As a promising and non-contact detection technique, machine vision has been widely used in fault diagnosis of substation equipment. The rapid and accurate detection of substation equipment in infrared images is one of the key steps for automatic fault diagnosis. However, the complexity of image background, the low contrast of infrared images and the rotational targets in infrared images pose a great challenge to detection task. This study aims to improve the detection accuracy of the model while having real-time detection speed, and propose a lightweight power equipment detection network (PEDNet) based on YOLOv4-tiny. First, a novel global information aggregation module is constructed to guide the network to focus on the salient regions where the target equipment is located. Second, an improved spatial transformer network is introduced to reduce the impact of rotational targets on detection accuracy. Finally, a feature enhanced fusion network is designed through the use of a multi-scale feature cross-fusion structure. It can fully fuse the feature information of the salient region, the rotational targets and the strong semantic information. The experimental results show that the proposed PEDNet can reach 92.66% detection accuracy and 107.07 FPS real-time detection speed on testing datasets. Compared with YOLOv4-tiny, there is a small sacrifice in detection speed, but the detection accuracy is improved and significantly higher than the existing state of the art (SOTA) object detection models.

Field Programmable Gate Array (FPGA) has become an efficient accelerator for convolutional neural network (CNN) inference due to its high performance and flexibility. To further improve the performance of CNN inference on FPGAs, an Intellectual Property core (IP core) called Deep Learning Processor Unit (DPU) is released by Xilinx. Unlike previous FPGA-based hardware designs focusing on specific functions or CNNs, the DPU IP supports ample basic functions of deep learning, and the developers can take advantage of DPUs to accelerate CNN inference conveniently. In DPU-based CNN acceleration platform, an encapsulated scheduler plays a crucial role in task scheduling between heterogeneous ARM and multiple DPUs. However, the current scheduler is unsatisfactory because its low schedule efficiency. This paper thus presents a high performance task assignment framework built upon Xilinx hybrid CPU-FPGA MPSoC devices. We first evaluate the main causes of low schedule efficiency problem. Then, we explore the scheduler rules and improve shedule efficiency through purposeful observations and analysis. Finally, we integrate our optimizations, and propose an efficient task assignment framework to maximize performance on DPU-based CNN acceleration platform. Experimental results on Xilinx Zynq UltraScale+ MPSoC zcu104 show that our efficient task assignment framework significantly boosts schedule efficiency for small-scale CNNs (from 36% to 70%), medium-scale CNNs (from 65% to 95%), and large-scale CNNs (from 77% to 99%) compared with original schedule strategy.

DC distribution network faults seriously affect the reliability of system power supply. Therefore, this paper proposes a fault recovery reconfiguration strategy for DC distribution networks, based on hybrid particle swarm optimization. The original particle swarm algorithm is improved by simplifying the distribution network structure, introducing Lévy Flight, and designing an adaptive coding strategy. First, the distribution network structure is equivalently simplified to reduce the problem dimensionality. Further, the generated branch groups are ensured to satisfy the radial constraints based on the adaptive solution strategy. Subsequently, Lévy flight is introduced to achieve intra-group optimality search for each branch group. The method is simulated in several distribution systems and analyzed in comparison with the particle swarm algorithm, genetic algorithm, and cuckoo algorithm. Finally, the results validate the accuracy and efficiency of the proposed method.

Multiple demand responses and electric vehicles are considered, and a micro-grid day-ahead dispatch optimization model with photovoltaic is constructed based on stochastic optimization theory. Firstly, an interruptible load model based on incentive-based demand response is introduced, and a demand response mechanism for air conditioning load is constructed to implement an optimal energy consumption curve control strategy for air conditioning units. Secondly, considering the travel demand and charging/discharging rules of electric vehicles, the electric vehicle optimization model is built. Further, a stochastic optimization model of micro-grid with demand response and electric vehicles is developed because of the uncertainty of photovoltaic power output. Finally, the simulation example verifies the effectiveness of the proposed model. The simulation results show that the proposed model can effectively tackle the uncertainty of photovoltaic, as well as reduce the operating cost of micro-grid. Therefore, the effective interaction between users and electric vehicles can be realized.

As an indispensable material, aluminum requires real-time surface defect detection to effectively control quality. However, defects on the surface of aluminum have the inherent properties of tiny size, diverse morphology, and multiscale variations, which present a significant challenge for defect detection. This article aims to improve the performance and efficiency of a defect detection model and proposes an efficient multiscale perceptual enhancement network for aluminum defect detection (ADDet). First, to compensate for the loss of deep features and mitigate information interference from the background, a multilayer semantic-aware enhancement module (MSEM) is proposed for the feature extraction network. Second, a proximity feature pyramid (PFP) structure in the neck is designed to effectively merge multiscale features with low computational cost and enhance feature representation across spatial coordinates, channels, and scales. Finally, a large kernel-size decoupled head (LKSDH) with a large receptive field and multiple detection branches is constructed in the detection head to improve detection accuracy. Experimental results show that the mean average precision (mAP) of the proposed ADDet is 80.85% with a real-time detection speed of 63.41 frames per second (FPS). Compared to YOLOv7, ADDet improves the mAP by 3.02%, and the number of parameters and FLOPs are reduced by 23.8% and 13%, respectively. Moreover, ADDet is shown to be superior to state-of-the-art detection methods in terms of detection accuracy.

Traditional reactive power optimization mainly considers the constraints of active management elements and ignores the randomness and volatility of distributed energy sources, which cannot meet the actual demand. Therefore, this paper establishes a reactive power optimization model for active distribution networks, which is solved by a second-order cone relaxation method and interval optimization theory. On the one hand, the second-order cone relaxation technique transforms the non-convex optimal dynamic problem into a convex optimization model to improve the solving efficiency. On the other hand, the interval optimization strategy can solve the source–load uncertainty problem in the distribution network and obtain the interval solution of the optimization problem. Specially, we use confidence interval estimation to shorten the interval range, thereby improving the accuracy of the interval solution. The model takes the minimum economy as the objective function and considers a variety of active management elements. Finally, the modified IEEE 33 node arithmetic example verifies the feasibility and superiority of the interval optimization algorithm.