Published date: Nov 19 2024

Journal Title: KnE Social Sciences

Issue title: The 1st International Conference on Creative Design, Business and Society (1st ICCDBS) 2023

Pages: 239–254

DOI: 10.18502/kss.v9i32.17439

Muhammad Vito Hamzavito.hamza@geodipa.co.idPT Geo Dipa Energi (Persero)

Fransisco T.P. SimamoraPT Geo Dipa Energi (Persero)

Efrata Pratenta MelialaPT Geo Dipa Energi (Persero)

R. Fuad Satrio AjiePT Geo Dipa Energi (Persero)

Hanifah Nur AzizahPT Geo Dipa Energi (Persero)

Fajar Khamim MustofaPT Geo Dipa Energi (Persero)

Adi SuparyantoPT Geo Dipa Energi (Persero)

Geothermal power plants are crucial for sustainable energy generation, necessitating the reliable maintenance of their operating assets. This research proposes an approach for asset maintenance through anomaly detection using the Locality- Sensitive Hashing (LSH) algorithm. The accuracy and coverage of traditional anomaly detection approaches in geothermal power plants may be constrained by sensor monitoring systems. The LSH algorithm is used to improve detection skills and get a full understanding of the state of important assets. The proposed method utilizes historical sensor data collected during geothermal power plant operations. This data is transformed into hash codes using LSH, effectively capturing similarities between various operational states and asset conditions. By comparing the hash codes of the current operational state with a library of precomputed hash codes representing typical operating conditions, the LSH algorithm can identify deviations indicating potential irregularities. This facilitates early detection of anomalies, even in large-scale databases, enabling prompt maintenance interventions. The application of anomaly detection using the LSH algorithm provides benefits such as improved asset maintenance planning, reduced downtime, and increased operational safety. By leveraging data-driven analysis and the effectiveness of LSH, geothermal operators can detect faults early, enabling prompt interventions and optimizing reliability and efficiency. By leveraging historical sensor data and the efficient similarity approximation capabilities of LSH, the proposed approach enables early diagnosis of problems, improving maintenance planning and optimizing geothermal operations.

Keywords: geothermal assets, locality-sensitive hashing, asset condition, fault detection, reliability

[1] IESR. Indonesia energy transition outlook 2023: Tracking progress of energy transition in indonesia: Pursuing energy security in the time of transition. Institute for Essential Services Reform (IESR), Jakarta. 2022

[2] ESDM. National Electricity Supply Business Plan (RUPTL) PT PLN Persero. Ministry of Energy and Mineral Resources (ESDM), Jakarta; 2021.

[3] Purba D, Chandra VR, Fadhillah FR, et al. Drilling infrastructure construction challenges in geothermal exploration project in Eastern Indonesia. In: Proceedings World Geothermal Congress 2020+1; International Geothermal Association (IGA), Iceland. 2021.

[4] Fadhillah FR, Asyari MR, Bagaskara A, et al. Challenges in Getting Public Acceptance on Geothermal Project in Indonesia. PROCEEDINGS, 48th Workshop on Geothermal Reservoir Engineering. Stanford University, California. 2023.

[5] Institute for Essential Services Reform (IESR). A roadmap for Indonesia’s power sector: How renewable energy can power Java-Bali and Sumatra. Jakarta: IESR; 2019.

[6] ISO Technical Committee 251. ISO 55000:2014 - Asset management. International Organization for Standardization (ISO) 2014. Link:https://www.iso.org/obp/ui/#iso:std:iso:55000:ed-1:en

[7] Mitchell JS. Physical asset management Handbook Fourth Edition. Jakarta: MTS Indonesia; 2009.

[8] Gulati R. Maintenance and reliability best practices. Third edition. Industrial Press, Inc.; 2020.

[9] Bisset C, Venter PV, Coetzer R. A systematic literature review on machine learning applications at coal-fired thermal power plants for improved energy efficiency. Int J Sustain Energy. 2023;42(1):845–72.

