Network Loss Research Articles

Security constrained optimal power flow solution for practical transmission grid using hybrid use of generating plant and network restructuring

This paper presented a study on the security constrained optimal power flow (SCOPF) of a practical transmission utility grid network (TUGN). The objectives of reduction in the transmission losses of TUGN (TLGN) and to minimize the bus voltage deviations (BVD) are achieved by addition of thermal power plant (TPP) and network restructuring (NR). Identification of best fit node for generation injection is achieved using the method based on hybrid combination of analytic approach and genetic algorithm (GA). Efficacy of the proposed study is evaluated using the computation of energy equivalent to network loss saving, percentage derating of the load (PDOL), voltage profile, bus voltage deviations, and financial analysis (FA). Study is performed without and with contingency (N-1) for the base case network, TUGN with addition of TPP and TUGN with addition of TPP and NR. This is established that addition of TPP, addition of TPP with NR are effective to reduce the TLGN, improve the voltage profile and minimize the bus voltage deviations. This study is more efficient compared to various studies reported in literature.

An integrated approach to evaluating water contaminants and evaporation in agricultural water distribution systems

This study presents an innovative approach for assessing water quality in agricultural irrigation networks, integrating stable isotope analysis, in vivo zebrafish screening, and comprehensive chemical profiling to investigate the occurrence, transformation, and potential toxicity of organic contaminants. Stable isotope analysis was used to measure evaporation as a proxy for water residence time in the canal, while liquid chromatography-high resolution mass spectrometry (LC-HRMS) identified a range of organic compounds in water samples collected from both the irrigation canal and its source river. Results indicated a reduction in contaminant levels in the canal compared to the river, with the most significant evaporation and concentration changes occurring at a holding reservoir, suggesting that managing residence time could help reduce water loss in arid irrigation networks. The data also highlighted how evaporation, particularly during the dry, hot season, influences contaminant dynamics. Hierarchical clustering of LC-HRMS results showed notable differences between the chemical profiles of canal and river samples, indicating that irrigation systems may contribute to the degradation or removal of certain compounds. Over 60 % of detected compounds were naturally derived, with anthropogenic contaminants like pesticides and personal care products further highlighting human impacts. Priority contaminants, including DEET and 2-naphthalene sulfonic acid, likely originated from urban activities upstream. Initial screening using zebrafish embryos showed bioactivity across sites, confirming the presence of contaminants needing further examination. Correlation analysis linked natural compounds to evaporation rates, suggesting that flora and fauna play significant roles in the chemical makeup of canal water. Overall, this approach provides a comprehensive framework for monitoring irrigation water, offering insights into contaminant behavior and supporting the development of standardized methods for assessing chemical fate and ecological risks in agricultural irrigation systems.

Prediction of laser beam spatial profiles in a high-energy laser facility by use of deep learning

We adapt the significant advances achieved recently in the field of generative artificial intelligence/machine-learning to laser performance modeling in multipass, high-energy laser systems with application to high-shot-rate facilities relevant to inertial fusion energy. Advantages of neural-network architectures include rapid prediction capability, data-driven processing, and the possibility to implement such architectures within future low-latency, low-power consumption photonic networks. Four models were investigated that differed in their generator loss functions and utilized the U-Net encoder/decoder architecture with either a reconstruction loss alone or combined with an adversarial network loss. We achieved inference times of 1.3 ms for a 256 × 256 pixel near-field beam with errors in predicted energy of the order of 1% over most of the energy range. It is shown that prediction errors are significantly reduced by ensemble averaging the models with different weight initializations. These results suggest that including the temporal dimension in such models may provide accurate, real-time spatiotemporal predictions of laser performance in high-shot-rate laser systems.

