A Battery Digital Twin From Laboratory Data Using Wavelet Analysis and Neural Networks

IEEE Transactions on Industrial Informatics

Published On 2024/1/26

Lithium-ion (Li-ion) batteries are the preferred choice for energy storage applications. Li-ion performances degrade with time and usage, leading to a decreased total charge capacity and to an increased internal resistance. In this article, the wavelet analysis is used to filter the voltage and current signals of the battery to estimate the internal complex impedance as a function of state of charge (SoC) and state of health (SoH). The collected data are then used to synthesize a battery digital twin (BDT). This BDT outputs a realistic voltage signal as a function of SoC and SoH inputs. The BDT is based on feedforward neural networks trained to simulate the complex internal impedance and the open-circuit voltage generator. The effectiveness of the proposed method is verified on the dataset from the prognostics data repository of NASA.

Journal

IEEE Transactions on Industrial Informatics

Published On

2024/1/26

Authors

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Position

Professor at

H-Index(all)

104

H-Index(since 2020)

72

I-10 Index(all)

0

I-10 Index(since 2020)

0

Citation(all)

0

Citation(since 2020)

0

Cited By

0

Research Interests

Power Electronics

Smart Batteries

AI

University Profile Page

Pallavi Bharadwaj

Pallavi Bharadwaj

Massachusetts Institute of Technology

Position

Postdoctoral Research Associate

H-Index(all)

10

H-Index(since 2020)

9

I-10 Index(all)

0

I-10 Index(since 2020)

0

Citation(all)

0

Citation(since 2020)

0

Cited By

0

Research Interests

Smart power electronics

Green energy optimization

Net zero transition

Other Articles from authors

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

IEEE Transactions on Industrial Electronics

Health Prediction for Lithium-Ion Batteries Under Unseen Working Conditions

Battery health prediction is significant while challenging for intelligent battery management. This article proposes a general framework for both short-term and long-term predictions of battery health under unseen dynamic loading and temperature conditions using domain-adaptive multitask learning (MTL) with long-term regularization. First, features extracted from partial charging curves are utilized for short-term state of health predictions. Then, the long-term degradation trajectory is directly predicted by recursively using the predicted features within the multitask framework, enhancing the model integrity and lowering the complexity. Then, domain adaptation (DA) is adopted to reduce the discrepancies between different working conditions. Additionally, a long-term regularization is introduced to address the shortcoming that arises when the model is extrapolated recursively for future health predictions. Thus, the short …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

IEEE/ASME Transactions on Mechatronics

Online Sensorless Temperature Estimation of Lithium-Ion Batteries Through Electro-Thermal Coupling

Owing to the nonnegligible impacts of temperature on the safety, performance, and lifespan of lithium-ion batteries, it is essential to regulate battery temperature to an optimal range. Temperature monitoring plays a fundamental role in battery thermal management, yet it is still challenged by limited onboard temperature sensors, particularly in large-scale battery applications. As such, developing sensorless temperature estimation is of paramount importance to acquiring the temperature information of each cell in a battery system. This article proposes an estimation approach to obtain the cell temperature by taking advantage of the electrothermal coupling effect of batteries. An electrothermal coupled model, which captures the interactions between the electrical and the thermal dynamics, is established, parameterized, and experimentally validated. A closed-loop observer is then designed based on this coupled model …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Artificial Intelligence-Based State-of-Health Estimation of Lithium-Ion Batteries

The State of Health (SOH) estimation for automotive batteries is currently assessed with different techniques which may involve long testing procedure or require costly hardware to be implemented. This paper aims at contributing to this domain by exploiting the response of a lead-acid battery with respect to a short-term current profile using an Artificial Neural Network (ANN) classifier for SOH estimation. The method is applicable onboard the vehicle and no additional instrumentation is required on the retained vehicle. The design and validation of a SOH method with a short-term current profile using Artificial Intelligence (AI) in lead-acid batteries, which are commonly used in heavy-duty vehicles for cranking and cabin systems, are presented. The paper validates the considered approach with experimental data, which are representative of actual vehicle operations. In detail, the paper describes the retained …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

arXiv preprint arXiv:2402.07777

Novel Low-Complexity Model Development for Li-ion Cells Using Online Impedance Measurement

