Expert Systems with Applications最新文献

{"title":"Dynamic LADRC-Based CFOA-LSTM MPPT optimizer for enhancing Grid-Connected renewable energy sources","authors":"AL-Wesabi Ibrahim ,&nbsp;Jiazhu Xu ,&nbsp;Khaled Ameur ,&nbsp;Riadh Al Dawood ,&nbsp;Zhenglu Shi ,&nbsp;Yang He ,&nbsp;Yuqing Yang ,&nbsp;Mbula Ngoy Nadège ,&nbsp;Imad Aboudrar","doi":"10.1016/j.eswa.2025.128114","DOIUrl":"10.1016/j.eswa.2025.128114","url":null,"abstract":"<div><div>In the past decade, interest in hybrid photovoltaic/wind energy conversion systems (PV/WECS) has grown due to its nonpolluting and non-depleting nature. However, these power sources rely on ecological circumstances to generate electricity. And the fluctuations of internal variables or computational errors as inner disturbances and the instability of the grid as outer disturbances are generated either via voltage dips or frequency droops. Therefore, this study introduces a dynamic linear active disturbance rejection control (LADRC) strategy for PV/WECS (Low-level control). The system integrates a wind turbine equipped with a permanent magnet synchronous generator (PMSG), which is connected to the electrical grid. Furthermore, to enhance the performance of maximum power point tracking (MPPT), an efficient metaheuristic technique called catch fish optimization algorithm (CFOA) with long short-term memory (LSTM) (CFOA-LSTM) is established (High-level control). Given that the PV/WECS is a complex nonlinear system, an enhanced version of LADRC is developed that utilizes an extended state observer (ESO) for estimating both interior and exterior disturbances, including modeling faults and parameter fluctuations. This approach regulates the DC-Link voltage and controls the active and reactive power by adjusting the utility grid currents to ensure a unity power factor. The simulation findings demonstrate that the proposed approach outperforms conventional control strategies with respect to of speedy tracking and robustness to both interior and exterior disturbances.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"285 ","pages":"Article 128114"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A knowledge and data augmentation-based method for combustion state recognition in cogeneration systems","authors":"Zhifei Sun ,&nbsp;Defeng He ,&nbsp;Xiuli Wang ,&nbsp;Wei Zhu ,&nbsp;Hongtian Chen ,&nbsp;Kai Wang","doi":"10.1016/j.eswa.2025.127969","DOIUrl":"10.1016/j.eswa.2025.127969","url":null,"abstract":"<div><div>Accurate recognition of combustion states is crucial for the safe operation of cogeneration systems. However, in actual operation, the quantity of training samples across different combustion states is severely imbalanced, making it relatively difficult to train a deep model for accurate combustion state recognition. This paper proposes a knowledge and data augmentation-based combustion state recognition method of cogeneration systems to accurately identify imbalanced combustion states samples. Firstly, combustion states are labeled by combining the temperature characteristics of the system boiler with the environmental regulations on pollutant emissions. Simultaneously, input variables are selected based on the mechanism knowledge of pollutant formation. Then, a refined auxiliary classifier generative adversarial network (RACGAN), incorporating an independent classifier and self-attention module, is designed to obtain high-quality multi-class combustion states samples. Next, evaluation criteria are established to adaptively filter the generated samples, ensuring their accuracy and diversity. Finally, an improved Residual network (IResNet) model is trained using both generated and real samples to recognize combustion states. Experiments based on actual operational data from a heat and power company show that this method achieves high accuracy, stability and potential in combustion state recognition.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"285 ","pages":"Article 127969"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143928649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coordinated multi lane-changing on highways for connected and automated vehicles in mixed traffic","authors":"Abdullah Alshakhs ,&nbsp;Muhammad Mysorewala ,&nbsp;Ali Nasir","doi":"10.1016/j.eswa.2025.127890","DOIUrl":"10.1016/j.eswa.2025.127890","url":null,"abstract":"<div><div>This paper presents a decentralized coordination algorithm for multi-vehicle lane changing in mixed traffic composed of Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs). Unlike approaches based on traditional single-vehicle decision-making, centralized control, or learning-based methods that depend on iterative exploration, the proposed framework employs a decentralized Markov Decision Process (MDP)-based model to compute a ready-to-use policy for each CAV. Assuming known reward structures, this model enables policy computation in advance. The framework is further extended with a priority-based mechanism for resolving trajectory conflicts, vehicle-to-vehicle communication for synchronized decision-making, smooth trajectory generation, and a Proportional–Integral–Derivative (PID) controller to ensure smooth longitudinal control during lane changes.