouling in a plate heat exchanger[J]. Applied Thermal Engineering, 2023, 224: 120116.

[6] 周泽楷, 侯宏娟, 孙莉, 等. 基于CNN和BiLSTM神经网络模型的太阳能供暖负荷预测研究[J]. 太阳能学报, 2024, 45(10): 415-422.

[7] Zhao Y, Zhang C, Zhang Y, et al. A review of data mining technologies in building energy systems: Load prediction, pattern identification, fault detection and diagnosis[J]. Energy and Built Environment, 2020, 1(2): 149-164.

[8] Hou G, Zhang D, Yan Q, et al. Application of machine learning algorithms in real-time fouling monitoring of plate heat exchangers[J]. International Communications in Heat and Mass Transfer, 2025, 164: 108809.

[9] Ilyunin O, Bezsonov O, Rudenko S, et al. The neural network approach for estimation of heat transfer coefficient in heat exchangers considering the fouling formation dynamic[J]. Thermal Science and Engineering Progress, 2024, 51: 102615.

[10] Galeazzo F C C, Miura R Y, Gut J A W, et al. Experimental and numerical heat transfer in a plate heat exchanger[J]. Chemical Engineering Science, 2006, 61(21): 7133-7138.

[11] Reynoso-Jardón E, Tlatelpa-Becerro A, Rico-Martínez R, et al. Artificial neural networks (ANN) to predict overall heat transfer coefficient and pressure drop on a simulated heat exchanger[J]. International Journal of Applied Engineering Research, 2019, 14(13): 3097-3103.

[12] Liang Y, Zhu L, Wang Y, et al. Fouling prediction of a heat exchanger based on wavelet neural network optimized by improved particle swarm optimization algorithm[J]. Processes, 2024, 12(11): 2412.

[13] Leidenfrost J G. De aquae communis nonnullis quqlitatibus tractatus[M]. Duisburg: Ovenius, 1756: 1153–1166.

[14] 杨善让，徐志明，孙灵芳. 换热设备污垢与对策[M]. 北京: 科学出版社，2004: 38-42.

[15] 蒋宁, 张元毅, 范伟, 等. 基于智能预测和机理模型的换热网络清洗决策[J]. 化工进展, 2022, 41(4): 1781-1792.

[16] 黄雄虎, 顾敦罡, 陆嘉麒, 等. 污水源热泵技术在城市污水热能回收中的应用现状与研究进展[J]. 应用化工, 2023, 52(3): 922-928.

[17] Melo L. F., Bott T. R., Bernardo C. A. Fouling Science and Technology [M]. Dordrecht: Springer Netherlands, 2012.

[18] Liu X, Zhang X, Lu T, et al. Numerical simulation of sub-cooled boiling flow with fouling deposited inside channels[J]. Applied Thermal Engineering, 2016, 103: 434-442.

[19] Helalizadeh A, Müller-Steinhagen H, Jamialahmadi M. Mixed salt crystallisation fouling[J]. Chemical Engineering and Processing: Process Intensification, 2000, 39(1): 29-43.

[20] Ben-Mansour R, El-Ferik S, Al-Naser M, et al. Experimental/Numerical investigation and prediction of fouling in multiphase flow heat exchangers: A review[J]. Energies, 2023, 16(6): 2812.

[21] Wang J, Xu X, Xu Y, et al. Fouling prediction of heat exchanger surface under alternating magnetic field based on IGWO–SVR[J]. International Journal of Thermal Sciences, 2023, 184: 108018.

[22] Wang J, Sun L, Li H, et al. Prediction model of fouling thickness of heat exchanger based on TA-LSTM structure[J]. Processes, 2023, 11(9): 2594.

[23] Jradi R, Marvillet C, Jeday M R. Multi-objective optimization and performance assessment of response surface methodology (RSM), artificial neural network (ANN) and adaptive neuro-fuzzy interfence system (ANFIS) for estimation of fouling in phosphoric acid/steam heat exchanger[J]. Applied Thermal Engineering, 2024, 248: 123255.

[24] Sundar S, Rajagopal M C, Zhao H, et al. Fouling modeling and prediction approach for heat exchangers using deep learning[J]. International Journal of Heat and Mass Transfer, 2020, 159: 120112.

[25] Tang S Z, Li M J, Wang F L, et al. Fouling potential prediction and multi-objective optimization of a flue gas heat exchanger using neural networks and genetic algorithms[J]. International Journal of Heat and Mass Transfer, 2020, 152: 119488.

