Electrical Engineering最新文献

A printed circuit board (PCB) is one of the important components in every single electronic device, which assists in connecting each component for many purposes. Somehow, the PCB can be affected due to spurs, short circuits, mouse bites, and so on. Therefore, the detection strategy for such defects is very important and also complicated. So, this research concentrates on developing a deep learning model, a multi-level attention-based printed circuit board with a generative adversarial network, and a YOLOv5 (MuAP-GAN-YOLOv5) model for defect detection in PCB. The contribution of this research is to enhance image quality using the proposed multi-level attention-based PCB-GAN (MuAP-GAN) method, which is embedded with a multi-level attention mechanism to enhance image quality. Therefore, the model can efficiently learn and train for accurate defection as well as localize the defected area in PCB. Here, the YOLOv5 model plays an important role in training based on enhanced features and, therefore provides accurate results. In addition, this model requires less computational expenses, is quite reliable, also provides a maximum accuracy of 95.24% compared to other traditional methods.

Load frequency control (LFC) is necessary to maintain the power system frequency and tie line power to its nominal value. In modern power system, importance of LFC is increased due to inevitable use of communication channel, intermittent nature of renewable sources, computer-based control strategies, model uncertainties and cyber-attack. An effective LFC is required to mitigate various uncertainties and disturbances including the delay for which active disturbance rejection control (ADRC) control schemes have been explored in this work. An ADRC consists of an extended state observer and a state feedback controller. In the present work, predictive structure like Smith predictor has been proposed for different variants of ADRCs. Additionally, a new variant of ADRC, namely, reduced order generalized active disturbance rejection control (RGADRC) has been proposed along with the predictive structure. These controllers are designed considering system uncertainties and with or without non-minimum phase. To show the efficacy of the proposed schemes examples of single-area non-reheat, reheat, and two-area thermal and photovoltaic-wind micro-grid system are demonstrated. The robustness of the proposed approach is examined while taking system parameter variation, random fluctuation of solar power (0–0.001 p.u.), wind power (0–0.0012 p.u.), and load disturbance (0–0.01 p.u.), and cyber-attack (2 p.u.). The predictive RGADRC shows superior performances compared with other predictive ADRCs as well as some methods prevalent in the literature for LFC systems with nonlinearities like generation rate constraint of 0.1 p.u./min, governor dead band of 0.05%, and communication delay of 2.28 s. The predictive RGADRC maintains stability of the LFC system with satisfactory transient for + 50% change in gain and time constant of the generator and load, along with random fluctuations as mentioned above.

This paper presents an innovative method for forecasting power consumption in the power system using an artificial neural network (ANN). The method was validated in the case of predicting power consumption for the Sarajevo region in Bosnia and Herzegovina. Power consumption is planned daily for the day-ahead with hourly resolution. Measured data on air temperature, wind speed, and insolation for 2017 to 2020 were utilized as input variables in the proposed power consumption forecasting method. The influence of these input variables on power consumption was analyzed using the Pearson correlation coefficient. The neural network underwent training with data on input variables and power consumption from 2017 to 2020 and was subsequently applied to forecast day-ahead power consumption for 2021. Due to the implementation of a neural network with a greater number of input variables, a smaller error in the power consumption forecast for 2021 was achieved compared to the forecast performed by the Electric Power Company. Therefore, the proposed method can be used as a more reliable tool for day-ahead power consumption forecasting. Additionally, the continual increase in the historical data on power consumption and influencing variables over time is expected to further enhance the reliability of power consumption forecasting using ANN.

This paper presents a new algorithm designed for the allocation of daily energy losses in radial distribution networks (DNs) featuring distributed generations. The algorithm employs a branch-oriented approach, widely acknowledged as the most effective method for addressing issues in radial DNs. To implement this methodology in a DN environment, a current summation algorithm is utilized. The statistical approach is employed to organize the complete data of daily load and generation curves, providing key quantities necessary for determining energy loss allocation. By leveraging the statistical characteristics, the proposed method calculates the single equivalent values of the collected data from the daily load and generation curves. These values are then used to perform power flow calculations and subsequently allocate the energy losses. To validate the effectiveness of the proposed method, tests are conducted on 33-node and 69-node radial DNs under load levels of 30%, 100%, and 150%. The obtained results were compared with those from the energy summation method of loss allocation and the repetitive geometric scheme of loss allocation method. Under all three load levels, the proposed method demonstrated an unbiased approach, avoiding the addition of economic burden of losses over groups of consumers and generators individually. Moreover, the performance of the proposed method was found to be well-suited and highly acceptable. This underscores the reliability of the proposed approach, achieving results within an acceptable error margin of 10%.

