{"title":"Data selection framework for battery state of health related parameter estimation under system uncertainties","authors":"Jackson Fogelquist, Xinfan Lin","doi":"10.1016/j.etran.2023.100283","DOIUrl":null,"url":null,"abstract":"<div><p>Data selection is a practical technique for improving parameter estimation accuracy through the strategic selection of information-rich data for use in the estimation algorithm. Traditional selection criteria have been either heuristic or sensitivity-based, without consideration of uncertainties in measurement, model, or parameter. In this paper, we propose an uncertainty-aware data selection framework that selects data segments based on the potential of the ingrained data structures to mitigate the influence of system uncertainties on the estimation result. The framework comprises two components: the data quality rating and data selection algorithm. The data quality rating is a metric for evaluating the uncertainty-propagating data structures of a data segment, and the data selection algorithm automatically integrates the data selection into the estimation procedure. Furthermore, a novel adaptive approximation of model/measurement uncertainty is derived and implemented in the data quality rating formula to enhance performance in the presence of time-varying sensor bias/noise and unmodeled system dynamics. The framework is validated through an advanced battery management system application, where two lithium-ion battery health-related electrochemical parameters are separately estimated under random drive-cycle input data to emulate battery state of health monitoring for an electric vehicle. We show that the drive-cycle data, which are frequently used for battery state of health estimation as the only available data during battery operation, may not provide accurate estimation results due to the existence of large portions of low-quality data (low sensitivity and high uncertainty). By extracting the high-quality data segments, the data selection framework reduced experimental estimation errors by one order of magnitude when compared with the conventional approach of estimating without data selection.</p></div>","PeriodicalId":36355,"journal":{"name":"Etransportation","volume":null,"pages":null},"PeriodicalIF":15.0000,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Etransportation","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2590116823000589","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0
Abstract
Data selection is a practical technique for improving parameter estimation accuracy through the strategic selection of information-rich data for use in the estimation algorithm. Traditional selection criteria have been either heuristic or sensitivity-based, without consideration of uncertainties in measurement, model, or parameter. In this paper, we propose an uncertainty-aware data selection framework that selects data segments based on the potential of the ingrained data structures to mitigate the influence of system uncertainties on the estimation result. The framework comprises two components: the data quality rating and data selection algorithm. The data quality rating is a metric for evaluating the uncertainty-propagating data structures of a data segment, and the data selection algorithm automatically integrates the data selection into the estimation procedure. Furthermore, a novel adaptive approximation of model/measurement uncertainty is derived and implemented in the data quality rating formula to enhance performance in the presence of time-varying sensor bias/noise and unmodeled system dynamics. The framework is validated through an advanced battery management system application, where two lithium-ion battery health-related electrochemical parameters are separately estimated under random drive-cycle input data to emulate battery state of health monitoring for an electric vehicle. We show that the drive-cycle data, which are frequently used for battery state of health estimation as the only available data during battery operation, may not provide accurate estimation results due to the existence of large portions of low-quality data (low sensitivity and high uncertainty). By extracting the high-quality data segments, the data selection framework reduced experimental estimation errors by one order of magnitude when compared with the conventional approach of estimating without data selection.
期刊介绍:
eTransportation is a scholarly journal that aims to advance knowledge in the field of electric transportation. It focuses on all modes of transportation that utilize electricity as their primary source of energy, including electric vehicles, trains, ships, and aircraft. The journal covers all stages of research, development, and testing of new technologies, systems, and devices related to electrical transportation.
The journal welcomes the use of simulation and analysis tools at the system, transport, or device level. Its primary emphasis is on the study of the electrical and electronic aspects of transportation systems. However, it also considers research on mechanical parts or subsystems of vehicles if there is a clear interaction with electrical or electronic equipment.
Please note that this journal excludes other aspects such as sociological, political, regulatory, or environmental factors from its scope.