Study on two‐stage optimal scheduling of DC distribution networks considering flexible load response

AbstractDue to the aim of developing sustainable energy systems, promoting the large‐scale accommodation of distributed renewable energy sources (DRESs) and flexible loads in DC distribution networks (DCDNs) is significant. The uncertainty of DRESs and the insufficient use of flexible loads pose a considerable challenge to the economical and safe operation of DCDNs. To address the challenge, this paper puts forward a two‐stage optimal scheduling model for the DCDNs considering flexible load response. The proposed model realises joint economic optimisation and reactive power optimisation, which is solved by the hybrid NSGAII‐MOPSO algorithm and the CPLEX. The performance of the proposed model in the modified Institute of Electrical and Electronics Engineers 33‐node system with the DCDNs is validated under different scenarios. The hybrid NSGAII‐MOPSO performs better in obtaining the Pareto front than the NSGA‐II and MOPSO individually. Compared to the traditional scheduling model, the proposed model can realise the power coordination of the flexible loads and energy storage systems to reduce the negative impact of uncertainty of DRESs while decreasing operating costs and carbon emissions by 3.94% and 36.4%. In addition, the proposed model can alleviate the network losses and ensure the node voltage for the DCDNs. Hence, the efficiency of the proposed model has been confirmed.

Distributed photovoltaic reactive power control strategy based on improved multiobjective particle swarm algorithm

AbstractDistributed power supply access to the distribution network, although it can effectively support the band voltage, will also cause problems such as voltage overruns at the point of grid connection and large network losses, so this paper establishes a reactive power optimization model containing three objectives: network loss, voltage fluctuation rate, and static reactive power generator (SVG) installation capacity in distributed photovoltaic power generation scenarios by taking advantage of the characteristics of SVG that both absorb and send out reactive power. A multiobjective particle swarm algorithm with an adaptive grid and roulette mechanism is introduced to ensure the uniformity and diversity of the Pareto boundaries under the constraint that the output of each device does not exceed the constraints, and to obtain the optimal set of solutions capable of coping with the stochastic fluctuations of distributed power sources. When the algorithm is compared with three other algorithms, such as nondominated sorting genetic algorithm‐II, the results show that it reduces the network loss by about 25% and significantly improves the voltage fluctuation rate.

Load forecasting based on multi-core learning Support Vector Machine (SVM)

Abstract The development of smart grids requires enhanced data integration, robust risk assessment, and dynamic response optimization. In this paper, a multi-core learning Support Vector Machine (SVM) model is presented to improve the accuracy and efficiency of load and photovoltaic output forecasting. The model leverages kernel function optimization and parallel computing frameworks to handle large-scale data efficiently. Additionally, a comprehensive risk assessment system is developed to quantify risks such as overvoltage, undervoltage, line overload, and load loss in distribution networks. An adaptive genetic algorithm-based risk control model is also proposed, optimized in two stages—day-ahead and intra-day—to achieve minimal comprehensive risk through real-time adjustments in distributed power output and electric vehicle charging strategies. Furthermore, an integrated virtual synchronous control online verification method for source-network-load-storage is introduced, enhancing system response speed and control accuracy. These innovations collectively provide a solid theoretical foundation and technical support for the efficient and safe operation of smart grids, addressing the increasing demands of modern energy systems.

Prediction of the sound transmission loss of shape-varied sonic crystals: A transfer matrix approach.

This paper proposes a transfer matrix method (TMM) for modeling sonic crystals to predict the transmission loss of noise exiting an air extraction system. Because the crystals may be of different shapes (e.g., square, circular, or standardized airfoil profile to minimize airflow resistance) and must account for thermo-viscous losses, a discrete version of the TMM is used. Similar to the finite element method, a discretization of the geometry is first performed. Each element is modeled with a transfer matrix (TM) that includes the local thermo-viscous losses which attenuate the sound wave. For each element in parallel, the parallel TMM is employed. For the subsequently created elements in series, the classic TMM is used. This generates a global TM from which the sound transmission loss of the crystal network is deduced. The predictions obtained by the proposed method are compared to measurements in an acoustic tube for three different shapes of sonic crystals. The results show that a geometric tortuosity correction is necessary for the predicted bandgap center frequency to match the measurement. A correction is proposed, but this requires a possible refinement for more complicated profiles.