Modeling of Li-ion cells is used in battery management systems (BMS) to determine key states such as state-of-charge (SoC), state-of-health (SoH), etc. Accurate models are also useful in developing a cell-level digital-twin that can be used for protection and diagnostics in the BMS. In this paper, a low-complexity model development is proposed based on the equivalent circuit model (ECM) of the Li-ion cells. The proposed approach uses online impedance measurement at discrete frequencies to derive the ECM that matches closely with the results from the electro-impedance spectroscopy (EIS). The proposed method is suitable to be implemented in a microcontroller with low-computational power, typically used in BMS. Practical design guidelines are proposed to ensure fast and accurate model development. Using the proposed method to enhance the functions of a typical automotive BMS is described. Experimental validation is performed using large prismatic cells and small-capacity cylindrical cells. Root-mean-square error (RMSE) of less than 3\% is observed for a wide variation of operating conditions.

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

IEEE Transactions on Industry Applications

Small-Sample-Learning-Based Lithium-Ion Batteries Health Assessment: An Optimized Ensemble Framework

Machine Learning is widely studied in battery state of health (SOH) estimation due to its advantage in establishing the non-linear mapping between measurements and SOH. However, the requirement of a big dataset and the lack of robustness limit the practical application, especially in small sample learning. To tackle these challenges, an optimal ensemble framework called BaggELM (bagging extreme learning machine) is proposed for battery SOH estimation. Specifically, the required dataset is reduced by optimizing the input voltage and the hyperparameters of the BaggELM algorithm. Moreover, a statistical post-processing method is used to aggregate multiple ELMs, and the final estimate is determined by the maximum probability density value. As a result, the effects of random parameterization of ELM and the training data size on SOH estimation are suppressed, thus improving the robustness and accuracy of …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Intelligent Cell Balancing Control for Lithium-Ion Battery Packs

This study introduces a balancing control strategy that employs an Artificial Neural Network (ANN) to ensure State of Charge (SOC) balance across lithium-ion (Li-ion) battery packs, consistent with the framework of smart battery packs. The model targets a battery pack consisting of cells with diverse characteristics, reflecting real-world heterogeneous conditions. A fundamental aspect of this approach is the ability to bypass individual cells optimally. This key feature stops current flow to and from the cell, allowing it to rest and cool off while avoiding charging or discharging cycles. The implementation of ANN enables adaptive and dynamic management of SOC, which is essential for optimizing performance and extending the lifespan of battery packs. The results demonstrate the effectiveness of the proposed ANN-based balancing strategy in SOC balancing, demonstrating its potential as a critical solution in enhancing battery management systems for electric vehicles.

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Grid Impedance Shaping for Grid-Forming Inverters: A Soft Actor-Critic Deep Reinforcement Learning Algorithm

This paper proposed an advanced method for adjusting grid impedance in grid-forming inverters, utilizing the Soft Actor-Critic Deep Reinforcement Learning (SAC-DRL) algorithm. The approach contains a flexible strategy for controlling virtual impedance, supported by an equivalent grid impedance estimator. This facilitates accurate modifications of virtual impedance based on the grid’s X/R ratio and the converter’s power capacity, aiming to optimize power flow and maintain grid stability. A unique feature of this methodology is the division of virtual reactance into two segments: one adhering to standard control protocols and the other designated for precision enhancement via the SAC-DRL method. This strategy introduces a layer of intelligence to the system, strengthening its resilience against fluctuations in grid impedance. Experimental validations, executed on a laboratory setup, verify the robustness of this approach, highlighting its potential to significantly improve intelligent power grid management practices.

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Electric vehicle battery charging strategy

As a key enabler for transportation electrification and a contributor toward the net-zero carbon future, battery plays a pivotal role in determining the energy management performance of electric vehicles. Technical challenges facing the development of advanced automotive battery charging arise from various contradictory objectives, immeasurable internal states, and hard constraints. This chapter presents a critical introduction to the state-of-the-art charging strategies for the electric vehicle battery and their key enabling technologies. Specifically, battery charging solutions for electric vehicles are first classified and discussed. Then, the battery models on which these solutions rest are stated, the related charging frameworks are summarized, and the advantages and drawbacks of the adopted technologies are discussed. Suggestions for overcoming the limitations of the discussed charging strategies are proposed …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Thermal state monitoring of lithium-ion batteries: Progress, challenges, and opportunities

