{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T17:53:08Z","timestamp":1791913988518,"version":"4.3.4"},"reference-count":34,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T00:00:00Z","timestamp":1791849600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Major National Science and Technology Special Project for New Oil and Gas Exploration and Development in China","award":["2025ZD1408804"],"award-info":[{"award-number":["2025ZD1408804"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>To address the issue of data distortion in traditional well log density reconstruction methods caused by harsh borehole environments in complex heterogeneous formations, and the challenge for general machine learning models to capture high-frequency abrupt change features due to the lack of geophysical constraints, this paper proposes a density curve reconstruction framework integrating geologically-inspired feature engineering and a mutation-aware mechanism. The method first constructs a multi-dimensional enhanced feature space covering sedimentary cycle characteristics and fluid properties through multi-scale geologically-inspired analysis, utilizing robust range normalization to eliminate interference from outliers. Subsequently, a mutation-aware driven algorithm based on the gradient change rate is innovatively designed. By dynamically adjusting loss weights, it resolves the drawback where traditional smoothing algorithms tend to erase formation interface information. Finally, geological prior constraint terms are embedded into the XGBoost architecture to achieve multi-objective optimization that balances data-driven accuracy with consistency in geological laws. Under leave-one-well-out cross-validation over ten wells, the proposed method achieves a mean R\u00b2 of 0.98, an RMSE of 0.029, and an MAE of 0.022. Experimental results demonstrate that the proposed method not only improves reconstruction accuracy but also significantly enhances the physical consistency of the results. Furthermore, it precisely identifies thin interbedded boundaries, providing highly reliable basic data support for the refined evaluation of oil and gas reservoirs.<\/jats:p>","DOI":"10.1371\/journal.pone.0347378","type":"journal-article","created":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T17:28:42Z","timestamp":1791912522000},"page":"e0347378","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Geologically-inspired and mutation-aware driven well log density reconstruction method for complex heterogeneous formations"],"prefix":"10.1371","volume":"21","author":[{"given":"Yang","family":"Li","sequence":"first","affiliation":[{"name":"Sinopec","place":["Puyang, China"],"department":["Zhongyuan Oilfield Informatization Management Center Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haiwei","family":"Guo","sequence":"additional","affiliation":[{"name":"Sinopec","place":["Puyang, China"],"department":["Zhongyuan Oilfield Informatization Management Center Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7080-1098","authenticated-orcid":true,"given":"Xianfeng","family":"Gong","sequence":"additional","affiliation":[{"name":"Zhongyuan Oilfield Company, Sinopec","place":["Puyang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ping","family":"Liu","sequence":"additional","affiliation":[{"name":"Sinopec","place":["Puyang, China"],"department":["Zhongyuan Oilfield Informatization Management Center Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ning","family":"Hao","sequence":"additional","affiliation":[{"name":"Sinopec","place":["Puyang, China"],"department":["Zhongyuan Oilfield Informatization Management Center Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuping","family":"Meng","sequence":"additional","affiliation":[{"name":"Sinopec","place":["Puyang, China"],"department":["Zhongyuan Oilfield Informatization Management Center Department"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,13]]},"reference":[{"issue":"5","key":"pone.0347378.ref001","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0324793","article-title":"Analyzing the impact of clay minerals on the reservoir quality of the Lower Goru Formation using Unsupervised Machine Learning","volume":"20","author":"K Noreen","year":"2025","journal-title":"PLoS One"},{"issue":"4","key":"pone.0347378.ref002","doi-asserted-by":"crossref","first-page":"1282","DOI":"10.3390\/pr11041282","article-title":"Saturation Determination and fluid identification in carbonate rocks based on well logging data: A middle eastern case study","volume":"11","author":"J Guo","year":"2023","journal-title":"Processes"},{"issue":"3","key":"pone.0347378.ref003","doi-asserted-by":"crossref","first-page":"1047","DOI":"10.1007\/s11053-023-10184-6","article-title":"A novel machine learning approach for detecting outliers, rebuilding well logs, and enhancing reservoir characterization","volume":"32","author":"M Ali","year":"2023","journal-title":"Nat Resour