Hierarchical Optimal Dispatching of Electric Vehicles Based on Photovoltaic-Storage Charging Stations

Electric vehicles, known for their eco-friendliness and rechargeable–dischargeable capabilities, can serve as energy storage batteries to support the operation of the microgrid in certain scenarios. Therefore, photovoltaic-storage electric vehicle charging stations have emerged as an important solution to address the challenges posed by energy interconnection networks. However, electric vehicle charging loads exhibit notable randomness, potentially altering load characteristics during certain periods and posing challenges to the stable operation of microgrids. To address this challenge, this paper proposes a hierarchical optimal dispatching strategy based on photovoltaic-storage charging stations. The strategy utilizes a dynamic electricity pricing model and the adaptive particle swarm optimization algorithm to effectively manage electric vehicle charging loads. By decomposing the dispatching task into multiple layers, the strategy effectively solves the problems of the “curse of dimensionality” and slow convergence associated with large numbers of electric vehicles. Simulation results demonstrate that the strategy can effectively achieve peak shaving and valley filling, reducing the load variance of the microgrid by 24.93%, and significantly reduce electric vehicle charging costs and distribution network losses, with a reduction of 92.29% in electric vehicle charging costs and 32.28% in microgrid losses compared to unorganized charging. Additionally, this strategy can meet the travel demands of electric vehicle owners while providing convenient charging services.

Coordinated Optimal Operation Method for Snow-Shaped Distribution Networks Based on Game Theory

The snow-shaped distribution network is developed and upgraded from the traditional single-loop network and double-loop network structure. It is composed of three or four groups of single-loop networks in pairs that form a feeder cluster in the form of a ring network. Based on this topology, the coordinated optimization operation problem is studied. First, the game elements are analyzed, and then a cooperative game-based, coordinated optimization operation model is established by taking the minimum network loss and voltage deviation of the system as the utility function. Then, the alternating direction multiplier method (ADMM) is employed in Cplex solver to solve the established optimization model. Finally, a numerical case is given to verify the effectiveness of the proposed model and method. The results show that, by using the complementary difference relationship of source–charge characteristics between different regions in the snow-shaped distribution networks, the cooperative game model can realize the energy interconnection among different regions to improve the operation level.

LES-YOLO: efficient object detection algorithm used on UAV for traffic monitoring

Abstract The use of UAVs for traffic monitoring greatly facilitates people’s lives. Classical object detection algorithms struggle to balance high speed and accuracy when processing UAV images on edge devices. To solve the problem, the paper introduces an efficient and slim YOLO with low computational overhead, named LES-YOLO. In order to enrich the feature representation of small and medium objects in UAV images, a redesigned backbone is introduced. And C2f combined with Coordinate Attention is used to focus on key features. In order to enrich cross-scale information and reduce feature loss during network transmission, a novel structure called EMS-PAN (Enhanced Multi-Scale PAN) is designed. At the same time, to alleviate the problem of class imbalance, Focal EIoU is used to optimize network loss calculation instead of CIoU. To minimize redundancy and ensure a slim architecture, the P5 layer has been eliminated from the model. And verification experiments show that LES-YOLO without P5 is more efficient and slimmer. LES-YOLO is trained and tested on the VisDrone2019 dataset. Compared with YOLOv8n-p2, mAP@0.5 and Recall has increased by 7.4% and 7%. The number of parameters is reduced by over 50%, from 2.9 M to 1.4 M, but there is a certain degree of increase in FLOPS, reaching 18.8 GFLOPS. However, the overall computational overhead is still small enough. Moreover, compared with YOLOv8s-p2, both the number of parameters and FLOPS are significantly reduced , while the performance is similar . As for real-time, LES-YOLO reaches 138 fps on GPU and a maximum of 78 fps on edge devices of UAV.

Vision-Based Localization Method for Picking Points in Tea-Harvesting Robots

To address the issue of accurately recognizing and locating picking points for tea-picking robots in unstructured environments, a visual positioning method based on RGB-D information fusion is proposed. First, an improved T-YOLOv8n model is proposed, which improves detection and segmentation performance across multi-scale scenes through network architecture and loss function optimizations. In the far-view test set, the detection accuracy of tea buds reached 80.8%; for the near-view test set, the mAP0.5 values for tea stem detection in bounding boxes and masks reached 93.6% and 93.7%, respectively, showing improvements of 9.1% and 14.1% over the baseline model. Secondly, a layered visual servoing strategy for near and far views was designed, integrating the RealSense depth sensor with robotic arm cooperation. This strategy identifies the region of interest (ROI) of the tea bud in the far view and fuses the stem mask information with depth data to calculate the three-dimensional coordinates of the picking point. The experiments show that this method achieved a picking point localization success rate of 86.4%, with a mean depth measurement error of 1.43 mm. The proposed method improves the accuracy of picking point recognition and reduces depth information fluctuations, providing technical support for the intelligent and rapid picking of premium tea.

