Convolutional Neural Network-Based False Battery Data Detection and Classification for Battery Energy Storage Systems

Lee, Hyun Jun; Kim, Kyoung-Tak; Park, Joung‐Hu; Bere, Gomanth; Ochoa, Justin J.; Kim, Taesic

doi:10.1109/tec.2021.3061493

Cited by 49 publications

(18 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then the SOC estimation methods based on recurrent neural networks, such as: recurrent neural network (RNN), long short term memory (LSTM), and gated recurrent unit (GRU), have attracted the attention of many scholars [3] , it not only have good interpretability in theory, but also shows a better application in SOC, but when a sudden voltage change caused by large currents occurs, the accuracy of the RNN will decrease significantly. For this reason, this paper proposes a SOC estimation method based on an optimized deep convolutional neural networks (CNN) [4][5][6] . The average-pooling layer can reduce the impact of input mutation on SOC estimation, and CNN training requires fewer parameters training, which means a better practical applicability.…”

Section: Introductionmentioning

confidence: 99%

Optimized Deep Convolutional Neural Networks Based State of Charge Estimation for Lithium-Ion Battery

Qiao¹,

Wang²,

Ye³

et al. 2022

Preprint

View full text Add to dashboard Cite

To address the problem of the accuracy decrease of state of charge estimation caused by sudden high current impact, this paper proposes a lithium-ion battery SOC estimation method based on optimized deep convolutional neural network. Firstly, the 18650 battery was tested under actual driving conditions to obtain experimental data, and the experimental data was preprocessed by moving window to fit the two dimensional convolutional neural networks. Secondly, the proposed method was trained and tested, and the model parameters were further optimized. Thirdly, the proposed method is compared with sequence-tosequence methods such as long short term memory and gated recurrent unit, and the results verify the superiority of the proposed method. This article provides a method to the battery SOC estimation, which is more conducive to practical applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Optimized Deep Convolutional Neural Networks Based State of Charge Estimation for Lithium-Ion Battery

Qiao¹,

Wang²,

Ye³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, there are many technical and operational limits for ESSs. For example, there have been problems with modeling batteries used in ESSs, which causes over-charging and over-voltage problems, and this may result in causing severe explosion and fire [26,27]. Therefore, since ESSs may not provide PFR whenever required, PFR is still mainly provided by GUs, especially SGs.…”

Section: Introductionmentioning

confidence: 99%

A Study on Frequency Stability and Primary Frequency Response of the Korean Electric Power System Considering the High Penetration of Wind Power

et al. 2022

View full text Add to dashboard Cite

The high penetration of wind power decreases the system inertia and primary frequency reserve while replacing the conventional synchronous generators (SGs). Therefore, if the system operator does not take appropriate action on the remaining generation units (GUs) operation, high penetration of wind power will aggravate the frequency stability. To solve this problem, wind power plants (WPPs) may provide the inertial response and primary frequency response (PFR) to support the frequency stability. However, due to the variability of renewable energy, WPPs may not provide adequate frequency response whenever it is required. This paper proposes an algorithm to determine the operation of GUs to provide appropriate PFR for a power system with high penetration of wind power. Through the proposed algorithm, it calculates the required PFR to restore the decreased frequency stability caused by the high penetration of wind power. Then, while considering the available PFR from WPPs, it redetermines the droop coefficient of SGs governor to provide the sufficient PFR to recover the frequency stability. Finally, the effectiveness of the proposed algorithm is verified on the practical Korean electric power system.

show abstract

“…Sensor fault detection and diagnosis (SFDD) methods can be broadly divided into data-driven and model-based methods (Reppa et al, 2015;Lee et al, 2021). The model-based methods are usually easy to integrate into control systems, but they also need to set complex thresholds, and the modelbased methods are more difficult to apply in different fields, especially for a nonlinear system since the fault models are more difficult to establish (Kou et al, 2020;Wang et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, the datadriven methods do not rely on mathematical models and have attracted attention of many scholars (Ojo et al, 2021). Lee et al (2021) proposed a convolutional neural network (CNN)-based FDD method for battery energy storage systems to detect and classify false battery sensor data. Ojo et al (2021) proposed a long short-term memory recurrent neural network (LSTM-RNN)-based thermal fault diagnosis method for lithium-ion batteries, which is an easy-to-implement way and does not need to pay attention to the complex mathematical modeling and parameters of battery physics.…”

Section: Introductionmentioning

confidence: 99%

A Random Forest and Current Fault Texture Feature–Based Method for Current Sensor Fault Diagnosis in Three-Phase PWM VSR

et al. 2021

View full text Add to dashboard Cite

Three-phase PWM voltage-source rectifier (VSR) systems have been widely used in various energy conversion systems, where current sensors are the key component for state monitoring and system control. The current sensor faults may bring hidden danger or damage to the whole system; therefore, this paper proposed a random forest (RF) and current fault texture feature–based method for current sensor fault diagnosis in three-phase PWM VSR systems. First, the three-phase alternating currents (ACs) of the three-phase PWM VSR are collected to extract the current fault texture features, and no additional hardware sensors are needed to avoid causing additional unstable factors. Then, the current fault texture features are adopted to train the random forest current sensor fault detection and diagnosis (CSFDD) classifier, which is a data-driven CSFDD classifier. Finally, the effectiveness of the proposed method is verified by simulation experiments. The result shows that the current sensor faults can be detected and located successfully and that it can effectively provide fault locations for maintenance personnel to keep the stable operation of the whole system.

show abstract

Convolutional Neural Network-Based False Battery Data Detection and Classification for Battery Energy Storage Systems

Cited by 49 publications

References 36 publications

Optimized Deep Convolutional Neural Networks Based State of Charge Estimation for Lithium-Ion Battery

Optimized Deep Convolutional Neural Networks Based State of Charge Estimation for Lithium-Ion Battery

A Study on Frequency Stability and Primary Frequency Response of the Korean Electric Power System Considering the High Penetration of Wind Power

A Random Forest and Current Fault Texture Feature–Based Method for Current Sensor Fault Diagnosis in Three-Phase PWM VSR

Contact Info

Product

Resources

About