</div><div>Simulation results demonstrate significant gains in traffic efficiency, with cooperative vehicles achieving up to 40% reductions in travel time compared to those constrained to fixed-lane behavior and affected by the presence of slower, non-cooperative HDVs. Acceleration remained below 2 m/s<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span>, indicating smooth transitions and enhanced passenger comfort. The approach also minimized sudden braking and hesitation during lane merges, resulting in safer and more stable interactions. These findings highlight the framework’s potential to improve throughput, safety, and comfort in mixed-autonomy traffic, offering a scalable solution for real-time cooperative decision-making. Future work will explore online learning and model adaptation to better address highly dynamic environments, including unpredictable human driving behavior and varying conditions such as weather disturbances.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 127890"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144072103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radar M3-Net: Multi-scale, multi-layer, multi-frame network with a large receptive field for 3D object detection","authors":"Yunting Yang,&nbsp;Jun Liu,&nbsp;Hongsi Liu,&nbsp;Guangfeng Jiang","doi":"10.1016/j.eswa.2025.127515","DOIUrl":"10.1016/j.eswa.2025.127515","url":null,"abstract":"<div><div>4D millimeter-wave radar has demonstrated significant potential for 3D object detection in autonomous driving due to its cost-effectiveness and robustness. However, the inherent sparsity of radar data poses a significant challenge to achieving accurate 3D object detection, as it limits the amount of meaningful information available for feature learning. The key to addressing the performance degradation caused by sparsity lies in expanding the receptive field and enriching feature representations. To this end, we propose a multi-scale, multi-layer, multi-frame network with a large receptive field, named Radar <span><math><msup><mrow><mtext>M</mtext></mrow><mrow><mn>3</mn></mrow></msup></math></span>-Net. First, we design a multi-scale voxel feature encoding (MSVFE) module and a multi-layer attention (MLA) module, both of which significantly expand the receptive field and enrich features, effectively addressing the issue of sparsity. Then, a multi-frame fusion module is developed to further enhance features by utilizing the accumulation of temporal information. Simultaneously, we design a novel sparse dual-head within the sparse framework to address the decline in detection accuracy for large object categories caused by radar sparsity. Extensive experiments on the View-of-Delft and TJ4DRadSet datasets have confirmed the advancement and effectiveness of our network. Specifically, our method achieves state-of-the-art mean average precision (mAP) performance on both datasets, even outperforming some multimodal approaches in certain metrics.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 127515"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on rolling bearing fault diagnosis based on parallel depthwise separable ResNet neural network with attention mechanism","authors":"Xiaoli Zhang ,&nbsp;Xudong Zhang ,&nbsp;Wang Liang ,&nbsp;Feixiang He","doi":"10.1016/j.eswa.2025.128105","DOIUrl":"10.1016/j.eswa.2025.128105","url":null,"abstract":"<div><div>An accurate and efficient fault diagnosis method for rolling bearings significantly contributes to enhancing the safety and stability of mechanical systems. To address the limitations of existing intelligent diagnostic methods—such as low accuracy and efficiency under complex operating conditions, excessive model parameters leading to prolonged inference times, a high risk of overfitting, and poor robustness—a parallel deep separable ResNet neural network based on an attention mechanism (PDSResNet-AM) is proposed. The model leverages a dual-input strategy by transforming one-dimensional vibration signals into two-dimensional time–frequency representations using both Continuous Wavelet Transform (CWT) and Short-Time Fourier Transform (STFT), thereby enriching the input features. Additionally, Depthwise Separable Convolution (DSConv) and the Convolutional Block Attention Module (CBAM) are utilized to enhance relevant features, reduce the number of parameters, and mitigate the risk of overfitting. Furthermore, an optimized dilated residual convolution module is introduced to replace conventional convolutional modules, enhancing the model’s generalization capability. Extensive experiments conducted under varying noise levels, load conditions, and rotational speeds demonstrate that the proposed method achieves superior diagnostic accuracy, strong generalization capability, and noise robustness compared to existing deep learning models. These results underscore the feasibility of deploying PDSResNet-AM in industrial applications.