Transportation electrification is a promising solution to meet the ever-rising energy demand and realize sustainable development. Lithium-ion batteries, being the most predominant energy storage devices, directly affect the safety, comfort, driving range, and reliability of many electric mobilities. Nevertheless, thermal-related issues of batteries such as potential thermal runaway, performance degradation at low temperatures, and accelerated aging still hinder the wider adoption of electric mobilities. To ensure safe, efficient, and reliable operations of lithium-ion batteries, monitoring their thermal states is critical to safety protection, performance optimization, as well as prognostics, and health management. Given insufficient onboard temperature sensors and their inability to measure battery internal temperature, accurate and timely temperature estimation is of particular importance to thermal state monitoring. Toward …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Reliability Engineering & System Safety

Predictive health assessment for lithium-ion batteries with probabilistic degradation prediction and accelerating aging detection

Predictive health assessment is of vital importance for smarter battery management to ensure optimal and safe operations and thus make the most use of battery life. This paper proposes a general framework for battery aging prognostics in order to provide the predictions of battery knee, lifetime, state of health degradation, and aging rate variations, as well as the assessment of battery health. Early information is used to predict knee slope and other life-related information via deep multi-task learning, where the convolutional-long-short-term memory-bayesian neural network is proposed. The structure is also used for online state of health and degradation rate predictions for the detection of accelerating aging. The two probabilistic predicted boundaries identify the accelerating aging regions for battery health assessment. To avoid wrong and premature alarms, the empirical model is used for data preprocessing and …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Fractional-order control techniques for renewable energy and energy-storage-integrated power systems: A review

The worldwide energy revolution has accelerated the utilization of demand-side manageable energy systems such as wind turbines, photovoltaic panels, electric vehicles, and energy storage systems in order to deal with the growing energy crisis and greenhouse emissions. The control system of renewable energy units and energy storage systems has a high effect on their performance and absolutely on the efficiency of the total power network. Classical controllers are based on integer-order differentiation and integration, while the fractional-order controller has tremendous potential to change the order for better modeling and controlling the system. This paper presents a comprehensive review of the energy system of renewable energy units and energy storage devices. Various papers are evaluated, and their methods and results are presented. Moreover, the mathematical fundamentals of the fractional-order method are mentioned, and the various studies are categorized based on different parameters. Various definitions for fractional-order calculus are also explained using their mathematical formula. Different studies and numerical evaluations present appropriate efficiency and accuracy of the fractional-order techniques for estimating, controlling, and improving the performance of energy systems in various operational conditions so that the average error of the fractional-order methods is considerably lower than other ones.

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

IEEE Transactions on Vehicular Technology

Battery states monitoring for electric vehicles based on transferred multi-task learning

State/temperature monitoring is one of the key requirements of battery management systems that facilitates efficient and intelligent management to ensure the safe operation of batteries in electrified transportation. This paper proposes an online end-to-end state monitoring method based on transferred multi-task learning. Measurement data is directly used for sharing information generation with the convolutional neural network. Then, the multiple task-specific layers are added for state/temperature monitoring. The transfer learning strategy is designed to improve accuracy further under various application scenarios. Experiments under different working profiles, temperatures, and aging conditions are conducted to evaluate the method, which covers the wide usage ranges in electric vehicles. Comparisons with several benchmarks illustrate the superiority of the proposed method with better accuracy and …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Battery aging behavior evaluation under variable and constant temperatures with real loading profiles

Studying and analyzing battery aging behavior is crucial in battery health prognostic and management. This paper conducts novel and comprehensive experiments to evaluate battery aging under variable external stresses, including different dynamic load profiles and variable environmental temperatures. Respond analysis in the time and frequency domain is performed to account for the different aging rates under different current loadings, where the statistic calculation and fast Fourier transform are used for the analysis. The empirical model is used to fit the fade curve for the comparisons between constant and variable temperatures. The capacity decrease and internal resistance increase are extracted to evaluate capacity and power fade, respectively. The experimental results show that the urban dynamic operating conditions help to prolong the service life compared to the constant current aging case. In contrast …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Small Signal Model of Modular Multilevel Converter with Power Synchronization Control

Power synchronization control (PSC) is one of the popular control schemes in grid-forming control-based converters because it simulates the grid support capability of conventional synchronous generators. However, prior research is based on two-level converters which do not have complex internal circuits, and whether PSC can be directly applied to the modular multilevel converter (MMC) topology since MMC has sub-module capacitor voltage ripples and inherent second harmonic circulating current algorithm, has not been analyzed. This paper establishes the small signal model of MMC with PSC considering the MMC internal dynamic and circulating current suppression control (CCSC). The power oscillation phenomenon when grid short-circuit ratio (SCR) increases is also demonstrated with the closed-loop system eigenvalues calculation and verified with the experimental results.