Optimizing flotation froth image segmentation via parallel branch network and hybrid loss supervision

Flotation is a crucial technology for fine coal separation, and accurately acquiring bubble size information during the flotation process is essential for monitoring flotation conditions and achieving intelligent control. However, existing semantic segmentation models encountered issues with boundary disconnection when segmenting flotation bubbles, resulting in deviations between the extracted bubble sizes and their true values. To address the aforementioned challenges, a semantic segmentation model was proposed to maintain high-resolution feature maps throughout the network by designing a parallel branch network structure. Additionally, a ConvTranspose module was proposed to preserve the detailed feature information of images while gradually enhancing the resolution of feature maps. In the model training phase, a hybrid loss function combining pixel classification loss with shape similarity loss was proposed to alleviate the sample imbalance problem caused by the substantial difference in the number of pixels between bubble boundaries and the interior of bubbles. Moreover, since traditional semantic segmentation evaluation metrics, such as MIoU, lack a mechanism for measuring bubble boundary continuity and cannot effectively penalize the problem of boundary disconnection, this paper proposed a new evaluation method for assessing the segmentation performance of flotation froth images. To comprehensively evaluate the effectiveness of the proposed method, this paper conducted tests using flotation froth images collected from actual production processes. Compared with existing methods, the segmentation model proposed in this paper exhibited clear superiority in mitigating the problem of bubble boundary disconnection. The prediction error for the number of bubbles was 6.38 %, which is significantly better than other methods.

Flexibility-Constrained Energy Storage System Placement for Flexible Interconnected Distribution Networks

Configuring energy storage systems (ESSs) in distribution networks is an effective way to alleviate issues induced by intermittent distributed generation such as transformer overloading and line congestion. However, flexibility has not been fully taken into account when placing ESSs. This paper proposes a novel ESS placement method for flexible interconnected distribution networks considering flexibility constraints. An ESS siting and sizing model is formulated aiming to minimize the life-cycle cost of ESSs along with the annual network loss cost, electricity purchasing cost from the upper-level power grid, photovoltaic (PV) curtailment cost, and ESS scheduling cost while fulfilling various security constraints. Flexible ramp-up/-down constraints of the system are added to improve the ability to adapt to random changes in both power supply and demand sides, while a fluctuation rate of net load constraints is also added for each bus to reduce the net load fluctuation. The nonconvex model is then converted into a second-order cone programming formulation, which can be solved in an efficient manner. The proposed method is evaluated on a modified 33-bus flexible distribution network. The simulation results show that better flexibility can be achieved with slightly increased ESS investment costs. However, a large ESS capacity is needed to reduce the net load fluctuation to low levels, especially when the PV capacity is large.

LSL-SS-Net: level set loss-guided semantic segmentation networks for landslide extraction

ABSTRACT A landslide inventory is essential for numerous applications, including disaster prevention and mitigation. The emergence of advanced satellite technologies and the proliferation of extensive satellite imagery have greatly enhanced the role of deep learning-based semantic segmentation networks in the development of comprehensive landslide inventories. In deep learning, especially in semantic segmentation tasks, the loss function is crucial to guide the training of neural networks. However, typical loss functions have several challenges and shortcomings, including difficulty in handling objects of different scales and neglecting spatial correlation. These drawbacks hinder semantic segmentation networks from solely concentrating on accurately predicting landslide categories with rich semantic information. The spatial details (e.g. small landslides and precise landslide boundary information) in the segmentation results may be lost, reducing the accuracy of landslide extraction. Most studies only focus on training and detection in the local region. Consequently, detecting new unexplored region and detrimental to emergency response is challenging. This study presents a level set loss-guided semantic segmentation network, which can be integrated with different semantic segmentation networks, to address these issues. Three out of five study areas (e.g. Bijie, Jiuzhaigou, and Taitung) are used for model training and accuracy evaluation, while the rest, which are newly unexplored regions (e.g. Hokkaido and Haiti), are used to test the transferability of the proposed loss function in different regions. The study results validate the effectiveness of the designed level set loss in various semantic segmentation networks and in different study areas for landslide extraction.

