{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T17:47:19Z","timestamp":1782755239556,"version":"3.54.5"},"reference-count":40,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T00:00:00Z","timestamp":1782691200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    The accuracy of power load forecasting directly affects the stable operation of the power system and energy optimization configuration. In response to the complex nonlinear characteristics of load data and the tendency of existing methods to fall into local optima, this study selects time-series convolutional networks, Long Short-Term Memory (LSTM) networks, and bidirectional LSTM to propose a combined deep learning model through weighted reconstruction. On this basis, the parrot optimization algorithm is introduced for iteration and strategy optimization, and joint error correction is performed to propose a new power load forecasting model. The experimental results show that under the same conditions and with the same hyperparameters, the conventional gradient descent optimization method takes about 330 rounds to reach 0.85 and the highest is only about 0.89. While the proposed method reaches 0.85 in about 200 rounds and eventually approaches 0.95. The overall convergence is faster and the accuracy is higher. On the UK-NGED dataset, the long-term prediction mean absolute errors of Transformer-DGNN, EEMD-Attention and VMD-ResNet are 0.214, 0.238 and 0.257 respectively, and the research method is 0.181. Simultaneously, continuous learning-based parrot optimization and error correction model suppress weight drift and short-term bias during optimization, maintaining prediction stability under concept drift and non-steady load patterns. When prediction performance degrades, error-monitoring-based lightweight retraining rapidly restores model accuracy. Under extreme weather conditions, the root mean square error of the baseline method is 0.295\u20130.318, and that of the research method is 0.250. To confirm that the differences are not due to random fluctuations, a two-sample t-test is conducted on the root mean square error distributions from 10 independent experiments. After explicitly calculating the mean difference, variance, and degrees of freedom, the result yields\n                    <jats:italic>p<\/jats:italic>\n                    \u2009&lt;\u20090.05, indicating that the reduction in model error under extreme meteorological conditions is statistically significant. Therefore, the model can maintain high accuracy prediction in variable load environments and has stronger adaptability in high penetration rates of new energy and extreme weather conditions. This study provides more accurate and reliable decision support for smart grid scheduling and optimization.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0351428","type":"journal-article","created":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T17:24:34Z","timestamp":1782753874000},"page":"e0351428","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Power load forecasting combining deep learning models and improved CLPO algorithm"],"prefix":"10.1371","volume":"21","author":[{"given":"Jianguang","family":"Song","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiang","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhipeng","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xia","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guoheng","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-0807-7086","authenticated-orcid":true,"given":"Lu","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,29]]},"reference":[{"issue":"3","key":"pone.0351428.ref001","doi-asserted-by":"crossref","first-page":"1480","DOI":"10.3390\/en16031480","article-title":"Load forecasting techniques and their applications in smart grids","volume":"16","author":"H Habbak","year":"2023","journal-title":"Energies"},{"issue":"4","key":"pone.0351428.ref002","doi-asserted-by":"crossref","first-page":"2941","DOI":"10.3390\/su15042941","article-title":"Short term power load forecasting based on PSVMD-CGA model","volume":"15","author":"J