[10] Fung PL, Zaidan MA, Timonen H, et al. Evaluation of white-box versus black-box machine learning models in estimating ambient black carbon concentration. J. Aerosol Sci. 2021;152:105694. DOI: 10.1016/j.jaerosci.2020.105694

[11] Shen H, Li T, Schweiger T, et al. Locality sensitive hashing. In [eds]. Encyclopedia of Computer Science and Technology. CRC Press; 2016, pp. 562-574. Available from: https://www.routledgehandbooks.com/doi/10.1081/E-ECST2-120052698

[12] Gangwani P, Joshi S, Upadhyay H, Lagos L. AI-Based anomaly detection on IoT datadriven thermal power plants for condition monitoring and preventive maintenance. In: Bhardwaj T, Upadhyay H, Sharma TK, Fernandes SL, editors. Artificial intelligence in cyber security: Theories and applications. Intelligent systems reference library. Volume 240. Cham: Springer; 2023. DOI: 10.1007/978-3-031-28581-3_8

[13] S. Joshi, H. Upadhyay, L. Lagos, N. S. Akkipeddi, and V. Guerra, “Machine learning approach for malware detection using random forest classifier on process list data structure,” in ACM International Conference Proceeding Series, Association for Computing Machinery, Apr. 2018, pp. 98–102. doi: 10.1145/3206098.3206113.

[14] Xu D, Tian Y. A comprehensive survey of clustering algorithms. Ann. Data. Sci. 2015;2(2):165–93.

[15] Y. F. Wang, Y. Jiong, G. P. Su, and Y. R. Qian, “A new outlier detection method based on OPTICS,” Sustainable Cities and Society, vol. 45, pp. 197–212, Feb. 2019, doi: 10.1016/j.scs.2018.11.031.

[16] Hajdarevic A, Dzananovic I, Banjanovic-Mehmedovic L, et al. Anomaly detection in thermal power plant using probabilistic neural network. In: 38th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO); Croatia. 2015. pp. 1118-1123.

[17] Ronald. DiPippo, Geothermal power plants: principles, applications, case studies, and environmental impact. Butterworth-Heinemann, imprint of Elsevier, 2016.

[18] L. Ljung, “Black-box models from input-output measurements,” IMTC 2001. Proceedings of the 18th IEEE Instrumentation and Measurement Technology Conference. Rediscovering Measurement in the Age of Informatics (Cat. No.01CH 37188), Budapest, Hungary, 2001, pp. 138-146 vol.1, doi: 10.1109/IMTC.2001.928802.

[19] Loyola-Gonzales O. Black-box vs. white-box: Understanding their advantages and weaknesses from a practical point of view. IEEE Access. 2019;7:154096–113.

[20] Murdoch WJ, Singh C, Kumbier K, Abbasi-Asl R, Yu B. Definitions, methods, and applications in interpretable machine learning. Proc Natl Acad Sci USA. 2019 Oct;116(44):22071–80.

[21] V. Hassija et al., “Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence,” Cognitive Computation, vol. 16, no. 1. Springer, pp. 45–74, Jan. 01, 2024. doi: 10.1007/s12559-023-10179-8.

[22] Saeed W, Omlin C. Explainable AI (XAI): A systematic meta-survey of current challenges and future opportunities. Knowl.-Based Syst. 2023;263:110273. DOI: 10.1016/j.knosys.2023.110273.

[23] Aristides Gionis, Piotr Indyk, and Rajeev Motwani. 1999. Similarity Search in High Dimensions via Hashing. In Proceedings of the 25th International Conference on Very Large Data Bases (VLDB ’99). Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 518–529. DOI: https://dl.acm.org/doi/10.5555/645925.671516

[24] Meira J, Eiras-Franco C, Bolón-Canedo V, et al. Fast anomaly detection with localitysensitive hashing and hyperparameter autotuning. Inf. Sci. 2022;607:1245-1264. DOI: 10.1016/j.ins.2022.06.035.

[25] Mihnev A. Method and apparatus for detection of anomalies in integrated parameter systems. US PATENT. US 2015/0199224 AI. 2015. Available from: https://patents.google.com/patent/US20150199224A1/en