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 128105"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143948420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An improved end-to-end micro-expression recognition system for real-world videos via dual-input CNN architecture","authors":"Y.S. Gan ,&nbsp;Kun-Hong Liu ,&nbsp;Min-Huan Wu ,&nbsp;Gen-Bing Liong ,&nbsp;Sze-Teng Liong","doi":"10.1016/j.eswa.2025.128062","DOIUrl":"10.1016/j.eswa.2025.128062","url":null,"abstract":"<div><div>Micro-expression (ME) recognition reveals nonverbal emotions through subtle, involuntary facial muscle movements. However, the development and commercialization of ME recognition systems have been hindered by the lack of databases that accurately reflect real-world conditions. This study addresses this challenge by proposing a robust end-to-end system designed to operate effectively in unconstrained environments. Existing methods typically rely on a single apex frame, which may be unreliable due to noise, occlusions, or lighting variations. To address these issues, a 3D facial reconstruction technique is applied as a pre-processing step to normalize pose and lighting. A novel dual-peak frame detection strategy is then introduced to extract two expressive optical flow frames, reducing the impact of noise from any single frame. Finally, a Shallow and Small-size Dual-input (SSD) CNN architecture is designed to jointly process the two frames for improved emotion classification. The proposed system achieves strong performance on the challenging in-the-wild MEVIEW dataset, with accuracy and F1-score of 75 % and 77.68 %, respectively. Comprehensive evaluations further validate the effectiveness of the pipeline, highlighting its potential for real-world ME recognition applications.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 128062"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144069214","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Event-triggered-based output-feedback control for uncertain nonlinear discrete-time systems: a reinforcement learning method","authors":"Jianwei Ren ,&nbsp;Ping Li ,&nbsp;Zhibao Song","doi":"10.1016/j.eswa.2025.128094","DOIUrl":"10.1016/j.eswa.2025.128094","url":null,"abstract":"<div><div>This paper develops a novel event-triggered (ET) output-feedback control algorithm that utilizes the reinforcement learning (RL) method for uncertain nonlinear discrete-time (DT) systems with unknown control directions. Unlike existing approaches, this work addresses the challenges of a nonstrict-feedback system. To estimate unmeasured system states, a radial basis function neural networks (RBF NNs) based observer is established. An efficient ET mechanism is then proposed to reduce communication redundancy. To tackle the issue of unknown control directions, the ET-based DT Nussbaum gain is employed. Under the Lyapunov stability theorem, it is demonstrated that semi-global ultimate uniform boundedness (SGUUB) is achieved for tracking errors and all signals in closed-loop systems. Simulations are provided to illustrate the effectiveness of the algorithm in handling nonlinear DT systems with unknown control directions.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 128094"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DiT-SFDA: A source-free domain adaptation method for intelligent diagnosis of cardiovascular diseases with limited heart sound samples","authors":"Suiyan Wang ,&nbsp;Yang Liu ,&nbsp;Zhixiang Liu ,&nbsp;Xiaoming Yuan ,&nbsp;Yun Ji ,&nbsp;Pengfei Liang","doi":"10.1016/j.eswa.2025.128118","DOIUrl":"10.1016/j.eswa.2025.128118","url":null,"abstract":"<div><div>In recent years, the application of deep learning in intelligent diagnosis (ID) of cardiovascular diseases (CVDs) has significantly improved diagnostic efficiency and accuracy. However, in practice, owing to data privacy constraints, high labeling cost and specialized medical knowledge, collecting adequate labeled samples continues to present substantial technical difficulties, which makes ID of CVDs under limited samples a challenging issue. In this paper, a novel source-free domain adaptation (SFDA) approach for ID of CVDs, named DiT-SFDA, is proposed by integrating an improved diffusion model based on transformer (DiT) and a semi-supervised domain adaptation network (SDAN). Specifically, the method first converts heart sound (HS) signals into Mel spectrograms that can represent their time–frequency characteristics. Then, more realistic labeled samples are generated through DiT using limited real labeled data, effectively solving training data insufficiency. Subsequently, the generated labeled samples serve as the source domain, while the real samples serve as the limited labeled data in the target domain, and the SDAN based on minimax entropy is employed to further improve the performance of the model. Finally, experimental validation demonstrates that the DiT-SFDA method achieves significantly better diagnostic performance than other methods on two datasets. This innovative approach not only effectively addresses the critical challenge of data scarcity, but also provides an efficient and robust solution for the early screening and precise diagnosis of CVDs.