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Hyperparameter optimization in bagging-based ELM algorithm for lithium-ion battery state of health estimation

Artificial neural networks are widely studied for the state of health (SOH) estimation of Lithium-ion batteries because they can recognize global features from the raw data and are able to cope with multi-dimensional data. But the performance of the model depends to some extent on the selection of the hyperparameters, which remain constant during model training. To improve the generalization performance as well as accuracy, an ensemble learning framework is proposed for battery SOH estimation, where multiple extreme learning machines are trained combined with bagging technology. The numbers of bags and neurons of the base model are then tuned by five commonly used hyperparameter optimization methods. Moreover, the SOH value with maximum probability density is selected as the output estimate to further improve the estimation accuracy. Finally, experimental results on both NMC and LPF batteries …

Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Ieee Transactions on Industrial Electronics

Lithium-ion battery lifetime extension with positive pulsed current charging

Some studies verified that the pulsed current charging technique could extend the battery lifetime and improve the charging performance of lithium-ion batteries. However, some researchers are skeptical of this opinion because their studies have not found any advantages of pulsed current charging over conventional constant current (CC) charging. Positive pulsed current (PPC), the most common pulsed current mode, was selected for the investigation in this work. The effect of the PPC with various parameters, including the duty cycle, amplitude, and frequency, on the performance and lifetime of lithium-ion batteries, are investigated by experiments. According to the experimental results, the charging speed, charging capacity, and maximum rising temperature are mainly determined by the duty cycle and amplitude of the PPC. The battery lifetime with PPC under the frequency range from 0.05 Hz to 2 kHz has been …

Pallavi Bharadwaj

Pallavi Bharadwaj

Massachusetts Institute of Technology

IEEE Journal of Emerging and Selected Topics in Industrial Electronics

Incorporating Uncertainty and Reliability for Battery Temperature Prediction using Machine Learning Methods

Temperature prediction of lithium-ion batteries is essential to prevent aging and degradation of batteries while ensuring safe and reliable operation.The use of simple machine learning methods for battery temperature prediction given current and voltage inputs is challenging due to battery cycling-induced aging. Since these methods do not generalize to the data outside the training domain, reliability issue arises in battery thermal management. As it is tough to train the battery temperature prediction models with all the possible data the battery is expected to see during real-world usage, uncertainty-aware models are needed. To address this, instead of point prediction, range prediction is performed using the conformal prediction method. The proposed method provides a band of temperature predictions for the input of current, voltage, and operational time. Results show high prediction accuracy with 79% of actual …

2023/10/23

Article Details
Remus Teodorescu

Remus Teodorescu

Aalborg Universitet

Energy Conversion and Management

Spatial–temporal data-driven full driving cycle prediction for optimal energy management of battery/supercapacitor electric vehicles

For multi-energy storage vehicles, the performance of online predictive energy management strategies largely relies on the length and effective utilization of predictive information. In this context, this paper proposes a novel velocity prediction method for the full driving cycle of electric vehicles based on the spatial–temporal commuting data, then the predicted velocity is applied to predictive energy management in electric vehicles with battery/supercapacitor hybrid energy storage system. Firstly, an one-year real-world commuting data set is collected on a Chinese arterial road with 10 intersections, 225 records are classified into 79 categories. Then, a real-time two-stage full driving cycle prediction method is proposed, where a medium-term prediction based on a long–short term memory (LSTM) network and a long-term prediction generated by a spatial–temporal interpolation method (STIM) are spliced for each …

Other articles from IEEE Transactions on Industrial Informatics journal

Shuping He

Shuping He

Anhui University

IEEE Transactions on Industrial Informatics

Distributed Asynchronous Optimization of Multiagent Systems: Convergence Analysis and Its Application

This article focuses on solving a distributed convex optimization problem of multiagent systems with multiple inequality constraints. Considering communications between agents are prone to failures and not synchronized among themselves in some cases, a novel asynchronous adaptive step sizes-DIGing algorithm is proposed through integrating the projection operators and logic- and framework. In specific, a bilaterally adjustable adaptive step size mechanism is introduced to automatically abandon the irrational evolutionary route which relieves the limitations of traditional DIGing algorithm. By adopting the operator theory, the almost sure convergence of the proposed algorithm under asynchronous communication is proved. Finally, the theoretical and simulation results for the plantwide optimization problem in the ethylene production process illustrate the effectiveness of the proposed algorithm.

