Predictive Strategy and Technology for Operation & Maintenance Decision-Making

In power generation must have robust decision-making, either for operation & maintenance (O&M) decision-making or capital investment decision-making. This aims to increase reliability and efficiency with life cycle cost (LCC) as low as reasonably practicable (ALARP). To achieve these goals, power generation should implement asset health management (AHM) through digital power plant (DPP), where one of the functions is prognostic health management.

In previous research related to prognostic health management, the asset condition criteria were based on the asset health index (AHI) of the asset. The smaller the AHI, then the criterion is danger. Even though every asset that has the same AHI, it doesn't necessarily mean that the remaining uptime is the same. So that the recommendations generated based on the criteria have a low effect on improving asset condition and power generation performance. Therefore, it is not possible to use AHI as the asset condition criteria.

In this research, the prognostic (prediction) in DPP requires strategy and technology, where there are 3 (three) predictive strategies, namely predictions obtained from the input parameters of online performance monitoring (asset performance management / APM), online / offline conditions-based monitoring (APM), and computerized maintenance management systems (enterprise asset management / EAM). APM is used for short-term planning ( < 1 year), where criteria developed based on the remaining uptime of power generation (PF-curve) are used for decision-making. Meanwhile, EAM is used for long-term planning ( ≥ 1 year), where risk cost criteria are used for decision-making.

Keywords: Predictive strategy and technology, Operation & maintenance decision-making, Capital investment decision-making, Prognostic health management, Asset health management, Asset health index, Asset condition criteria.

Download PDF