Power Loss Minimization and Voltage Profile Improvement on Nigeria Distribution Network Using Whale Optimization Algorithm

The power grid has significant power losses, unstable voltage, and large voltage drops during high demand due to its high resistance. One way to reduce power loss is by strategically placing and sizing Distributed Generation (DG) within the distribution network. However, improper installation of DG can increase power loss. To address this issue, various optimization techniques have been used, but finding the best solution and dealing with complex issues has been challenging. It is important to make efforts to optimally place and size DG in the distribution network to minimize power loss. To this end, a research study aimed to minimize power loss and improve the voltage profile on the DG network using the Whale Optimization Algorithm (WOA) was done. Objective functions were formulated and integrated into WOA, and power flow analysis on the standard IEEE 33-bus and 33-bus Ilorin industrial distribution feeder with and without DG was conducted using the forward and backward sweep distribution load flow algorithm. The optimal size and location of DG for power network loss reduction using sensitivity analysis (SA) and the whale optimization algorithm were determined. The results of the power flow analysis indicated that the total active losses and reactive power losses were reduced to varying extents with the integration of DG using both SA and WOA. The validation results showed that WOA is more efficient and provides high-quality solutions in terms of system loss reduction and best placement for DG in power systems compared to the application of SA. Additionally, WOA was found to offer accurate and high-quality solutions for power loss reduction compared to other existing techniques such as Loss Sensitivity Analysis (LSA), Grey Wolf Optimizer (GWO), Adaptive Shuffled Frogs Leaping Algorithm (ASFLA), and One Rank Cuckoo Search Algorithm (ORCSA). Keywords: Optimization, Distributed Generation (DG), Whale Optimization Algorithm (WOA), Sensitivity Analysis (SA)

Power flow analysis and volt/var control strategy of the active distribution network based on data-driven method

In order to solve the problems of low power flow calculation accuracy and voltage overcrossing of the distribution network caused by large-scale distributed power supply, a data-driven power flow analysis and volt/var optimization control strategy for distribution network are proposed in this paper. Firstly, a data-driven power flow analysis model for the distribution network is proposed, and the nonlinear mapping relationship between the distribution network state and power flow results is described from the data-driven perspective. Secondly, the paper analyzes the influence of photovoltaic power supply on the voltage of the distribution network, and establishes the volt/var optimization model based on photovoltaic power supply. Then, on the basis of data-driven power flow analysis of the distribution network, the distribution network reactive voltage optimization strategy based on data-driven power flow analysis is proposed to ensure the safe and stable operation of distribution network voltage, reduce the operating network loss of distribution network, and realize the distribution network reactive voltage optimization is not affected by factors such as distribution network line parameters. Finally, IEEE 33 node power distribution system is used to verify the effectiveness of the proposed strategy.

Research on multi-time scale Volt/VAR optimization in active distribution networks based on NSDBO and MPC approach

To explore the potential of all-round and multiangle voltage and reactive power control to reduce losses in a new power system with a high proportion of distributed resources connected to a distribution network, this study proposes a Volt/VAR optimization method for distribution networks using a nondominated sorting dung beetle optimizer (NSDBO) and model predictive control (MPC). According to the characteristics of the various adjustable resources in the network, they are divided into day-ahead and intraday optimizations. The optimization process in the day-ahead stage uses the NSDBO to create a discrete equipment scheduling plan. The optimization process in the intraday stage combines the discrete equipment scheduling plan with the MPC to create the scheduling plan for photovoltaics, energy storage, and vehicle charging stations. Two-stage optimization scheduling is achieved with the lowest discrete equipment regulation cost and minimization of network loss and node voltage deviation penalty cost as the optimization goal. The feasibility of this method for coordinating various resources in the network, processing discrete and continuous variables, and coping with the volatility and uncertainty of high-proportion distributed resources is verified through case analysis. The effectiveness of this method in improving the security and economy of distribution networks is demonstrated. The superiority of the solution speed and quality is also confirmed.