[26] Davoudi E, Vaferi B. Applying artificial neural networks for systematic estimation of degree of fouling in heat exchangers[J]. Chemical Engineering Research and Design, 2018, 130: 138-153.

[27] Hosseini S, Khandakar A, Chowdhury M E H, et al. Novel and robust machine learning approach for estimating the fouling factor in heat exchangers[J]. Energy Reports, 2022, 8: 8767-8776.

[28] Wang Y, Li Z, Liu J, et al. A novel combined model for heat load prediction in district heating systems[J]. Applied Thermal Engineering, 2023, 227: 120372.

[29] Zhu H, Cheng X, Liu X, et al. Advancing regional heat load forecasting through sophisticated data-driven methodologies integrated with robust adversarial training strategies[J]. Journal of Building Engineering, 2025, 112101.

[30] Nellis G F. A heat exchanger model that includes axial conduction, parasitic heat loads, and property variations[J]. Cryogenics, 2003, 43(9): 523-538.

[31] Gong M, Zhao Y, Sun J, et al. Load forecasting of district heating system based on Informer[J]. Energy, 2022, 253: 124179.

[32] Chung W H, Gu Y H, Yoo S J. District heater load forecasting based on machine learning and parallel CNN-LSTM attention[J]. Energy, 2022, 246: 123350.

[33] Wang Y, Zhan C, Li G, et al. Physics-guided LSTM model for heat load prediction of buildings[J]. Energy and Buildings, 2023, 294: 113169.

[34] Harb H, Boyanov N, Hernandez L, et al. Development and validation of grey-box models for forecasting the thermal response of occupied buildings[J]. Energy and Buildings, 2016, 117: 199-207.

[35] Thulukkanam K. Heat exchanger design handbook[M]. Boca Raton: CRC Press, 2013.

[36] 魏欣宇, 方松, 滕钧杰, 等. 连续转化式低温氢气换热器换热与催化匹配特性研究[J]. 工程热物理学报, 2025, 46(3): 703-713.

[37] Sadeghianjahromi A, Wang C C. Heat transfer enhancement in fin-and-tube heat exchangers–A review on different mechanisms[J]. Renewable and Sustainable Energy Reviews, 2021, 137: 110470.

[38] Hesselgreaves J E, Law R, Reay D. Compact heat exchangers: Selection, design and operation[M]. Oxford: Butterworth-Heinemann, 2016.

[39] Shah R K, Sekulic D P. Fundamentals of heat exchanger design[M]. Hoboken: John Wiley & Sons, 2003.

[40] Flynn A M, Akashige T, Theodore L. Kern's process heat transfer[M]. Hoboken: John Wiley & Sons, 2019.

[41] Kakaç S, Liu H, Pramuanjaroenkij A. Heat exchangers: Selection, rating, and thermal design[M]. CRC press, 2002.

[42] Jalili B, Aghaee N, Jalili P, et al. Novel usage of the curved rectangular fin on the heat transfer of a double-pipe heat exchanger with a nanofluid[J]. Case Studies in Thermal Engineering, 2022, 35: 102086.

[43] Xia G, Ma D, Zhai Y, et al. Experimental and numerical study of fluid flow and heat transfer characteristics in microchannel heat sink with complex structure[J]. Energy Conversion and Management, 2015, 105: 848-857.

[44] Kalinin E K, Dreitser G A. Heat transfer enhancement in heat exchangers[J]. Advances in heat transfer. Elsevier, 1998, 31: 159-332.

[45] Jang J Y, Chen L K. Numerical analysis of heat transfer and fluid flow in a three-dimensional wavy-fin and tube heat exchanger[J]. International Journal of Heat and Mass Transfer, 1997, 40(16): 3981-3990.

[46] Pham D C, Mercère G, Ouvrard R, et al. Fouling detection in a parallel flow heat exchanger via a Roesser model identification procedure[J]. IFAC-PapersOnLine, 2017, 50(1): 12866-12871.

[47] Gautam R K, Parmar K N S, Vyas B G. Effect of fouling on thermal and hydraulic parameter of Shell and Tube Heat exchanger[C]. Proc. Student Conf. 2017.

[48] Jradi R, Marvillet C, Jeday M R. Estimation and sensitivity analysis of fouling resistance in phosphoric acid/steam heat exchanger using artificial neural networks and regression methods[J]. Scientific Reports, 2023, 13(1): 17889.

[49] Godasiaei S H. Exploring the influence of crystallization fouling on microscale heat exchangers through machine learning analysis[J]. Numerical Heat Transfer, Part A: Applications, 2024, 1-27.