Res"},{"key":"pone.0347378.ref004","doi-asserted-by":"crossref","first-page":"105789","DOI":"10.1016\/j.cageo.2024.105789","article-title":"Imputation in well log data: A benchmark for machine learning methods","volume":"196","author":"PHT Gama","year":"2025","journal-title":"Computers & Geosciences"},{"issue":"3","key":"pone.0347378.ref005","doi-asserted-by":"crossref","DOI":"10.1190\/geo2011-0454.1","article-title":"Improved estimation of mineral and fluid volumetric concentrations from well logs in thinly bedded and invaded formations","volume":"77","author":"Z Heidari","year":"2012","journal-title":"Geophysics"},{"issue":"8","key":"pone.0347378.ref006","doi-asserted-by":"crossref","first-page":"5932","DOI":"10.1109\/TGRS.2020.2973171","article-title":"Physics-constrained deep learning of geomechanical logs","volume":"58","author":"Y Chen","year":"2020","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"issue":"6","key":"pone.0347378.ref007","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1190\/1.1440465","article-title":"Formation velocity and density; The diagnostic basics for stratigraphic traps","volume":"39","author":"G Gardner","year":"1974","journal-title":"Geophysics"},{"issue":"16","key":"pone.0347378.ref008","article-title":"Well logging curve reconstruction based on kernel ridge regression","volume":"14","author":"P Fan","year":"2021","journal-title":"Arab J Geosci"},{"key":"pone.0347378.ref009","doi-asserted-by":"crossref","first-page":"104836","DOI":"10.1016\/j.jappgeo.2022.104836","article-title":"Robust data smoothing algorithms and wavelet filter for denoising sonic log signals","volume":"206","author":"S Aftab","year":"2022","journal-title":"Journal of Applied Geophysics"},{"issue":"3","key":"pone.0347378.ref010","doi-asserted-by":"crossref","first-page":"385","DOI":"10.17014\/ijog.8.3.385-399","article-title":"The application of parametric and nonparametric regression to predict the missing well log data","volume":"8","author":"M Mordekhai","year":"2021","journal-title":"Indonesian J Geosci"},{"key":"pone.0347378.ref011","doi-asserted-by":"crossref","first-page":"106106","DOI":"10.1016\/j.jappgeo.2026.106106","article-title":"A deep learning-based method for well log data reconstruction in marine carbonate reservoirs","volume":"246","author":"Z Kang","year":"2026","journal-title":"Journal of Applied Geophysics"},{"issue":"4","key":"pone.0347378.ref012","doi-asserted-by":"crossref","first-page":"629","DOI":"10.1016\/S1876-3804(18)30068-5","article-title":"Synthetic well logs generation via Recurrent Neural Networks","volume":"45","author":"D Zhang","year":"2018","journal-title":"Petroleum Exploration and Development"},{"key":"pone.0347378.ref013","doi-asserted-by":"crossref","unstructured":"Chang J, Li J, Liu H, Kang Y, Lv W. Well Logging Reconstruction Based on Bidirectional GRU. In: Proceedings of the 2022 2nd International Conference on Control and Intelligent Robotics, 2022. 525\u20138. https:\/\/doi.org\/10.1145\/3548608.3559256","DOI":"10.1145\/3548608.3559256"},{"issue":"5","key":"pone.0347378.ref014","doi-asserted-by":"crossref","first-page":"4901","DOI":"10.1007\/s12145-024-01425-0","article-title":"Deep learning-aided simultaneous missing well log prediction in multiple stratigraphic units: A case study from the Bhogpara oil field, Upper Assam, Northeast India","volume":"17","author":"B Mukherjee","year":"2024","journal-title":"Earth Sci Inform"},{"key":"pone.0347378.ref015","doi-asserted-by":"crossref","first-page":"125270","DOI":"10.1016\/j.eswa.2024.125270","article-title":"Missing well logs reconstruction based on cascaded bidirectional long short-term memory network","volume":"259","author":"W Zhou","year":"2025","journal-title":"Expert Systems with Applications"},{"key":"pone.0347378.ref016","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1016\/j.neucom.2020.07.026","article-title":"Attention-based bidirectional gated recurrent unit neural networks for well logs prediction and lithology identification","volume":"414","author":"L Zeng","year":"2020","journal-title":"Neurocomputing"},{"key":"pone.0347378.ref017","doi-asserted-by":"crossref","unstructured":"Li W, Fu L, Al Tammar MJ. Well log prediction using deep sequence learning. In: ARMA\/DGS\/SEG International Geomechanics Symposium, 2022. ARMA-IGS-2022-184.","DOI":"10.56952\/IGS-2022-184"},{"issue":"2","key":"pone.0347378.ref018","doi-asserted-by":"crossref","first-page":"100536","DOI":"10.1016\/j.engeos.2026.100536","article-title":"Toward basin-agnostic well log imputation and anomaly detection via a pre-trained time-series foundation model","volume":"7","author":"A Koeshidayatullah","year":"2026","journal-title":"Energy Geoscience"},{"key":"pone.0347378.ref019","doi-asserted-by":"crossref","first-page":"104763","DOI":"10.1016\/j.cageo.2021.104763","article-title":"Imputation of missing well log data by random forest and its uncertainty analysis","volume":"152","author":"R Feng","year":"2021","journal-title":"Computers & Geosciences"},{"issue":"2","key":"pone.0347378.ref020","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1093\/jge\/gxaf131","article-title":"Synthetic well log generation based on SBOA-XGBoost and SHAP","volume":"23","author":"W Zeng","year":"2025","journal-title":"Journal of Geophysics and Engineering"},{"issue":"1","key":"pone.0347378.ref021","doi-asserted-by":"crossref","first-page":"11000","DOI":"10.1038\/s41598-025-95709-0","article-title":"Well log data generation and imputation using sequence based generative adversarial networks","volume":"15","author":"A Al-Fakih","year":"2025","journal-title":"Sci Rep"},{"issue":"4","key":"pone.0347378.ref022","doi-asserted-by":"crossref","DOI":"10.1007\/s12145-025-02030-5","article-title":"A method for reconstructing logging curves based on characterization-enhanced variational autoencoder","volume":"18","author":"X Zhang","year":"2025","journal-title":"Earth Sci Inform"},{"key":"pone.0347378.ref023","doi-asserted-by":"crossref","first-page":"106106","DOI":"10.1016\/j.cageo.2026.106106","article-title":"Deep-learning-based hybrid model with iterative lithology constraints for the enhanced prediction of missing well-logs","volume":"209","author":"J Park","year":"2026","journal-title":"Computers & Geosciences"},{"issue":"02","key":"pone.0347378.ref024","doi-asserted-by":"crossref","first-page":"153","DOI":"10.2118\/6859-PA","article-title":"Theoretical and experimental bases for the dual-water model for interpretation of shaly sands","volume":"24","author":"C Clavier","year":"1984","journal-title":"Society of Petroleum Engineers Journal"},{"issue":"3","key":"pone.0347378.ref025","doi-asserted-by":"crossref","first-page":"251","DOI":"10.17014\/ijog.5.3.251-263","article-title":"Determination of hydrocarbon zones using logging data analysis in a sandstone reservoir (case study: Structure \u2018TL\u2019 Basin North West Java)","volume":"5","author":"H Herianto","year":"2018","journal-title":"Indonesian J Geosci"},{"issue":"1","key":"pone.0347378.ref026","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1190\/1.1438217","article-title":"Elastic wave velocities in heterogeneous and porous media","volume":"21","author":"MRJ Wyllie","year":"1956","journal-title":"Geophysics"},{"issue":"2","key":"pone.0347378.ref027","doi-asserted-by":"crossref","first-page":"169","DOI":"10.1190\/1.1438855","article-title":"Some aspects of elastic wave propagation in fluid-saturated porous solids","volume":"26","author":"J Geertsma","year":"1961","journal-title":"Geophysics"},{"issue":"7","key":"pone.0347378.ref028","doi-asserted-by":"crossref","first-page":"1285","DOI":"10.1306\/03B5949A-16D1-11D7-8645000102C1865D","article-title":"Determination of organic-matter content of appalachian devonian shales from gamma-ray logs","volume":"65","author":"JW Schmoker","year":"1981","journal-title":"AAPG Bulletin"},{"key":"pone.0347378.ref029","doi-asserted-by":"crossref","unstructured":"Amaefule JO, Altunbay M, Tiab D, Kersey DG, Keelan DK. Enhanced Reservoir Description: Using Core and Log Data to Identify Hydraulic (Flow) Units and Predict Permeability in Uncored Intervals\/Wells. In: SPE Annual Technical Conference and Exhibition, 1993. https:\/\/doi.org\/10.2118\/26436-ms","DOI":"10.2118\/26436-MS"},{"issue":"05","key":"pone.0347378.ref030","article-title":"Response of neutron and formation density logs in hydrocarbon bearing formations","volume":"9","author":"R Gaymard","year":"1968","journal-title":"The Log Analyst"},{"issue":"3","key":"pone.0347378.ref031","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1190\/1.3567263","article-title":"Rock physics of organic shales","volume":"30","author":"L Vernik","year":"2011","journal-title":"The Leading Edge"},{"key":"pone.0347378.ref032","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016. 785\u201394.","DOI":"10.1145\/2939672.2939785"},{"issue":"6","key":"pone.0347378.ref033","article-title":"Missing well-log reconstruction using a sequence self-attention deep-learning framework","volume":"88","author":"L Lin","year":"2023","journal-title":"Geophysics"},{"issue":"4","key":"pone.0347378.ref034","doi-asserted-by":"crossref","first-page":"3071","DOI":"10.1007\/s12145-024-01323-5","article-title":"Missing log prediction using machine learning perspectives: A case study from upper Assam basin","volume":"17","author":"B Mukherjee","year":"2024","journal-title":"Earth Sci Inform"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0347378","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T17:28:48Z","timestamp":1791912528000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0347378"}},"subtitle":[],"editor":[{"given":"Xinyuan","family":"Gao","sequence":"first","affiliation":[{"name":"Northeastern University","place":["CHINA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,13]]},"references-count":34,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,13]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0347378","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,13]]}}}