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"285 ","pages":"Article 128118"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Anti-containment and α-anti-containment neighborhoods-based neighborhood rough sets and their classification models in medical application for infectious diseases","authors":"Xia Liu ,&nbsp;Xianyong Zhang ,&nbsp;Benwei Chen ,&nbsp;Hongyuan Gou ,&nbsp;Mawia Osman","doi":"10.1016/j.eswa.2025.127892","DOIUrl":"10.1016/j.eswa.2025.127892","url":null,"abstract":"<div><div>Uncertainty modeling aims to improve the accuracy and reliability of predictions by identifying and quantifying uncertainties through statistical and analytical methods. In particular, neighborhood rough set models have undergone significant development in the latest medical applications for infectious diseases, and they improve approximation accuracies and achieve risk classifications. However, there is a lack of clear semantic explanation of the existing containment neighborhoods in medical applications, and the corresponding classification methods are relatively simple and lack practicality. In this paper, the systemic <span><math><mi>α</mi></math></span>-containment, <span><math><mi>α</mi></math></span>-anti-containment and anti-containment neighborhoods are constructed by semantic analyses of posterior and conditional probabilities, and thus they not only deduce novel neighborhood rough sets but also drive more detailed classification models that can be flexibly applied to different medical scenarios. Firstly, posteriori probabilities and a threshold are introduced to propose the <span><math><mi>α</mi></math></span>-containment neighborhoods. Then, the <span><math><mi>α</mi></math></span>-anti-containment and anti-containment neighborhoods are constructed by using conditional probabilities. Accordingly, they can induce new neighborhood rough sets and obtain better approximation accuracies. In addition, the inclusion relationships between the proposed and existing neighborhoods are discussed, and the threshold monotonicity is studied through theoretical analysis and examples. Furthermore, the proposed neighborhoods are employed to classify individuals at some suspected risk of infectious diseases for different application scenarios (such as the transmission research of infected individuals and the tracing of the root causes of infectious diseases), based on the semantic analysis of posterior and conditional probabilities. By flexibly selecting thresholds, the <span><math><mi>α</mi></math></span>-containment and <span><math><mi>α</mi></math></span>-anti-containment neighborhoods can deduce more detailed classification models that are more practical for actual needs. Finally, several examples of medical application are implemented to illustrate the advantages of our classification models. The optimal accuracies and threshold monotonicity are validated through datasets experiments, showing that the three proposed classification models are superior to the existing models. Therefore, the whole research is beneficial to the development of neighborhoods, uncertainty modeling and medical applications.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"286 ","pages":"Article 127892"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel interpretability paradigm based on semantic features of time–frequency images for trustworthy cross-machine fault diagnosis","authors":"Xin Kang,&nbsp;Zhengyang Cheng,&nbsp;ChengCheng Duan,&nbsp;Junsheng Cheng,&nbsp;Yu Yang","doi":"10.1016/j.eswa.2025.128115","DOIUrl":"10.1016/j.eswa.2025.128115","url":null,"abstract":"<div><div>Recently, domain generalization-based fault diagnosis (DGFD) has garnered significant attention for recognizing faults without the accessibility of the target domain. However, most DGBD methods concentrate on enhancing generalization through improved inductive bias or learning bias, while overlooking observational bias. This oversight leads to limited interpretability and subpar cross-machine diagnostic performance, which hinders the practical deployment of intelligent fault diagnosis in industry. In response to the above issues, the core idea of this paper is to reduce observational bias by defining cross-domain consistent semantic features for different health states in time–frequency images based on fault mechanisms, then training the model to rely exclusively on these defined semantic features for diagnosis, ensuring both interpretability and generalization. To realize this idea, this paper conducts a detailed analysis of various time–frequency analysis methods, evaluating their effectiveness in extracting cross-domain consistent semantic features, and proposes improvements to enhance these features, ensuring the model generalization. Furthermore, the model’s behavior is thoroughly visualized using the class activation map (CAM), confirming that the model relies solely on the defined semantic features as its decision basis, thereby ensuring interpretability. Finally, the model generalization and interpretability are tested using a single-source domain training and multi-target domain testing approach. Notably, although this study uses rolling bearing as an example, it is applicable to other fault diagnosis scenarios, such as for gears and motors. Source code: <span><span>https://github.com/kangxin8/TF_based_CAM_beaeing_fault_detection</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50461,"journal":{"name":"Expert Systems with Applications","volume":"285 ","pages":"Article 128115"},"PeriodicalIF":7.5,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}