{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T14:37:30Z","timestamp":1775745450695,"version":"3.50.1"},"reference-count":50,"publisher":"Association for Computing Machinery (ACM)","issue":"3","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Cyber-Phys. Syst."],"published-print":{"date-parts":[[2025,7,31]]},"abstract":"<jats:p>Power consumption anomaly detection systems that use neural networks for prediction tasks are vulnerable to adversarial attacks, leading to unreliable performance and potential adverse effects on the power grid. Certain attack configurations can evade detection or trigger false alarms due to the neural networks\u2019 generalization tendencies, causing adaptation to attack values. This study investigates the effectiveness of adversarial training methods in improving detection performance against such attacks on power consumption data. Leveraging techniques like the Fast Gradient Sign Method (FGSM), Basic Iterative Method (BIM), and Projected Gradient Descent (PGD), we assess the resilience of anomaly detection models. Through empirical experiments, we evaluate detection accuracy, adjustment capabilities, and prediction errors of these adversarially trained models across three datasets. Our results show significant improvements in detection performance, particularly in attack scenarios that normal prediction models would not detect. Additionally, we analyze the models\u2019 adaptability to anomalous data and quantify prediction errors, providing insights into their robustness and limitations. Integrating adversarial training techniques into anomaly detection models for power grids can reduce over-generalization to attack data, enhancing the detection of malicious demand manipulation attacks.<\/jats:p>","DOI":"10.1145\/3736640","type":"journal-article","created":{"date-parts":[[2025,5,22]],"date-time":"2025-05-22T22:15:44Z","timestamp":1747952144000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Mitigating Over-Generalization in Anomalous Power Consumption Detection using Adversarial Training"],"prefix":"10.1145","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0117-8911","authenticated-orcid":false,"given":"Srinidhi","family":"Madabhushi","sequence":"first","affiliation":[{"name":"University of Denver, Denver, Colorado, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8332-2157","authenticated-orcid":false,"given":"Rinku","family":"Dewri","sequence":"additional","affiliation":[{"name":"University of Denver, Denver, Colorado, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,8,4]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","unstructured":"Baldemar Aguirre-Fraire Jessica Beltr\u00e1n and Valeria Soto. 2024. Household energy consumption enriched with weather data in northeast of Mexico. Data set. DOI:10.17632\/tvhygj8rgg.1","DOI":"10.17632\/tvhygj8rgg.1"},{"key":"e_1_3_2_3_2","unstructured":"Sheila Alemany and Niki Pissinou. 2021. The dilemma between data transformations and adversarial robustness for time series application systems. arXiv:2006.10885. Retrieved from https:\/\/arxiv.org\/abs\/2006.10885"},{"key":"e_1_3_2_4_2","first-page":"928","volume-title":"13th International Wireless Communications and Mobile Computing Conference","author":"Aloqaily Moayad","year":"2017","unstructured":"Moayad Aloqaily, Burak Kantarci, and Hussein T. Mouftah. 2017. Trusted third party for service management in vehicular clouds. In 13th International Wireless Communications and Mobile Computing Conference, 928\u2013933."},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2020.3012171"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1613\/jair.1.13543"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/2434020.2434043"},{"key":"e_1_3_2_8_2","unstructured":"Liron Bergman Niv Cohen and Yedid Hoshen. 2020. Deep nearest neighbor anomaly detection. arXiv:2002.10445. Retrieved from https:\/\/arxiv.org\/abs\/2002.10445"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1007\/s12053-020-09884-2"},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.rser.2014.01.088"},{"key":"e_1_3_2_11_2","first-page":"303","volume-title":"33rd Annual Computer Security Applications Conference","author":"Dabrowski Adrian","year":"2017","unstructured":"Adrian Dabrowski, Johanna Ullrich, and Edgar R. Weippl. 