Use of Machine Learning-Based Health Index With K-Nearest Neighbors Method to Maintain Desalination Plant Performance Gas and Steam Power Plants Applications
DOI:
https://doi.org/10.59888/ajosh.v3i7.549Keywords:
Machine learning, K-Nearest Neighbors (K-NN), desalination plant, predictive maintenance, power plant efficiencyAbstract
This study presents the implementation of a Machine Learning-Based Health Index utilizing the K-Nearest Neighbors (K-NN) algorithm for predictive maintenance in desalination plants within gas and steam power plants. The research focuses on optimizing the maintenance schedule of the Block 3 Priok Desalination Plant, which is critical for providing high-quality distilled water for power generation. This study aims to develop and integrate a predictive maintenance framework into PLN’s digitization system, allowing for automated monitoring and optimized servicing schedules. Unlike the previous application of K-NN in Block 4, which utilized five health indices for performance classification, Block 3 requires an expanded model incorporating at least seven input parameters due to its multi-effect desalination process. By refining the predictive model and increasing data parameterization, this study seeks to enhance maintenance accuracy, minimize operational downtime, and improve overall desalination efficiency. By leveraging historical operational data and real-time monitoring, the K-NN model predicts the health index of desalination components with 98% accuracy. Implementing this approach minimizes downtime, optimizes maintenance schedules, and enhances energy efficiency. The results demonstrate that AI-driven predictive maintenance significantly improves reliability, reduces costs, and supports energy sustainability goals.
Published
Issue
Section
License
Copyright (c) 2025 Udi Harmoko, Marcelinus Christwardana, Muhammad Rizkan

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International. that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.