Surveys on explainable artificial intelligence (XAI) are related to biology, clinical trials, fintech management, medicine, neurorobotics, and psychology, among others. Prognostics and health management (PHM) is the discipline that links the studies of failure mechanisms to system lifecycle management. There is a need, which is still absent, to produce an analytical compilation of PHM-XAI works. In this paper, we use preferred reporting items for systematic reviews and meta-analyses (PRISMA) to present a state of the art on XAI applied to PHM of industrial assets. This work provides an overview of the trend of XAI in PHM and answers the question of accuracy versus explainability, considering the extent of human involvement, explanation assessment, and uncertainty quantification in this topic. Research articles associated with the subject, since 2015 to 2021, were selected from five databases following the PRISMA methodology, several of them related to sensors. The data were extracted from selected articles and examined obtaining diverse findings that were synthesized as follows. First, while the discipline is still young, the analysis indicates a growing acceptance of XAI in PHM. Second, XAI offers dual advantages, where it is assimilated as a tool to execute PHM tasks and explain diagnostic and anomaly detection activities, implying a real need for XAI in PHM. Third, the review shows that PHM-XAI papers provide interesting results, suggesting that the PHM performance is unaffected by the XAI. Fourth, human role, evaluation metrics, and uncertainty management are areas requiring further attention by the PHM community. Adequate assessment metrics to cater to PHM needs are requested. Finally, most case studies featured in the considered articles are based on real industrial data, and some of them are related to sensors, showing that the available PHM-XAI blends solve real-world challenges, increasing the confidence in the artificial intelligence models’ adoption in the industry.

Mistrust, amplified by numerous artificial intelligence (AI) related incidents, is an issue that has caused the energy and industrial sectors to be amongst the slowest adopter of AI methods. Central to this issue is the black-box problem of AI, which impedes investments and is fast becoming a legal hazard for users. Explainable AI (XAI) is a recent paradigm to tackle such an issue. Being the backbone of the industry, the prognostic and health management (PHM) domain has recently been introduced into XAI. However, many deficiencies, particularly the lack of explanation assessment methods and uncertainty quantification, plague this young domain. In the present paper, we elaborate a framework on explainable anomaly detection and failure prognostic employing a Bayesian deep learning model and Shapley additive explanations (SHAP) to generate local and global explanations from the PHM tasks. An uncertainty measure of the Bayesian model is utilized as a marker for anomalies and expands the prognostic explanation scope to include the model’s confidence. In addition, the global explanation is used to improve prognostic performance, an aspect neglected from the handful of studies on PHM-XAI. The quality of the explanation is examined employing local accuracy and consistency properties. The elaborated framework is tested on real-world gas turbine anomalies and synthetic turbofan failure prediction data. Seven out of eight of the tested anomalies were successfully identified. Additionally, the prognostic outcome showed a 19% improvement in statistical terms and achieved the highest prognostic score amongst best published results on the topic.

This study’s primary goal is to evaluate the performance of a large thermal energy storage tank installed in a Gas District Cooling (GDC) plant. The performance parameters considered in this study include thermocline thickness (WTc), Cumulated Charge (Q_cum), and Half Figure of Merit (½ FOM). The operation sensor data of a large Thermal Energy Storage (TES) tank was acquired for this analysis. The recorded temperature sensor from the 1st to 7th January and from 12th to 17th October 2019 was considered in this research. GraphPad prism computer software was deployed for analyses, and the temperature distribution data were analyzed to determine the four temperature parameters (hot water temperature (Th), cool water temperature (Tc), cool water depth (C), and slope gradient (S)) using a non-linear regression curve fitting technique and sigmoid Dose Responses function as integrated with the software. At the end of this research, the relationship between the growth of the determined performance parameter with charging hours was analyzed and presented. The research results proved the ability of GraphPad Prism software to assess the temperature distribution in the TES tank and also the corresponding effects on the overall Tank performance. The software offers better advantages in evaluating the performance parameter of the TES tank accurately.

: Mistrust, amplified by numerous artificial intelligence (AI) related incidents, has caused the energy and industrial sectors to be amongst the slowest adopter of AI methods. Central to this issue is the black-box problem of AI, which impedes investments and fast becoming a legal hazard for users. Explainable AI (XAI) is a recent paradigm to tackle this challenge. Being the backbone of the industry, the prognostic and health management (PHM) domain has recently been introduced to XAI. However, many deficiencies, particularly lack of explanation assessment methods and uncertainty quantification, plague this young field. In this paper, we elaborate a framework on explainable anomaly detection and failure prognostic employing a Bayesian deep learning model to generate local and global explanations from the PHM tasks. An uncertainty measure of the Bayesian model is utilized as marker for anomalies expanding the prognostic explanation scope to include model’s confidence. Also, the global explanation is used to improve prognostic performance, an aspect neglected from the handful of PHM-XAI publications. The quality of the explanation is finally examined employing local accuracy and consistency properties. The method is tested on real-world gas turbine anomalies and synthetic turbofan data failure prediction. Seven out of eight of the tested anomalies were successfully identified. Additionally, the prognostic outcome showed 19% improvement in statistical terms and achieved the highest prognostic score amongst best published results on the topic.

Corrosion under insulation (CUI) is one of the increasing industrial problems, especially in chemical plants that have been running for an extended time. Prediction modeling, which is one of the solutions for this issue, has attracted increasing attention and has been considered for several industrial applications. The main objective of this work was to investigate the effect of combined data input in prediction modeling, which could be applied to improve the existing CUI rate prediction model. Experimental data and field historical data were gathered and simulated using an artificial neural network separately. To analyze the effect of data sources on the final corrosion rate under the insulation prediction model, both sources of data from experiment and field data were then combined and simulated again using an artificial neural network. Results exhibited the advantages of combined input data type from the experiment and field in the final prediction model. The model developed clearly shows the occurrence of corrosion by phases, which are uniform corrosion at the early phases and pitting corrosion at the later phases. The prediction model will enable better mitigation actions in preventing loss of containment due to CUI, which in turn will improve overall sustainability of the plant.

In this study, the viability of harnessing wind energy for offshore oil and gas (O&G) platforms as a micro-grid solution in low wind speed regions to power O&G platforms is explored. However, wind, as a renewable energy, is known to be highly variable and is unable to provide standalone power reliably within a micro-grid solution due to the variation of the wind speed at hub height, which subsequently leads to a variation of the power outcome. Here, a hybrid power generation concept is developed in which one gas turbine generator (GTG) is replaced with a floating horizontal axis wind turbine (WT) system. By setting up this system, the reduction of the maintenance costs of the GTGs and the reduction of fuel gas consumption reduces carbon dioxide (CO₂) emissions. In addition to this, the fuel gas savings in terms of the business side of such a solution provide a positive revenue impact. In this feasibility study, a technical framework is developed, followed by an economic framework. In the technical framework, wind assessments are performed to obtain the annual energy production for the selected field. Furthermore, an economic framework is established for both conventional and hybrid concepts in two scenarios: greenfield and brownfield, where the incremental net present value (NPV) and levelized cost of energy are calculated. The resultant difference in NPV for hybrid power generation compared to conventional power generation was found to be between 22% and 37%. The levelized cost of energy (LCOE) for WT is USD 165.52/MWh, which is 39% lower than for conventional, gas turbine-only operations. The LCOE for the hybrid approach is lower than for the conventional scenario by 22%. In conclusion, the hybrid micro-grid concept solution can harness wind energy from low wind regions with better economic benefits compared to conventional methods through the proper selection of the WT system, its floating substructure, and efficient micro-grid system for powering oil and gas platforms.