Hao Sheng

Hao Sheng

Beihang University

IEEE Transactions on Industrial Informatics

Distributed Collaborative Object Retrieval With Blockchain-Based Edge Computing

In the current industrial informatics society, the numerous cameras deployed in the modern city promote the development of various video services, such as security monitoring and object retrieval. However, traditional methods encounter data leakage risks. Some camera owners are reluctant to share their data since the video contains confidential information. Meanwhile, domain diversities between cameras bring obstacles to practical object retrieval applications. To deal with these dilemmas, we propose a blockchain-based collaborative object retrieval (BCOR) system that can protect privacy as much as possible. BCOR includes two core components: multicamera reidentification framework (MC-ReF) and multicamera collaborative chain (M2C-Chain). Specifically, MC-ReF leverages visual relevance attention net (VRANet) to distinguish object identities in edge nodes. Through domain adaptation gradient …

Jingwen Wei

Jingwen Wei

University of Science and Technology of China

IEEE Transactions on Industrial Informatics

Optimal Charging of Lithium-Ion Battery Using Distributionally Robust Model Predictive Control With Wasserstein Metric

Developing a fast and safe charging strategy has been one of the key breakthrough points in lithium battery development owing to its range anxiety and long charging time. The majority of current model-based charging strategies are developed for deterministic systems. Real battery dynamics are, however, affected by model mismatches and process uncertainties, which may lead to constraint violations and even premature aging. This article proposes a fast charging scheme based on distributionally robust model predictive control (DRMPC) against uncertainty. Specifically, a coupled electrothermal-aging model is first introduced to describe the battery behavior, and electrothermal parameters of the adopted model are identified online based on the recursive least-squares algorithm. Subsequently, an online DRMPC-based charging framework is proposed, utilizing the Wasserstein ball centered on the empirical …

Abdollah Kavousi Fard

Abdollah Kavousi Fard

Shiraz University of Technology

IEEE Transactions on Industrial Informatics

A Blockchain-Based Mutual Authentication Method to Secure the Electric Vehicles' TPMS

Despite the widespread use of radio frequency identification and wireless connectivity such as near field communication in electric vehicles, their security and privacy implications in Ad-Hoc networks have not been well explored. This article provides a data protection assessment of radio frequency electronic system in the tire pressure monitoring system (TPMS). It is demonstrated that eavesdropping is completely feasible from a passing car, at an approximate distance up to 50 m. Furthermore, our reverse analysis shows that the static n -bit signatures and messaging can be eavesdropped from a relatively far distance, raising privacy concerns as a vehicles’ movements can be tracked by using the unique IDs of tire pressure sensors. Unfortunately, current protocols do not use authentication, and automobile technologies hardly follow routine message confirmation so sensor messages may be spoofed remotely. To …

Guoqing Xiao

Guoqing Xiao

Hunan University

IEEE Transactions on Industrial Informatics

A Black-Box Attack Algorithm Targeting Unlabeled Industrial AI Systems With Contrastive Learning

Adversarial attack algorithms are useful for testing and improving the robustness of industrial AI models. However, attacking black-box models with limited queries and unknown real labels remains a significant challenge. To overcome this challenge, we propose using contrastive learning to train a generated substitute model called attack contrastive learning network (ACL-Net) to attack black-box models with very few queries and no real labels. ACL-Net achieves end-to-end contrastive learning during training without labels, which differs from previous contrastive learning methods that required separate training for the classification layer with labels. We improve ACL-Net's robustness by using adversarial examples to train it during the attack stage. This approach results in more effective adversarial examples generated by ACL-Net. We conducted extensive experiments to validate the effectiveness of ACL-Net …

Wei Xiang

Wei Xiang

La Trobe University

IEEE Transactions on Industrial Informatics

Advanced Manufacturing in Industry 5.0: A Survey of Key Enabling Technologies and Future Trends