Automatic detection and segmentation of lesions in 18 F-FDG PET/CT imaging of patients with Hodgkin lymphoma using 3D dense U-Net.

The accuracy of automatic tumor segmentation in PET/computed tomography (PET/CT) images is crucial for the effective treatment and monitoring of Hodgkin lymphoma. This study aims to address the challenges faced by certain segmentation algorithms in accurately differentiating lymphoma from normal organ uptakes due to PET image resolution and tumor heterogeneity. Variants of the encoder-decoder architectures are state-of-the-art models for image segmentation. Among these kinds of architectures, U-Net is one of the most famous and predominant for medical image segmentation. In this study, we propose a fully automatic approach for Hodgkin lymphoma segmentation that combines U-Net and DenseNet architectures to reduce network loss for very small lesions, which is trained using the Tversky loss function. The hypothesis is that the fusion of these two deep learning models can improve the accuracy and robustness of Hodgkin lymphoma segmentation. A dataset with 141 samples was used to train our proposed network. Also, to test and evaluate the proposed network, we allocated two separate datasets of 20 samples. We achieved 0.759 as the mean Dice similarity coefficient with a median value of 0.767, and interquartile range (0.647-0.837). A good agreement was observed between the ground truth of test images against the predicted volume with precision and recall scores of 0.798 and 0.763, respectively. This study demonstrates that the integration of U-Net and DenseNet architectures, along with the Tversky loss function, can significantly enhance the accuracy of Hodgkin lymphoma segmentation in PET/CT images compared to similar studies.

Methodology for selecting optimal wire grade and cross-section based on an integrated technical and economic criterion

The aim was to develop a methodology for selecting an optimal wire grade and cross-section for overhead power lines with a voltage of above 1 kV under intensive development of power grid systems. This aim was solved using the authors’ previously developed methods: selection of an optimal wire grade based on hierarchy analysis and selection of an optimal wire cross-section by optimizing the specific discounted costs during the entire period of construction and operation of overhead power lines. In the present study, these methods are integrated into a single methodology. It is proposed to implement wire inspections prior to the stage of technical and economic calculations, since the necessary data has already been taken into account during the selection of a wire grade and its cross-section. The proposed methodology was tested on a typical example of reconstruction of the Zapadnaya– Davydovka 110 kV overhead line, which creates limitations in the power supply of Primorsky Krai due to insufficient wire capacity and its operation beyond the normative period. SENILEK AT3/C 150/24 wire was selected as a solution for the example under consideration. The application of this wire grade allows not only the overhead line capacity to be increased by 151% without replacing the existing supports, but also active and reactive power losses in the electric network to be reduced by 18.9 and 2.5%, respectively. The proposed methodology enables selection of an optimal grade and cross-section of any wire design, taking the dynamically changing operational conditions of power grid systems into account. The solutions found by the proposed methodology may contribute to a more efficient operation of power lines due to increasing their transmission capacity, reducing the number of used supports, replacing the line wire without replacing supports, decreasing ice formation on wires, and reducing power losses.

Physics-informed neural networks for localized assessment in Euler-Bernoulli beams

This study employs the Physics-Informed Neural Networks (PINNs) to address an inverse problem, specifically evaluating local changes in material properties within an Euler-Bernoulli beam through numerical simulations. In our numerical model, a viscoelastic material is strategically introduced to a localized zone of the beam, inducing non-constant and complex alterations in the storage modulus. The training process incorporates the partial differential equation into the neural network's loss function, coupled with essential stability conditions for the machine learning model, enhancing the efficiency of the training process. Notably, this methodology excels in characterizing structural behavior influenced by these intricate local changes in material properties, bypassing the need to identify the entire structure and consequently to deal with uncertainties related to boundary conditions. The effectiveness of the approach is validated through numerical simulations, underscoring its potential for applications in efficiently assessing and understanding local material alterations within complex structural systems.