[50] Xue G, Qi C, Li H, et al. Heating load prediction based on attention long short term memory: A case study of Xingtai[J]. Energy, 2020, 203: 117846.

[51] Zhao Z, Alzubaidi L, Zhang J, et al. A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations[J]. Expert Systems with Applications, 2024, 242: 122807.

[52] Rani V, Nabi S T, Kumar M, et al. Self-supervised learning: A succinct review[J]. Archives of Computational Methods in Engineering, 2023, 30(4): 2761-2775.

[53] Tarmanini C, Sarma N, Gezegin C, et al. Short term load forecasting based on ARIMA and ANN approaches[J]. Energy Reports, 2023, 9: 550-557.

[54] Jagtap A D, Karniadakis G E. How important are activation functions in regression and classification? A survey, performance comparison, and future directions[J]. Journal of Machine Learning for Modeling and Computing, 2023, 4(1).

[55] Dubey S R, Singh S K, Chaudhuri B B. Activation functions in deep learning: A comprehensive survey and benchmark[J]. Neurocomputing, 2022, 503: 92-108.

[56] Moller M. Efficient training of feed-forward neural networks[M]. Boca Raton: CRC Press, 2024: 136-173.

[57] Tian Y, Zhang Y. A comprehensive survey on regularization strategies in machine learning[J]. Information Fusion, 2022, 80: 146-166.

[58] Salehin I, Kang D K. A review on dropout regularization approaches for deep neural networks within the scholarly domain[J]. Electronics, 2023, 12(14): 3106.

[59] Hu T, Wang W, Lin C, et al. Regularization matters: A nonparametric perspective on overparametrized neural network[C]. International Conference on Artificial Intelligence and Statistics. PMLR, 2021, 829-837.

[60] Goh G, Cammarata N, Voss C, et al. Multimodal neurons in artificial neural networks[J]. Distill, 2021, 6(3): e30.

[61] Madhiarasan M, Deepa S N. Comparative analysis on hidden neurons estimation in multi layer perceptron neural networks for wind speed forecasting[J]. Artificial Intelligence Review, 2017, 48: 449-471.

[62] Ahmad M W, Mourshed M, Rezgui Y. Trees vs neurons: Comparison between random forest and ANN for high-resolution prediction of building energy consumption[J]. Energy and Buildings, 2017, 147: 77-89.

[63] Sheela K G, Deepa S N. Review on methods to fix number of hidden neurons in neural networks[J]. Mathematical Problems in Engineering, 2013, 2013(1): 425740.

[64] Wright L G, Onodera T, Stein M M, et al. Deep physical neural networks trained with backpropagation[J]. Nature, 2022, 601(7894): 549-555.

[65] Lee C, Sarwar S S, Panda P, et al. Enabling spike-based backpropagation for training deep neural network architectures[J]. Frontiers in Neuroscience, 2020, 14: 497482.

[66] Lillicrap T P, Santoro A, Marris L, et al. Backpropagation and the brain[J]. Nature Reviews Neuroscience, 2020, 21(6): 335-346.

[67] Sekhar C, Meghana P S. A study on backpropagation in artificial neural networks[J]. Asia-Pacific Journal of Neural Networks and Its Applications, 2020, 4(1): 21-28.

[68] Romanowicz T, Taler J, Jaremkiewicz M, et al. Determination of heat transfer correlations for fluids flowing through plate heat exchangers needed for online monitoring of district heat exchanger fouling[J]. Energies, 2023, 16(17): 6264.

[69] Sharma S, Sharma S, Athaiya A. Activation functions in neural networks[J]. Towards Data Sci, 2017, 6(12): 310-316.

[70] Mhaskar H N, Micchelli C A. How to choose an activation function[J]. Advances in neural information processing systems, 1993, 6.

[71] Han J, Moraga C. The influence of the sigmoid function parameters on the speed of backpropagation learning[C]. International workshop on artificial neural networks. Berlin, Heidelberg: Springer Berlin Heidelberg, 1995, 195-201.

[72] Alhojilan Y, Ahmed H M. Novel analytical solutions of stochastic Ginzburg-Landau equation driven by Wiener process via the improved modified extended tanh function method[J]. Alexandria Engineering Journal, 2023, 72: 269-274.

[73] Bai Y. Relu-function and derived function review[C]. SHS web of conferences. EDP Sciences, 2022, 144: 02006.

[74] Mienye I D, Sun Y. A survey of ensemble learning: Concepts, algorithms, applications, and prospects[J]. Ieee Access, 2022, 10: 99129-99149.