A revolution in advanced manufacturing has been driven by digital technology in the fourth industrial revolution, also known as Industry 4.0, and has resulted in a substantial increase in profits for the industry. In a new paradigm of Industry 5.0, advanced manufacturing will step further and be capable of offering customized products and a better user experience. A number of key enabling technologies are expected to play crucial roles in assisting Industry 5.0 in meeting higher demands of data acquisition and processing, communications, and collaborative robots in the advanced manufacturing process. The aim of this survey is to provide novel insights into advanced manufacturing in Industry 5.0 by summarizing the latest progress of key enabling technologies, such as artificial intelligence of things (AIoT), beyond 5G communications, and collaborative robotics. Finally, key directions for future research to enable this …

Matthew M.Y. Kuo

Matthew M.Y. Kuo

Auckland University of Technology

IEEE Transactions on Industrial Informatics

Actively Detecting Multiscale Flooding Attacks & Attack Volumes in Resource-Constrained ICPS

The significant growth in modern communication technologies has led to an increase in zero-day vulnerabilities that degrade the performance of industrial cyber-physical systems (ICPS). Distributed denial of service (DDoS) attacks are one such threat that overwhelms a target with floods of packets, posing a severe risk to the normal operations of the ICPS. Current solutions to detect DDoS attacks are unsuitable for resource-constrained ICPS. This study proposes actively detecting multiscale flooding DDoS attacks in resource-constrained ICPS by analyzing network traffic in the frequency domain. A two-phased technique detects attack presence and attack volume. Both phases use a novel combination of light-weight and theoretically sound statistical methods. The effectiveness of the proposed technique is evaluated using mainstream metrics like true and false positive rates, accuracy, and precision using BOUN …

Alois Knoll

Alois Knoll

Technische Universität München

IEEE Transactions on Industrial Informatics

Enhancing Local–Global Representation Learning for Image Restoration

Vision systems are the core element in industrial systems, such as intelligent transportation systems and inspection robots. However, undesired degradations caused by bad weather or low-end devices reduce the visibility of images. Image restoration aims to reconstruct a sharp image from a degraded counterpart and plays an important role in industrial systems. Recent transformer-based architectures leverage the self-attention unit and convolutions to model long-range dependencies and local connectivity, respectively, achieving promising performance for image restoration. However, these methods have quadratic complexity with respect to the input size. In addition, convolution operators are ineffective enough to recover the local details. This article presents a joint local and global representation learning framework for image restoration, called LoGoNet. Specifically, to enhance global contexts, we excavate the …

Changxin Liu

Changxin Liu

Northeastern University, China

IEEE Transactions on Industrial Informatics

Fast Sensitivity-Analysis-Based Online Self-Organizing Broad Learning System

Modern industrial process modeling requires models to adapt quickly to real-time operating conditions. To this end, this article proposes a fast sensitivity analysis (SA)-based self-organizing broad learning system (SASO-BLS) that offers a paradigm for theory-guided online structural self-adaptation of differentiable models. Specifically, SASO-BLS is implemented by embedding in BLS an interpretable and efficient model compression method called fast partial differential-based SA (FPD-SA). Unlike conventional SA methods that require iterative evaluation of SA indexes for all samples, FPD-SA exploits the deduced chain rule across categories, effectively mitigating the computational burden imposed by industrial Big Data and reducing computational time. In addition, we derive the offline SASO-BLS algorithm for discrete data and extend it to the online scenario for real-time streaming data processing. Note that both …

Mourad Debbabi

Mourad Debbabi

Concordia University

IEEE Transactions on Industrial Informatics

Spatial-Temporal Data-Driven Model for Load Altering Attack Detection in Smart Power Distribution Networks

The widespread deployment of information and communication technologies in smart power distribution networks (SPDNs) exposes them to cyber threats. Among different types of cyber-attacks in such ICT-based SPDNs, load-altering attacks (LAAs) against high-wattage devices have received significant attention in recent years. In this context, this article proposes a data-driven detection model tailored for identifying and localizing LAAs in SPDNs. In this pursuit, first, the graph structure of an SPDN, which is obtained from the grid topology, and node features, i.e., measurements of the load's power, are fed to a graph attention network (GAT), and the spatial correlations among the nodes are captured. Alongside, the temporal correlations are captured using a long short-term memory model trained based on the graph representation obtained from the GAT. These spatial and temporal correlations are used by prediction …

