This paper proposes an AC bus-based distributed battery management system, where the secondary windings of modular equalizers are series-connected. The proposed system enables rapid and flexible power balancing from any cells to any cells. The modular equalizers are designated as either active or inactive, depending on their involvement in power balancing. By employing loosely coupled transformers and resonant networks, it ensures decoupling between active and inactive equalizers, allowing balancing to occur exclusively among unbalanced cells and thereby improving efficiency. The distributed control strategy enables each equalizer to determine its power target based on SOC, improving balancing accuracy. This target is tracked via locally generated control signals, eliminating the need for external synchronization or global timing, thereby enhancing system scalability. Finally, simulation results from a four-equalizer prototype demonstrate the effectiveness of the theoretical analysis.

Talkative Power Conversion (TPC) is a cost-effective communication solution for DC microgrids, enabling power converters to transmit both power and data simultaneously by embedding data into their output voltage ripples. However, due to its broadcast nature, TPC is susceptible to eavesdropping. To address this security vulnerability, this paper proposes a novel physical-layer data carrier encryption strategy. The embedded data ripple generated by the transmitter is mixed and encrypted with a noise ripple generated by the receiver, making unauthorized interception infeasible. In addition, closed-loop impedance models are derived to construct and analyze the small signal model of the proposed system. Finally, a DC microgrid based on bidirectional buck-boost converters is implemented, and theoretical analysis is validated through simulation results.

With the flourishing of remote sensing (RS) platform techniques, very high-resolution (VHR) images have become more and more popular in recent years, which benefit the task of semantic segmentation but bring new challenges as well. Small objects, such as cars and trees, only occupy a few pixels in VHR images and are usually hard to segment. Moreover, the overlap problem about similar ground objects, such as low vegetation and trees, always results in underperformance. In this article, a content-guided and class-oriented network (CGCO-Net) for VHR image semantic segmentation is proposed to tackle this problem. Specifically, an adaptive content-guided fusion (ACGF) module with deformable convolution is introduced to capture long-distance dependencies and spatial aggregation effectively. With the guidance of the high-level features, the semantic content knowledge is gradually aggregated into low-level features and the details of the original features could be preserved. In addition, a multiscale channel alignment module is introduced into the encoder–decoder structure to further extract the long-range context information and reduce the calculation consumption. In order to improve the ability of pixel-level classification, a class-oriented representation learning (CORL) way is designed with transformer blocks by class embedding and deep supervision, which gradually enhance the discrimination and benefit the final segmentation. Furthermore, a weighted loss function and a threshold optimization strategy are employed to alleviate the sample imbalance problem. Tested on three public datasets and compared with several state-of-the-art methods, the proposed CGCO-net achieves good performance in both qualitative and quantitative analysis.

The communication of the Battery Management System (BMS) is crucial to the regular operation of the BMS. This paper proposes a communication-integrated battery equalization strategy based on bidirectional flyback converters. In comparison to the strategies currently in use, the proposed strategy sends data by modulating the drive signal of the switching devices in the bidirectional flyback converter rather than using a separate signal transmission circuit, which reduces hardware complexity and lowers cost. Additionally, the receiver can demodulate data by analyzing the voltage ripple of the input and output ports in the bidirectional flyback converter. In this paper, an experimental system prototype with three bidirectional flyback converters is built. The experiment achieves an 8.33kb/s communication rate and verifies the feasibility of the proposed strategy.

The domain shift between the source and target domains limits the performance of traditional convolutional neural networks (CNNs) for feature extraction in remote sensing tasks. We propose an image translation network that uses generative adversarial networks (GANs) to transfer spectral distributions from training to test data, enhancing cross-domain semantic segmentation. Our approach fine-tunes the DeepLab-V3 framework on synthetic training data generated by the proposed network. Experimental results show improved performance in cross-domain semantic segmentation tasks for remote sensing images.