[75] Yang Y, Lv H, Chen N. A survey on ensemble learning under the era of deep learning[J]. Artificial Intelligence Review, 2023, 56(6): 5545-5589.

[76] Khan A A, Chaudhari O, Chandra R. A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation[J]. Expert Systems with Applications, 2024, 244: 122778.

[77] Sammil S, Sridharan M. Employing ensemble machine learning techniques for predicting the thermohydraulic performance of double pipe heat exchanger with and without turbulators[J]. Thermal Science and Engineering Progress, 2024, 47: 102337.

[78] Chandran V V, Adepu R. Reduced order modeling of a heat exchanger with a stacking ensemble to reduce computational inefficiencies[C]. 2022 IEEE International Symposium on Systems Engineering (ISSE). IEEE, 2022, 1-5.

[79] Ali Z A, Abduljabbar Z H, Tahir H A, et al. Extreme gradient boosting algorithm with machine learning: A review[J]. Academic Journal of Nawroz University, 2023, 12(2): 320-334.

[80] Ngo G, Beard R, Chandra R. Evolutionary bagging for ensemble learning[J]. Neurocomputing, 2022, 510: 1-14.

[81] Prasad B S V. The performance prediction of multistream plate-fin heat exchangers based on stacking pattern[J]. Heat Transfer Engineering, 1991, 12(4): 58-70.

[82] Rainio O, Teuho J, Klén R. Evaluation metrics and statistical tests for machine learning[J]. Scientific Reports, 2024, 14(1): 6086.

[83] Kumari S A, Srinivasan S. Ash fouling monitoring and soot-blow optimization for reheater in thermal power plant[J]. Applied Thermal Engineering, 2019, 149: 62-72.

[84] 张晗筱, 王瑞琪, 张亚婷. 基于CNN-LSTM的换热器污垢因子预测研究[J]. 化工学报, 2025, 1-14.

[85]Ridzuan F, Zainon W M N W. A review on data cleansing methods for big data[J]. Procedia Computer Science, 2019, 161: 731-738.

[86] Alshaher H. Studying the effects of feature scaling in machine learning[D]. North Carolina Agricultural and Technical State University, 2021.

[87] Madhu P K R, Subbaiah J, Krithivasan K. RF-LSTM-based method for prediction and diagnosis of fouling in heat exchanger[J]. Asia-Pacific Journal of Chemical Engineering, 2021, 16(5): e2684.

[88] Zhang M L, Zhou Z H. ML-KNN: A lazy learning approach to multi-label learning[J]. Pattern Recognition, 2007, 40(7): 2038-2048.

[89] Rigatti S J. Random forest[J]. Journal of Insurance Medicine, 2017, 47(1): 31-39.

[90] Altman N, Krzywinski M. Ensemble methods: Bagging and random forests[J]. Nature Methods, 2017, 14(10): 933-935.

[91] Zhou Z H. Ensemble methods: foundations and algorithms[M]. CRC press, 2025.

[92] Chen T, Guestrin C. Xgboost: A scalable tree boosting system[C]. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016: 785-794.

[93] Schulz E, Speekenbrink M, Krause A. A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions[J]. Journal of Mathematical Psychology, 2018, 85: 1-16.

[94] Ke G, Meng Q, Finley T, et al. Lightgbm: A highly efficient gradient boosting decision tree[J]. Advances in Neural Information Processing Systems, 2017, 30.

[95] Durgam S, Bhosale A, Bhosale V, et al. Temperature prediction of heat sources using machine learning techniques[J]. Heat Transfer, 2021, 50(8): 7817-7838.

[96] Guo Y, Wang J, Chen H, et al. Machine learning-based thermal response time ahead energy demand prediction for building heating systems[J]. Applied Energy, 2018, 221: 16-27.

[97] Guo X, Gao Y, Zheng D, et al. Study on short-term photovoltaic power prediction model based on the stacking ensemble learning[J]. Energy Reports, 2020, 6: 1424-1431.

[98] 盛科荣, 李晓瑞, 吴石磊, 等. 中国城市多层网络嵌入对能源利用效率的影响[J]. 地理科学, 2025, 45(03): 495-505.

[99] 江荣霞, 谢秀娟, 邓笔财, 等. 独立参数对氦低温制冷系统循环效率的影响[J]. 低温工程, 2016, (06): 24-28.

[100] Hughes M T, Kini G, Garimella S. Status, challenges, and potential for machine learning in understanding and applying heat transfer phenomena[J]. Journal of Heat Transfer, 2021, 143(12): 120802.