Xudong Wang

Xudong Wang

Harbin Institute of Technology

IEEE Transactions on Industrial Informatics

Data-Based Guaranteed Trajectory Estimation for Unmanned Surface Vehicles

This article concerns the guaranteed trajectory estimation problem for unmanned surface vehicles (USVs) via set-membership estimation technique. Taking both rigid-body and hydrostatics kinetics into consideration, the nonlinear dynamic model of USV system is derived, where the parameters of system are all unknown. Considering external disturbance and nonlinearities, an offline data-based set-membership estimation algorithm of unknown system parameters is proposed to obtain the set representation of system parameters, which contain the actual parameters of USV. Then, based on the obtained parameter sets, an online guaranteed trajectory estimation algorithm of USV is constructed to provide guaranteed sets enclosing actual trajectory points of USV, which consists of a time-update step and measurement-update step. To tackle the nonlinear transformation of zonotopes, both interval arithmetic and …

Nianyin Zeng (曾念寅)

Nianyin Zeng (曾念寅)

Xiamen University

IEEE Transactions on Industrial Informatics

DPMSN: A Dual-Pathway Multiscale Network for Image Forgery Detection

Multimedia images have become an important way for the sharing of digital information. However, advanced editing tools provide easy methods for malicious content modification, resulting in less viable information. Therefore, there is an urgent need to design an algorithm for image tampering detection and localization, which is capable of locating the authenticity region of the received image, thus delivering the accurate information and assisting in correct decision making in industries and other fields. In this article, a novel dual-pathway multiscale network (DPMSN) is proposed for the image forgery detection, which mainly focuses on extracting the edge information. In particular, a dual-pathway structure is deployed to align visual features in red, green and blue (RGB) space and edge information in LAB space, where a coarse prediction mask is generated to promote accurate localization of the forged regions. By …

Gen Li

Gen Li

Cardiff University

IEEE Transactions on Industrial Informatics

Complex Domain Analysis Based Fault Detection in VSC Interfaced Multi-terminal LVDC System

In voltage source converter-based low-voltage dc systems (LVDC), the fault current of the dc-link capacitor is considerably high and destructive to system infrastructure. It is necessary to develop effective fault detection methods with sufficient sensitivity and accuracy. To achieve these targets, a complex domain analysis-based fault detection method is proposed in this article. Specifically, the proposed method first fits the transient current into a linear combination of exponential functions, which is solved in the Z-domain-based on the Padé approximation. Second, exponents of the fitted function are projected into the complex plane. A state circle centered on the origin is defined on the complex plane to detect dc faults according to the position of projection points relative to the state circle. The proposed method can differentiate several typical situations via theoretical analysis, including fault line transients, healthy line …

Chao Jiang

Chao Jiang

East China University of Science and Technology

IEEE Transactions on Industrial Informatics

Transferable Deep Slow Feature Network With Target Feature Attention for Few-Shot Time-Series Prediction

Data-driven methods for predicting quality variables in wastewater treatment processes (WWTPs) have mostly ignored the slow time-varying nature of WWTP, and they are data-consuming that need a large amount of independent and homogeneously distributed data, which makes it difficult to collect. To address this issue with few-shot and inconsistent distribution, a transfer learning method called transferable deep slow feature network (TDSFN) for time-series prediction is proposed by leveraging the knowledge of relevant datasets. TDSFN extracts nonlinear slow features of WWTP with inertia from the time series through a deep slow feature network and constructs the domain invariant features based on them. Target feature attention is designed in TDSFN to enhance the predictor adaptability to the target domain by assigning weights to the source features based on their similarity to target features. Furthermore, a …

Alejandro Castillo Atoche

Alejandro Castillo Atoche

Universidad Autónoma de Yucatán

IEEE Transactions on Industrial Informatics

Adhesion Testing System Based on Convolutional Neural Networks for Quality Inspection of Flexible Strain Sensors

Manufacturing reliable strain sensors based on nanostructured materials faces several challenges, such as ensuring quality inspection of the adhesion between the active sensor and its substrate. In order to overcome this, the use of a deep learning-based technique is proposed herein to determine the in-situ adhesion strength. This study conducts an adhesion strength analysis between carbon nanotubes (CNTs) over a polymeric substrate (as components of a strain sensor), using image analysis. In line with the edge-computing paradigm, a novel inspection system is presented. An embedded processor equipped with a convolutional neural network architecture is used to inspect the CNT adhesion deposited on a substrate surface using deep learning semantic segmentation. This determines the relative concentration of CNTs covering the peeled area and its spatial probabilistic distribution map. Experimental …