2017. Grid shock: Coordinated load-changing attacks on power grids: The non-smart power grid is vulnerable to cyber attacks as well. In 33rd Annual Computer Security Applications Conference. ACM, 303\u2013314."},{"key":"e_1_3_2_12_2","unstructured":"Islam Debicha Thibault Debatty Jean-Michel Dricot and Wim Mees. 2021. Adversarial training for deep learning-based intrusion detection systems. arXiv:2104.09852. Retrieved from https:\/\/arxiv.org\/abs\/2104.09852"},{"key":"e_1_3_2_13_2","unstructured":"U.S.-Canada Power System Outage Task Force. 2004. Final Report on the August 14 2003 Blackout in the United States and Canada: Causes and Recommendations. Retrieved from https:\/\/www.energy.gov\/sites\/prod\/files\/oeprod\/DocumentsandMedia\/BlackoutFinal-Web.pdf"},{"issue":"5","key":"e_1_3_2_14_2","first-page":"2978","article-title":"Tsmae: A novel anomaly detection approach for internet of things time series data using memory-augmented autoencoder","volume":"10","author":"Gao Honghao","year":"2022","unstructured":"Honghao Gao, Binyang Qiu, Ramon J. Duran Barroso, Walayat Hussain, Yueshen Xu, and Xinheng Wang. 2022. Tsmae: A novel anomaly detection approach for internet of things time series data using memory-augmented autoencoder. IEEE Transactions on Network Science and Engineering 10, 5 (2022), 2978\u20132990.","journal-title":"IEEE Transactions on Network Science and Engineering"},{"key":"e_1_3_2_15_2","doi-asserted-by":"crossref","unstructured":"Dong Gong Lingqiao Liu Vuong Le Budhaditya Saha Moussa Reda Mansour Svetha Venkatesh and Anton van den Hengel. 2019. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. arXiv:1904.02639. Retrieved from https:\/\/arxiv.org\/abs\/1904.02639","DOI":"10.1109\/ICCV.2019.00179"},{"key":"e_1_3_2_16_2","unstructured":"Ian J. Goodfellow Jonathon Shlens and Christian Szegedy. 2015. Explaining and harnessing adversarial examples. arXiv:1412.6572. Retrieved from https:\/\/arxiv.org\/abs\/1412.6572"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.24432\/C58K54"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","unstructured":"M. Hofmann and T. Siebenbrunner. 2023. A rich dataset of hourly residential electricity consumption data and survey answers from the iFlex dynamic pricing experiment. Data set. DOI:10.5281\/zenodo.8248802","DOI":"10.5281\/zenodo.8248802"},{"key":"e_1_3_2_19_2","first-page":"1","volume-title":"International Conference on Computer Communications and Networks","author":"Jadidi Zahra","year":"2022","unstructured":"Zahra Jadidi, Shantanu Pal, Nithesh Nayak, Arawinkumaar Selvakkumar, Chih-Chia Chang, Maedeh Beheshti, and Alireza Jolfaei. 2022. Security of machine learning-based anomaly detection in cyber physical systems. In International Conference on Computer Communications and Networks, 1\u20137."},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijcip.2021.100452"},{"issue":"10","key":"e_1_3_2_21_2","first-page":"3309","volume":"43","author":"Karim Fazle","year":"2020","unstructured":"Fazle Karim, Somshubra Majumdar, and Houshang Darabi. 2020. Adversarial attacks on time series. IEEE Transactions on Pattern Analysis and Machine Intelligence 43, 10 (2020), 3309\u20133320.","journal-title":"Adversarial attacks on time series. IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"e_1_3_2_22_2","first-page":"481","volume-title":"International Conference on Intelligent Data Engineering and Automated Learning","author":"Kim Tae-Young","year":"2018","unstructured":"Tae-Young Kim and Sung-Bae Cho. 2018. Predicting the household power consumption using CNN-LSTM hybrid networks. In International Conference on Intelligent Data Engineering and Automated Learning, 481\u2013490."},{"key":"e_1_3_2_23_2","unstructured":"Alexey Kurakin Ian Goodfellow and Samy Bengio. 2017. Adversarial examples in the physical world. arXiv:1607.02533. Retrieved from https:\/\/arxiv.org\/abs\/1607.02533"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2020.2975654"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/3433210.3437513"},{"key":"e_1_3_2_26_2","first-page":"220","volume-title":"International Conference on Information Systems Security","author":"Madabhushi Srinidhi","year":"2022","unstructured":"Srinidhi Madabhushi and Rinku Dewri. 