[101] Zhu J, Han Z, Rao M, et al. Identification of heat load loops and downstream paths in heat exchanger networks[J]. The Canadian Journal of Chemical Engineering, 1996, 74(6): 876-882.

[102] Cui Xuyang, Zhu Junda, Jia Lifu, et al. A novel heat load prediction model of district heating system based on hybrid whale optimization algorithm (WOA) and CNN-LSTM with attention mechanism[J]. Energy, 2024, 312: 133536.

[103] Thilker C A, Bacher P, Bergsteinsson H G, et al. Non-linear grey-box modelling for heat dynamics of buildings[J]. Energy and Buildings, 2021, 252: 111457.

[104] Khan O, Parvez M, Seraj M, et al. Optimising building heat load prediction using advanced control strategies and Artificial Intelligence for HVAC system[J]. Thermal Science and Engineering Progress, 2024, 49: 102484.

[105] Neubauer A, Brandt S, Kriegel M. Relationship between feature importance and building characteristics for heating load predictions[J]. Applied Energy, 2024, 359: 122668.

[106] Morteza A, Yahyaeian A A, Mirzaeibonehkhater M, et al. Deep learning hyperparameter optimization: Application to electricity and heat demand prediction for buildings[J]. Energy and Buildings, 2023, 289: 113036.

[107] Ding Y, Zhu H, Chen R, et al. An efficient AdaBoost algorithm with the multiple thresholds classification[J]. Applied Sciences, 2022, 12(12): 5872.

[108] Huang X, Li Z, Jin Y, et al. Fair-AdaBoost: Extending AdaBoost method to achieve fair classification[J]. Expert Systems with Applications, 2022, 202: 117240.

[109] Zhang L, Jánošík D. Enhanced short-term load forecasting with hybrid machine learning models: CatBoost and XGBoost approaches[J]. Expert Systems with Applications, 2024, 241: 122686.

[110] Zhang Yu, Chen Guangshu. A building heat load prediction method driven by a multi-component fusion LSTM ridge regression ensemble model[J]. Applied Sciences, 2024, 14(9): 3810.

[111] Chaganti R, Rustam F, Daghriri T, et al. Building heating and cooling load prediction using ensemble machine learning model[J]. Sensors, 2022, 22(19): 7692.

[112] Herodotou H, Chen Y, Lu J. A survey on automatic parameter tuning for big data processing systems[J]. ACM Computing Surveys (CSUR), 2020, 53(2): 1-37.

[113] Ding N, Qin Y, Yang G, et al. Parameter-efficient fine-tuning of large-scale pre-trained language models[J]. Nature Machine Intelligence, 2023, 5(3): 220-235.

[114] Sun Y, Ding S, Zhang Z, et al. An improved grid search algorithm to optimize SVR for prediction[J]. Soft Computing, 2021, 25: 5633-5644.

[115] Chicco D, Warrens M J, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation[J]. PeerJ Computer Science, 2021, 7: e623.

[116] Zhou M, Wang L, Hu F, et al. ISSA-LSTM: A new data-driven method of heat load forecasting for building air conditioning[J]. Energy and Buildings, 2024, 114698.

[117] Bergsteinsson H G, Møller Ja K, Nystrup P, et al. Heat load forecasting using adaptive temporal hierarchies[J]. Applied Energy, 2021, 292: 116872.

[118] Zhu G, Wen T, Zhang D. Machine learning-based approach for the prediction of flow boiling/condensation heat transfer performance in mini channels with serrated fins[J]. International Journal of Heat and Mass Transfer, 2021, 166: 120783.

[119] Gao Z, Yu J, Zhao A, et al. A hybrid method of cooling load forecasting for large commercial building based on extreme learning machine[J]. Energy, 2022, 238: 122073.

[120] Dang L M, Shin J, Li Y, et al. Toward explainable heat load patterns prediction for district heating[J]. Scientific Reports, 2023, 13(1): 7434.

[121] Cakiroglu C, Aydın Y, Bekdaş G, et al. Cooling load prediction of a double-story terrace house using ensemble learning techniques and genetic programming with SHAP approach[J]. Energy and Buildings, 2024, 313: 114254.

[122] Singh N K, Nagahara M. LightGBM-, SHAP-, and Correlation-Matrix-Heatmap-Based Approaches for Analyzing Household Energy Data: Towards Electricity Self-Sufficient Houses[J]. Energies, 2024, 17(17): 4518.