Soumya Ranjan Mohanty

Soumya Ranjan Mohanty

Indian Institute of Technology, BHU

IEEE Transactions on Industrial Informatics

Intelligent Fault Detection and Classification for an Unbalanced Network With Inverter-Based DG Units

In this article, a machine-learning-based fault detection and classification method is proposed. Two supervised learning-based protection modules are developed for the relays considered in the study—one to detect the fault and discriminate between the symmetrical or unsymmetrical nature of the fault and another to detect the faulty phase(s). A robust set of features using both relay voltage and current signals is utilized for developing the modules. The features are obtained using the multiresolution decomposition based on the empirical wavelet transform. The modules are tested on the unbalanced IEEE 13-node network integrated with inverter-based distributed generation systems capable of reactive power injection in the low-voltage ride-through mode of operation. Varying penetration levels and intermittent output of distributed generation, fault resistance, fault inception time, noise in the signals, and switching …

Yan Wang 王龑

Yan Wang 王龑

Fudan University

IEEE Transactions on Industrial Informatics

MGR3Net: Multigranularity Region Relation Representation Network for Facial Expression Recognition in Affective Robots

Automatic facial expression recognition (FER) based on face images is essential for affective robots, which are designed for interactive companions and intelligent healthcare. Although existing DL-based FERs have made significant progress, an accurate FER model in robots is challenging due to the subtle differences in facial expressions across various scenarios. To address this issue, we propose a multigranularity region relation representation network (MGR 3 Net) to improve the robustness and generalization of FER via attention-guided global-local fusion. The MGR 3 Net is composed of three modules: multigranularity attention (MGA), holistic-regional feature extractor (HRFE), and hybrid feature fusion. In the MGA module, we first process each holistic cropped face image into three granularity of face regions from coarse to fine, which are four region-cropped faces, face partitions, and face partitions …

Hongli Dong

Hongli Dong

Universität Duisburg-Essen

IEEE Transactions on Industrial Informatics

Quantized Distributed Economic Dispatch for Microgrids: Paillier Encryption–Decryption Scheme

This article is concerned with the secure distributed economic dispatch (DED) problem of microgrids. A quantized distributed optimization algorithm using the Paillier encryption–decryption scheme is developed. This algorithm is designed to optimally coordinate the power outputs of a collection of distributed generators (DGs) in order to meet the total load demand at the lowest generation cost under the DG capacity limits while ensuring communication efficiency and security. First, to facilitate data encryption and reduce data release, a novel dynamic quantization scheme is integrated into the DED algorithm, through which the effects of quantization errors can be eliminated. Next, utilizing matrix norm analysis and mathematical induction, a sufficient condition is provided to demonstrate that the developed DED algorithm converges precisely to the optimal solution under finite quantization levels (and even the three …

Hongseok Kim

Hongseok Kim

Sogang University

IEEE Transactions on Industrial Informatics

FedAND: Federated Learning Exploiting Consensus ADMM by Nulling Drift

In this article, we propose FedAND, a unified federated learning optimization algorithm, to tackle client drift and server drift issues under partial client participation. Federated learning is gaining popularity due to privacy concerns and mobile computing, but it still faces challenges due to heterogeneous and distributed data. FedAND leverages consensus alternating direction method of multipliers (ADMM) and resolves the server drift caused by the server state in the global update. Under partial participation, we prove that FedAND preserves the strong convergence properties of ADMM while suppressing the server drift, which in turn reduces the client drift and thus achieves better convergence. Our empirical results demonstrate superior performance compared to other methods such as FedProx, FedADMM, FedPD, and FedDyn in diverse scenarios of statistical and system heterogeneity under partial client participation.

Jun Yang

Jun Yang

Loughborough University

IEEE Transactions on Industrial Informatics

Multistep dual control for exploration and exploitation in autonomous search with convergence guarantee

Inspired by the concept of recently proposed dual control for exploration and exploitation, this article presents a multistep dual control for exploration and exploitation with guaranteed convergence in search for autonomous sources. To deal with an unknown source position and environment, the proposed dual control algorithm faces significant challenges in demonstrating its recursive feasibility and convergence. With the help of the properties of Bayesian estimators, we redesign a multistep dual control for exploitation and exploration algorithm with necessary terminal ingredients and show that the recursive feasibility and the convergence of the modified dual control algorithm are guaranteed. Two simulation scenarios are conducted, which demonstrate that the proposed algorithm outperforms the stochastic model-predictive control approach and the informative path planning approach in terms of searching …