Su","year":"2023","journal-title":"Sustainability"},{"issue":"5","key":"pone.0351428.ref003","doi-asserted-by":"crossref","first-page":"2283","DOI":"10.3390\/en16052283","article-title":"Electrical load forecasting using LSTM, GRU, and RNN algorithms","volume":"16","author":"M Abumohsen","year":"2023","journal-title":"Energies"},{"issue":"4","key":"pone.0351428.ref004","doi-asserted-by":"crossref","first-page":"1925","DOI":"10.3390\/en16041925","article-title":"Two-stage short-term power load forecasting based on RFECV feature selection algorithm and a TCN\u2013ECA\u2013LSTM neural network","volume":"16","author":"H Liang","year":"2023","journal-title":"Energies"},{"issue":"4","key":"pone.0351428.ref005","doi-asserted-by":"crossref","first-page":"6130","DOI":"10.1109\/TII.2023.3341256","article-title":"Nonvolatile CMOS memristor, reconfigurable array, and its application in power load forecasting","volume":"20","author":"Q Deng","year":"2024","journal-title":"IEEE Trans Ind Inf"},{"issue":"4","key":"pone.0351428.ref006","doi-asserted-by":"crossref","first-page":"625","DOI":"10.1007\/s12273-023-1087-0","article-title":"Enhancing source domain availability through data and feature transfer learning for building power load forecasting","volume":"17","author":"F Qian","year":"2024","journal-title":"Build Simul"},{"issue":"1","key":"pone.0351428.ref007","doi-asserted-by":"crossref","first-page":"1932","DOI":"10.1109\/TPWRS.2023.3271325","article-title":"A novel sequence to sequence data modelling based CNN-LSTM algorithm for three years ahead monthly peak load forecasting","volume":"39","author":"O Rubasinghe","year":"2024","journal-title":"IEEE Trans Power Syst"},{"issue":"6","key":"pone.0351428.ref008","doi-asserted-by":"crossref","first-page":"3202","DOI":"10.3390\/s23063202","article-title":"Optimized EWT-Seq2Seq-LSTM with attention mechanism to insulators fault prediction","volume":"23","author":"ACR Klaar","year":"2023","journal-title":"Sensors (Basel)"},{"issue":"3","key":"pone.0351428.ref009","doi-asserted-by":"crossref","first-page":"321","DOI":"10.3390\/biomimetics8030321","article-title":"A novel bio-inspired optimization algorithm design for wind power engineering applications time-series forecasting","volume":"8","author":"FK Karim","year":"2023","journal-title":"Biomimetics (Basel)"},{"issue":"10","key":"pone.0351428.ref010","doi-asserted-by":"crossref","first-page":"10145","DOI":"10.1109\/TII.2022.3230726","article-title":"Optimized forecasting model to improve the accuracy of very short-term wind power prediction","volume":"19","author":"MA Hossain","year":"2023","journal-title":"IEEE Trans Ind Inf"},{"issue":"14","key":"pone.0351428.ref011","doi-asserted-by":"crossref","first-page":"10757","DOI":"10.3390\/su151410757","article-title":"A review of modern wind power generation forecasting technologies","volume":"15","author":"W-C Tsai","year":"2023","journal-title":"Sustainability"},{"issue":"5","key":"pone.0351428.ref012","doi-asserted-by":"crossref","first-page":"237","DOI":"10.3390\/systems11050237","article-title":"Development of a hybrid support vector machine with grey wolf optimization algorithm for detection of the solar power plants anomalies","volume":"11","author":"QI Ahmed","year":"2023","journal-title":"Systems"},{"issue":"1","key":"pone.0351428.ref013","doi-asserted-by":"crossref","first-page":"213","DOI":"10.3390\/forecast5010012","article-title":"A day-ahead photovoltaic power prediction via transfer learning and deep neural networks","volume":"5","author":"SM Miraftabzadeh","year":"2023","journal-title":"Forecasting"},{"issue":"11","key":"pone.0351428.ref014","doi-asserted-by":"crossref","first-page":"2559","DOI":"10.3390\/en17112559","article-title":"An Improved WOA (Whale Optimization Algorithm)-Based CNN-BIGRU-CBAM Model and its application to short-term power load forecasting","volume":"17","author":"L Dai","year":"2024","journal-title":"Energies"},{"issue":"14","key":"pone.0351428.ref015","doi-asserted-by":"crossref","first-page":"5966","DOI":"10.3390\/app14145966","article-title":"Short-term electricity load forecasting based on improved data decomposition and hybrid deep-learning models","volume":"14","author":"J Chen","year":"2024","journal-title":"Applied