Jinlong Kang, C. Varnier, A. Mosallam, N. Zerhouni, Fares Ben Youssef, Na-Na Shen
{"title":"Risk Level Estimation for Electronics Boards in Drilling and Measurement Tools Based on the Hidden Markov Model","authors":"Jinlong Kang, C. Varnier, A. Mosallam, N. Zerhouni, Fares Ben Youssef, Na-Na Shen","doi":"10.1109/PHM2022-London52454.2022.00093","DOIUrl":null,"url":null,"abstract":"The electronic boards in drilling and measurement (D&M) tools provide multiple functions, such as data acquisition, signal processing, operation control, and data storage. However, due to the harsh downhole operating conditions; i.e., high temperature, dynamic vibration, and extensive shocks, the boards are likely to suffer from complex failure modes and result in failed jobs. Estimating the risk level of the boards can tolerate and provide support for maintenance decision making and job planning, this paper presents a statistical method for risk assessment of the electronic boards. The method first selects relevant channels from D&M tool measurement data and extracts histogram features based on those selected channels. The histogram features are then enhanced based on a linear interpolation method and aggregated using weighted sum. Finally, hidden Markov models (HMMs) with different parameter settings are trained using the processed features. The best HMM is chosen according to the Akaike information criterion and Bayesian information criterion. The proposed HMM-based method is tested on a real-world data set of failed control processing unit boards that were assembled for a specific D&M tool. The experimental results show that this method is effective in estimating the risks as a sequence of events, which in turn, helps to achieve consistent risk estimation. The work presented in this paper is also part of a long-term project with the aim to construct a risk-based decision advisor for D&M tools used in the oil and gas industry.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 Prognostics and Health Management Conference (PHM-2022 London)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PHM2022-London52454.2022.00093","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
The electronic boards in drilling and measurement (D&M) tools provide multiple functions, such as data acquisition, signal processing, operation control, and data storage. However, due to the harsh downhole operating conditions; i.e., high temperature, dynamic vibration, and extensive shocks, the boards are likely to suffer from complex failure modes and result in failed jobs. Estimating the risk level of the boards can tolerate and provide support for maintenance decision making and job planning, this paper presents a statistical method for risk assessment of the electronic boards. The method first selects relevant channels from D&M tool measurement data and extracts histogram features based on those selected channels. The histogram features are then enhanced based on a linear interpolation method and aggregated using weighted sum. Finally, hidden Markov models (HMMs) with different parameter settings are trained using the processed features. The best HMM is chosen according to the Akaike information criterion and Bayesian information criterion. The proposed HMM-based method is tested on a real-world data set of failed control processing unit boards that were assembled for a specific D&M tool. The experimental results show that this method is effective in estimating the risks as a sequence of events, which in turn, helps to achieve consistent risk estimation. The work presented in this paper is also part of a long-term project with the aim to construct a risk-based decision advisor for D&M tools used in the oil and gas industry.