The task of image-text matching refers to measuring the visual-semantic similarity between an image and a sentence. Recently, the fine-grained matching methods that explore the local alignment between the image regions and the sentence words have shown advance in inferring the image-text correspondence by aggregating pairwise region-word similarity. However, the local alignment is hard to achieve as some important image regions may be inaccurately detected or even missing. Meanwhile, some words with high-level semantics cannot be strictly corresponding to a single-image region. To tackle these problems, we address the importance of exploiting the global semantic consistence between image regions and sentence words as complementary for the local alignment. In this article, we propose a novel hybrid matching approach named Cross-modal Attention with Semantic Consistency (CASC) for image-text matching. The proposed CASC is a joint framework that performs cross-modal attention for local alignment and multilabel prediction for global semantic consistence. It directly extracts semantic labels from available sentence corpus without additional labor cost, which further provides a global similarity constraint for the aggregated region-word similarity obtained by the local alignment. Extensive experiments on Flickr30k and Microsoft COCO (MSCOCO) data sets demonstrate the effectiveness of the proposed CASC on preserving global semantic consistence along with the local alignment and further show its superior image-text matching performance compared with more than 15 state-of-the-art methods.

The remaining useful lifetime (RUL) and state of charge (SoC) of rechargeable lithium-ion batteries (LIBs) are two integral parts to ensure LIBs working reliably and safely for transportation electrification systems. The two together reflect the state of a battery in use. However, existing capacity estimation approaches focus on separately modeling one of them, and no one has proposed a unified estimation model that is applicable to both RUL and SoC estimation yet. In this article, we propose a unified deep learning method that can be implemented for both RUL and SoC estimation. The proposed method leverages long short-term memory recurrent neural networks to achieve state-of-the-art accurate capacity estimation for LIBs under complex operating conditions. Notably, the unified method can perform not only one-step-ahead prediction but also multistep-ahead estimation with high accuracy, achieving RUL estimation error within ten cycles and SoC estimation error within 0.13%. Experimental data collected from battery testing systems with simulated complex operating conditions are used to train the method. A series of comparative experiments are conducted to compare our method with other existing methods. The experimental results show that our method can increase estimation accuracy and robustness for LIBs estimation problems via capturing the long-term dependencies among battery degradation data.

Prognostic health management (PHM) has become a crucial part in building highly automated systems, whose primary task is to precisely predict the remaining useful life (RUL) of the system. Recently, deep models including convolutional neural network (CNN) and long short-term memory (LSTM) have been widely used to predict RUL. However, these models generally require sufficient labeled training data to guarantee fair performance, whereas a limited amount of labeled data overwhelmed by the abundance of unlabeled ones is the normality in industry, and the cost of full data annotation can be unaffordable. To attack this challenge, domain adaptation seeks to transfer the knowledge from a well-labeled source domain to another unlabeled target domain by mitigating their domain gap. In this article, we leverage domain adaptation for RUL prediction and propose a novel method by aligning distributions at both the feature level and the semantic level. The proposed method facilitates a large improvement of model performance and faster convergence. Besides, we propose to use the transformer as the backbone, which can capture long-term dependency more efficiently than the widely used recurrent neural network (RNN), and is thus critical for boosting the robustness of the model. We test our model on the commercial modular aero-propulsion system simulation (CMAPSS) dataset and its newly published variant new CMAPSS (N-CMAPSS) provided by National Aeronautics and Space Administration (NASA), achieving state-of-the-art results on both source-only RUL prediction and domain adaptive RUL prediction tasks.

Energy problem is one of the most important challenges in the application of sensor networks. With the development of wireless charging technology and intelligent mobile charger (MC), the energy problem can be solved by wireless charging strategy. In the practical application of Wireless Rechargeable Sensor Networks (WRSNs), the energy consumption rate of nodes is dynamically changed due to many uncertainties such as the death and different transmission tasks of sensor nodes. However, existing works focus on on-demand schemes, which not fully consider real-time global charging scheduling. In this paper, a novel Dynamic Charging Scheme (DCS) in WRSN based on Actor-Critic Reinforcement Learning (ACRL) algorithm is proposed. In the ACRL, we introduce Gated Recurrent Units (GRU) to capture the relationships of charging actions in time sequence. Using the Actor network with one GRU layer, we can pick up an optimal or near-optimal sensor node from candidates as the next charging target more quickly and speed up the training of the model. Meanwhile, we take the tour length and the number of dead nodes as the reward signal. Actor and Critic networks are updated by the error criterion function of R and V. Compared with current on-demand charging scheduling algorithms, extensive simulations show that the proposed ACRL algorithm surpasses heuristic algorithms such as the Greedy, DP, NJNP and TSCA in the average lifetime and tour length, especially against the size and complexity increasing of WRSNs.

Machine learning methods for state of charge (SOC) estimation of lithium-ion batteries (LiBs) face the problem of domain shift. Varying conditions, such as different ambient temperatures, can cause performance degradation of the estimators due to data distribution discrepancy. Some transfer learning methods have been utilized to tackle the problem. At real-time transfer, the source model is supposed to keep updating itself online. In the process, source domain data are usually absent because the storage and acquisition of all historical running data can involve violating the privacy of users. However, existing methods require coexistence of source and target samples. In this article, we discuss a more difficult yet more practical source-free setting where there are only the models pretrained in source domain and limited target data can be available. To address the challenges of the absence of source data and distribution discrepancy in cross-domain SOC estimation, we propose a novel source-free temperature transfer network (SFTTN), which can mitigate domain shift adaptively. In this article, cross-domain SOC estimation under source-free transfer setting is discussed for the first time. To this end, we define an effective approach named minimum estimation discrepancy (MED), which attempts to align domain distributions by minimizing the estimation discrepancy of target samples. Extensive transfer experiments and online testing at fixed and changing ambient temperatures are performed to verify the effectiveness of SFTTN. The experiment results indicate that SFTTN can achieve robust and accurate SOC estimation at different ambient temperatures under source-free scenario.

In industry, accurate remaining useful life (RUL) prediction is critical in improving system reliability and reducing downtime and accident risk. Numerous deep-learning approaches have been proposed and achieved impressive performance in RUL prediction. Nevertheless, most of them are based on an unrealistic assumption, that is, the training (source) and testing (target) data follow similar distributions. In real-world applications, the source and target domains usually have different data distributions, which degrades the model performance on the unlabeled target domain. Existing adversarial domain adaptive RUL prediction methods fail to consider the label information, which potentially aligns samples with very different semantic information and ultimately leads to negative transfer. To address the above issue, this article proposes a weighted adversarial loss (WAL) for cross-domain RUL prediction. To be specific, WAL utilizes the ground-truth labels of the source domain and the pseudo-labels of the target domain to calculate weight and then obtain the WAL. This proposed loss forces the adversarial model to align samples with similar RULs from the source and target domains. Therefore, WAL can enhance positive transfer while alleviating negative transfer. Extensive experiments demonstrate that the proposed loss can be effectively plugged into existing adversarial domain adaptation methods and yield state-of-the-art results.