2022. On the impact of model tolerance in power grid anomaly detection systems. In International Conference on Information Systems Security, 220\u2013234."},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10207-023-00720-z"},{"key":"e_1_3_2_28_2","unstructured":"Aleksander Madry Aleksandar Makelov Ludwig Schmidt Dimitris Tsipras and Adrian Vladu. 2019. Towards deep learning models resistant to adversarial attacks. arXiv:1706.06083. Retrieved from https:\/\/arxiv.org\/abs\/1706.06083"},{"key":"e_1_3_2_29_2","first-page":"467","volume-title":"19th IEEE International Conference on Machine Learning and Applications","author":"Gautam Raj Mode and Khaza Anuarul Hoque","year":"2020","unstructured":"Gautam Raj Mode and Khaza Anuarul Hoque. 2020. Crafting adversarial examples for deep learning based prognostics. In 19th IEEE International Conference on Machine Learning and Applications, 467\u2013472."},{"key":"e_1_3_2_30_2","doi-asserted-by":"crossref","unstructured":"Seyed-Mohsen Moosavi-Dezfooli Alhussein Fawzi and Pascal Frossard. 2016. DeepFool: A simple and accurate method to fool deep neural networks. arXiv:1511.04599. Retrieved from https:\/\/arxiv.org\/abs\/1511.04599","DOI":"10.1109\/CVPR.2016.282"},{"key":"e_1_3_2_31_2","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1109\/SERA57763.2023.10197774","volume-title":"2023 IEEE\/ACIS 21st International Conference on Software Engineering Research, Management and Applications (SERA)","author":"Mulo John","year":"2023","unstructured":"John Mulo, Pu Tian, Adamu Hussaini, Hengshuo Liang, and Wei Yu. 2023. Towards an adversarial machine learning framework in cyber-physical systems. In 2023 IEEE\/ACIS 21st International Conference on Software Engineering Research, Management and Applications (SERA). IEEE, 138\u2013143."},{"issue":"1","key":"e_1_3_2_32_2","doi-asserted-by":"crossref","first-page":"524","DOI":"10.1109\/COMST.2020.3036778","article-title":"Resilient machine learning for networked cyber physical systems: A survey for machine learning security to securing machine learning for CPS","volume":"23","author":"Olowononi Felix O.","year":"2020","unstructured":"Felix O. Olowononi, Danda B. Rawat, and Chunmei Liu. 2020. Resilient machine learning for networked cyber physical systems: A survey for machine learning security to securing machine learning for CPS. IEEE Communications Surveys & Tutorials 23, 1 (2020), 524\u2013552.","journal-title":"IEEE Communications Surveys & Tutorials"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2019.2903882"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cose.2022.102783"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01392"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.3390\/s23125459"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-63836-8_45"},{"key":"e_1_3_2_38_2","volume-title":"27th USENIX Security Symposium","author":"Soltan Saleh","year":"2018","unstructured":"Saleh Soltan, Prateek Mittal, and H. Vincent Poor. 2018. BlackIoT: IoT botnet of high wattage devices can disrupt the power grid. In 27th USENIX Security Symposium."},{"key":"e_1_3_2_39_2","first-page":"57947","article-title":"Memto: Memory-guided transformer for multivariate time series anomaly detection","volume":"36","author":"Song Junho","year":"2023","unstructured":"Junho Song, Keonwoo Kim, Jeonglyul Oh, and Sungzoon Cho. 2023. Memto: Memory-guided transformer for multivariate time series anomaly detection. In Advances in Neural Information Processing Systems, Vol. 36, 57947\u201357963.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_2_40_2","doi-asserted-by":"crossref","first-page":"530","DOI":"10.1109\/MASS56207.2022.00080","volume-title":"2022 IEEE 19th International Conference on Mobile Ad Hoc and Smart Systems (MASS)","author":"Sun Zhe","year":"2022","unstructured":"Zhe Sun and Jinguo Li. 2022. Anomaly detection for CPS via memory-augmented reconstruction and time series prediction. In 2022 IEEE 19th International Conference on Mobile Ad Hoc and Smart Systems (MASS). IEEE, 530\u2013536."