Sciences"},{"issue":"3","key":"pone.0351428.ref016","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0300229","article-title":"Short-term power load forecasting method based on Bagging-stochastic configuration networks","volume":"19","author":"X Pang","year":"2024","journal-title":"PLoS One"},{"issue":"7","key":"pone.0351428.ref017","doi-asserted-by":"crossref","first-page":"9939","DOI":"10.1007\/s11042-022-13462-2","article-title":"A load forecasting model based on support vector regression with whale optimization algorithm","volume":"82","author":"Y Lu","year":"2022","journal-title":"Multimed Tools Appl"},{"issue":"14","key":"pone.0351428.ref018","doi-asserted-by":"crossref","first-page":"11123","DOI":"10.3390\/su151411123","article-title":"Short-term power load forecasting in three stages based on CEEMDAN-TGA model","volume":"15","author":"Y Hong","year":"2023","journal-title":"Sustainability"},{"issue":"4","key":"pone.0351428.ref019","doi-asserted-by":"crossref","first-page":"4249","DOI":"10.1007\/s40747-022-00952-x","article-title":"A hybrid temporal convolutional network and Prophet model for power load forecasting","volume":"9","author":"J Mo","year":"2023","journal-title":"Complex Intell Syst"},{"issue":"12","key":"pone.0351428.ref020","doi-asserted-by":"crossref","first-page":"5436","DOI":"10.3390\/s23125436","article-title":"A machine learning model ensemble for mixed power load forecasting across multiple time horizons","volume":"23","author":"N Giamarelos","year":"2023","journal-title":"Sensors"},{"issue":"3","key":"pone.0351428.ref021","doi-asserted-by":"crossref","first-page":"963","DOI":"10.1007\/s40684-023-00537-0","article-title":"AI-empowered methods for smart energy consumption: A review of load forecasting, anomaly detection and demand response","volume":"11","author":"X Wang","year":"2023","journal-title":"Int J of Precis Eng and Manuf-Green Tech"},{"issue":"7","key":"pone.0351428.ref022","doi-asserted-by":"crossref","first-page":"2444","DOI":"10.1002\/ese3.1465","article-title":"Short\u2010term load forecasting based on a generalized regression neural network optimized by an improved sparrow search algorithm using the empirical wavelet decomposition method","volume":"11","author":"G Fan","year":"2023","journal-title":"Energy Science & Engineering"},{"issue":"4","key":"pone.0351428.ref023","first-page":"5714","article-title":"A quantum controlled-not neural network-based load forecast and management model for smart grid","volume":"17","author":"J Kumar","year":"2023","journal-title":"IEEE Syst J"},{"issue":"10","key":"pone.0351428.ref024","doi-asserted-by":"crossref","first-page":"2175","DOI":"10.3390\/electronics12102175","article-title":"Long-term power load forecasting using LSTM-informer with ensemble learning","volume":"12","author":"K Wang","year":"2023","journal-title":"Electronics"},{"issue":"3","key":"pone.0351428.ref025","doi-asserted-by":"crossref","first-page":"1434","DOI":"10.3390\/en16031434","article-title":"On the benefits of using metaheuristics in the hyperparameter tuning of deep learning models for energy load forecasting","volume":"16","author":"N Bacanin","year":"2023","journal-title":"Energies"},{"issue":"17","key":"pone.0351428.ref026","doi-asserted-by":"crossref","first-page":"6676","DOI":"10.1002\/cpe.6676","article-title":"Short-term load forecasting based on CNN-BiLSTM with Bayesian optimization and attention mechanism","volume":"35","author":"H Shi","year":"2023","journal-title":"Concurr Comput Pract Exp"},{"issue":"8","key":"pone.0351428.ref027","doi-asserted-by":"crossref","first-page":"10833","DOI":"10.1007\/s12652-022-04355-1","article-title":"A fuzzy adaptive symbiotic organism search based hybrid wavelet transform-extreme learning machine model for load forecasting of power system: A case study","volume":"14","author":"JR Nayak","year":"2022","journal-title":"J Ambient Intell Human Comput"},{"issue":"1","key":"pone.0351428.ref028","doi-asserted-by":"crossref","first-page":"1783","DOI":"10.32604\/cmc.2023.035736","article-title":"A Levenberg-Marquardt based neural network for short-term load forecasting","author":"S