Feature Analyses and Modeling of Lithium-Ion Battery Manufacturing Based on Random Forest Classification

Liu, Kailong; Hu, Xiaosong; Zhou, Huiyu; Tong, Lei; Widanalage, Dhammika; Marco, James

doi:10.1109/tmech.2020.3049046

Cited by 138 publications

(76 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, we have quantitatively validated the applicability of structured regressors to the estimation of LIB energy and power. Future work to be considered includes analyzing feature and parameter sensitivities [50] and the use of ML explainability/interpretability methods to ensemble models from an electrochemical perspective. Furthermore, model and data uncertainty management [51,52] should also be analyzed: the quantification of the risk from different sources is valuable for the real-world applicability of the developed methodology.…”

Section: Discussionmentioning

confidence: 99%

Ensemble Surrogate Models for Fast LIB Performance Predictions

et al. 2021

View full text Add to dashboard Cite

Battery Cell design and control have been widely explored through modeling and simulation. On the one hand, Doyle’s pseudo-two-dimensional (P2D) model and Single Particle Models are among the most popular electrochemical models capable of predicting battery performance and therefore guiding cell characterization. On the other hand, empirical models obtained, for example, by Machine Learning (ML) methods represent a simpler and computationally more efficient complement to electrochemical models and have been widely used for Battery Management System (BMS) control purposes. This article proposes ML-based ensemble models to be used for the estimation of the performance of an LIB cell across a wide range of input material characteristics and parameters and evaluates 1. Deep Learning ensembles for simulation convergence classification and 2. structured regressors for battery energy and power predictions. The results represent an improvement on state-of-the-art LIB surrogate models and indicate that deep ensembles represent a promising direction for battery modeling and design.

show abstract

Section: Discussionmentioning

confidence: 99%

Ensemble Surrogate Models for Fast LIB Performance Predictions

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Each of these individual ROMs had n = 5 states. All states are initialized to 0; the covariance matrix is initialized to a diagonal matrix having a value of 10 in every diagonal element, except for the element corresponding to the integration state which had a value of 6 . The process-noise covariance was chosen to be Σw = 0.4 and the sensor-noise covariance was chosen to be Σṽ = 0.1.…”

Section: B Sigma Point Kalman Filter Initializationmentioning

confidence: 99%

“…ROMs can be used in this context to feed AI algorithms efficiently, or to give feedback to the AI algorithms, while running in the BMS. Also AI techniques could be used to aid handling uncertainty in the predictions [5] or in model parameters [6].…”

Section: Introductionmentioning

confidence: 99%

Electrochemical Model and Sigma Point Kalman Filter Based Online Oriented Battery Model

et al. 2021

View full text Add to dashboard Cite

show abstract

“…RF was introduced in 2014 by Breiman [6]. RF had been recently used in various fields including rain prediction [4], drug resistance prediction [5], geospatial pattern alaysis [7], prediction of electricity production [8], battery modeling [9], and landslide susceptibility mapping [10].…”

Section: -03mentioning

confidence: 99%

Automatic Complaints Categorization Using Random Forest and Gradient Boosting

Anwar

Pratiwi

Udhayana

2021

ASSET

View full text Add to dashboard Cite

Capturing and responding to complaints from the public is an important effort to develop a good city/country. This project aims to utilize Data Mining to automatize complaints categorization. More than 35,000 complaints in Bangalore city, India, were retrieved from the “I Change My City” website (https://www.ichangemycity.com). The vector space of the complaints was created using Term Frequency–Inverse Document Frequency (TF-IDF) and the multi-class text classifications were done using Random Forest (RF) and Gradient Boosting (GB). Results showed that both RF and GB have similar performance with an accuracy of 73% on the 10-classes multi-class classification task. Result also showed that the model is highly dependent on the word usage in the complaint's description. Future research directions to increase task performance are also suggested.

show abstract

Feature Analyses and Modeling of Lithium-Ion Battery Manufacturing Based on Random Forest Classification

Abstract: Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

Cited by 138 publications

References 36 publications

Ensemble Surrogate Models for Fast LIB Performance Predictions

Ensemble Surrogate Models for Fast LIB Performance Predictions

Electrochemical Model and Sigma Point Kalman Filter Based Online Oriented Battery Model

Automatic Complaints Categorization Using Random Forest and Gradient Boosting

Contact Info

Product

Resources

About