},{"issue":"1","key":"e_1_3_2_41_2","doi-asserted-by":"crossref","first-page":"663","DOI":"10.1109\/TSG.2022.3193989","article-title":"Robust data-driven detection of electricity theft adversarial evasion attacks in smart grids","volume":"14","author":"Takiddin Abdulrahman","year":"2022","unstructured":"Abdulrahman Takiddin, Muhammad Ismail, and Erchin Serpedin. 2022. Robust data-driven detection of electricity theft adversarial evasion attacks in smart grids. IEEE Transactions on Smart Grid 14, 1 (2022), 663\u2013676.","journal-title":"IEEE Transactions on Smart Grid"},{"key":"e_1_3_2_42_2","first-page":"3959","volume-title":"2020 American control conference (ACC)","author":"Liang Tan Kai","year":"2020","unstructured":"Kai Liang Tan, Yasaman Esfandiari, Xian Yeow Lee, Aakanksha, Soumik Sarkar. 2020. Robustifying reinforcement learning agents via action space adversarial training. In 2020 American control conference (ACC). IEEE, 3959\u20133964."},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNSE.2021.3135565"},{"key":"e_1_3_2_44_2","first-page":"1","volume-title":"2024 IEEE Power & Energy Society General Meeting (PESGM)","author":"Wang Kebei","year":"2024","unstructured":"Kebei Wang and Manimaran Govindarasu. 2024. FGSM-based synthetic data generation technique and application to anomaly detection in smart grid. In 2024 IEEE Power & Energy Society General Meeting (PESGM). IEEE, 1\u20135."},{"key":"e_1_3_2_45_2","first-page":"31","volume-title":"Cyber-Physical Anomaly Detection for Power Grid with Machine Learning","author":"Wang Pengyuan","year":"2019","unstructured":"Pengyuan Wang and Manimaran Govindarasu. 2019. Cyber-Physical Anomaly Detection for Power Grid with Machine Learning. Springer International Publishing, 31\u201349."},{"key":"e_1_3_2_46_2","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2023.3344808"},{"key":"e_1_3_2_47_2","volume-title":"4th International Conference on Cloud Computing and Big Data Analysis","author":"Wang Xiaohui","year":"2019","unstructured":"Xiaohui Wang, Ting Zhao, He Liu, and Rong He. 2019. Power consumption predicting and anomaly detection based on long short-term memory neural network. In 4th International Conference on Cloud Computing and Big Data Analysis."},{"key":"e_1_3_2_48_2","volume-title":"ICML 2011 Workshop on Machine Learning for Global Challenge","author":"Wu Leon","year":"2011","unstructured":"Leon Wu, Gail Kaiser, Cynthia Rudin, David Waltz, Roger Anderson, Albert Boulanger, Ansaf Salleb-Aouissi, Haimonti Dutta, and Manoj Pooleery. 2011. Evaluating machine learning for improving power grid reliability. In ICML 2011 Workshop on Machine Learning for Global Challenge."},{"key":"e_1_3_2_49_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2021.11.007"},{"key":"e_1_3_2_50_2","first-page":"8","volume-title":"2021 10th International Conference on Internet Computing for Science and Engineering. ACM","author":"Xu Aidong","year":"2022","unstructured":"Aidong Xu, Xuechun Wang, Yunan Zhang, Tao Wu, and Xingping Xian. 2022. Adversarial attacks on deep neural networks for time series prediction. In 2021 10th International Conference on Internet Computing for Science and Engineering. ACM, 8\u201314."},{"key":"e_1_3_2_51_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2022.10.017"}],"container-title":["ACM Transactions on Cyber-Physical Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3736640","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,4]],"date-time":"2025-08-04T15:54:28Z","timestamp":1754322868000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3736640"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,31]]},"references-count":50,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,7,31]]}},"alternative-id":["10.1145\/3736640"],"URL":"https:\/\/doi.org\/10.1145\/3736640","relation":{},"ISSN":["2378-962X","2378-9638"],"issn-type":[{"value":"2378-962X","type":"print"},{"value":"2378-9638","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,31]]},"assertion":[{"value":"2024-11-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-04-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-08-04","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}