Ali","year":"2023","journal-title":"Comput. Mater. Continua"},{"issue":"6","key":"pone.0351428.ref029","first-page":"889","article-title":"Improved deep learning model for forecasting short-term load based on random forest algorithm and rough set theory","volume":"44","author":"F Yu","year":"2023","journal-title":"Power Gener Technol"},{"issue":"4","key":"pone.0351428.ref030","doi-asserted-by":"crossref","first-page":"421","DOI":"10.26599\/BDMA.2022.9020043","article-title":"Personalized federated learning for heterogeneous residential load forecasting","volume":"6","author":"C Guan","year":"2023","journal-title":"Big Data Min Anal"},{"key":"pone.0351428.ref031","first-page":"1","article-title":"Short-term load forecasting system based on sliding fuzzy granulation and equilibrium optimizer","author":"S Li","year":"2023","journal-title":"Appl Intell (Dordr)"},{"issue":"15","key":"pone.0351428.ref032","doi-asserted-by":"crossref","first-page":"11667","DOI":"10.3390\/su151511667","article-title":"Research on peak load prediction of distribution network lines based on prophet-LSTM model","volume":"15","author":"Z Chen","year":"2023","journal-title":"Sustainability"},{"issue":"1","key":"pone.0351428.ref033","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1007\/s40866-023-00188-9","article-title":"Multi-term electrical load forecasting of smart cities using a new hybrid highly accurate neural network-based predictive model","volume":"9","author":"A Safari","year":"2023","journal-title":"Smart Grids Sustain Energ"},{"issue":"2","key":"pone.0351428.ref034","first-page":"119","article-title":"Region-based convolutional neural network for segmenting text in epigraphical images","volume":"1","author":"P Preethi","year":"2023","journal-title":"Artif Intell Appl"},{"issue":"16","key":"pone.0351428.ref035","doi-asserted-by":"crossref","first-page":"6069","DOI":"10.3390\/en16166069","article-title":"The use of real energy consumption data in characterising residential energy demand with an inventory of UK datasets","volume":"16","author":"L Thomson","year":"2023","journal-title":"Energies"},{"issue":"1","key":"pone.0351428.ref036","doi-asserted-by":"crossref","first-page":"686","DOI":"10.1038\/s41597-023-02542-z","article-title":"ELMAS: A one-year dataset of hourly electrical load profiles from 424 French industrial and tertiary sectors","volume":"10","author":"K Bellinguer","year":"2023","journal-title":"Sci Data"},{"issue":"4","key":"pone.0351428.ref037","doi-asserted-by":"crossref","first-page":"3929","DOI":"10.1007\/s12652-022-04462-z","article-title":"Improved invasive weed bird swarm optimization algorithm (IWBSOA) enabled hybrid deep learning classifier for diabetic prediction","volume":"14","author":"CN Aher","year":"2022","journal-title":"J Ambient Intell Human Comput"},{"issue":"12","key":"pone.0351428.ref038","first-page":"217","article-title":"New graph-based and transformer deep learning models for river dissolved oxygen forecasting","volume":"10","author":"PAC Rocha","year":"2023","journal-title":"Environ"},{"issue":"1","key":"pone.0351428.ref039","doi-asserted-by":"crossref","first-page":"12028","DOI":"10.1088\/1742-6596\/2637\/1\/012028","article-title":"Crop water requirement prediction method based on EEMD-attention-LSTM mode","volume":"2637","author":"Y Ma","year":"2023","journal-title":"J Phys Conf Ser"},{"issue":"3","key":"pone.0351428.ref040","doi-asserted-by":"crossref","first-page":"36209","DOI":"10.1088\/1361-6501\/adb2b0","article-title":"Fault diagnosis of planetary gears in noisy environments using a VMTransformer model","volume":"36","author":"D Wang","year":"2025","journal-title":"Meas Sci Technol"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351428","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T17:24:43Z","timestamp":1782753883000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351428"}},"subtitle":[],"editor":[{"given":"Zeyar","family":"Aung","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,29]]},"references-count":40,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,29]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0351428","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,29]]}}}