This article explores the influence of energy storage devices (ESDs) like battery storage devices, aqua-equalizer-based fuel cells (FC) and electric vehicles as secondary sources for the improvement of frequency regulation of a dual-area hybrid power system (d-HPS) for its outstanding disturbance rejection capability. The d-HPS is a hybrid system, integrated with wind, solar and tidal systems, and a reheater interfaced thermal system enabled with GRC-GDB nonlinearities. Due to uncertain solar, wind and tidal power injection, primary and secondary load frequency control (LFC) frequently approaches incompetency in mitigating the power and frequency deviations due to inadequacy of controller action. To improve the inadequacy of the controller, the fractional-order scaled interval type 2 fuzzy PID controller (FO-T-II-FPID) control approach is integrated with LFC loops, which is optimally scaled by the improved equilibrium optimization algorithm (i-EOA) overcoming population diversity and local trapping issues with basic EOA. Furthermore, with the insertion of ESDs in steps the frequency responses are improved marginally. The efficacy of the i-EOA scaled FO-T-II-FPID controller is authenticated by contrasting it against some recent research approaches. Lastly, conferring to the sensitivity analysis and stability analysis, the proposed frequency regulation approach with ESDs is found to be an inventive contrast to parameters alternation, random loading conditions and uncertain power injection.

This paper introduces an approach that enhances the computational efficiency of reliability assessment for composite power systems by integrating machine learning (ML) techniques with sequential monte carlo simulation (SMCS). Integration of renewable energy resources (RERs) into power systems is increasing at a rapid pace. Evaluating the reliability of composite power systems is helpful in identifying any deficiencies in their operation. As power systems operation becomes more fluctuating and stochastic, it is necessary to update the tools used to analyse reliability. In this paper, SMCS is used as a conventional method, as it provides results by taking chronological nature of RERs. However, SMCS is highly computational. ML models fit for solving complex problems that require computational power. ML techniques, such as convolutional neural network (CNN) and hybrib models of Convolutional and Extreme Gradient Boosting (ConXGB), and Convolutional and Random Forest (ConRF) are proposed to determine the expectation of load curtailment and minimum amount of load curtailments. The proposed technique is applied on test system IEEE RTS-79. Results indicate the ConvXGB method is fast and accurate in computing composite reliability indices. For instance, it achieved a Loss of Load Probability (LOLP) of 0.0025 and an Expected Demand Not Supplied (EDNS) of 0.1850 MW, compared to SMCS’s LOLP of 0.0021 and EDNS of 0.1794 MW while reducing computational time from 12900 to 5414 s. These results confirm the proposed method’s speed and accuracy, making it a robust solution for modern power system reliability evaluation.

This paper explores the dynamics of large-scale offshore wind farms comprised of full-power variable frequency wind turbines, interconnected with VSC-HVDC (voltage source converter-based high-voltage direct current) converter stations. These systems are susceptible to broad frequency oscillations due to the rapid response characteristics of power electronic devices, potentially compromising their operational safety and stability under various conditions. To mitigate the impact of these oscillations on offshore wind turbines and the connected systems, the study first outlines the structure and operational mode of the offshore wind power electronic system with VSC-HVDC transmission. It then analyzes the mechanisms underlying these broad frequency oscillations. Subsequently, the paper presents a model construction and stability analysis for wind farms and transmission systems. It specifically focuses on offshore wind power systems based on symmetric monopolar topology, involving multiple branches and multiple wind farm access points. The research includes an in-depth oscillation analysis, supported by real-world case studies, demonstrating that strategically optimized control and protection strategies can effectively reduce the oscillation risks associated with wind farms connected through VSC-HVDC systems, thereby ensuring their safe and stable operation.

The excessive electric field on the surface of DC gas-insulated metal-enclosed transmission lines (GIL) basin insulators is one of the main factors leading to insulation failure. In this paper, we parameterized and reconstructed the shape of the insulator based on the Bernstein polynomial under the Bessel curve and established an optimization model for the geometric shape of 500 kV DC GIL insulators considering the surface charge accumulated on the insulator under temperature gradient. We obtained the optimal parameters of the contour function of the basin insulator surface using the Levenberg–Marquardt optimization algorithm and explored the optimization effect of insulators by the insulation tests. The results show that the optimized basin insulator has a maximum electrical strength of 4.39 kV/mm along the insulator surface, and the flashover voltage of the optimized insulator is 15.37% higher than that of the original structure, laying a foundation for the production of a new type of high-electrical performance basin insulator.

This paper presents two improved and adaptive models based on functional neural network and Autoregression (FNNAR) analysis. These models have been developed for estimating the fundamental component of nonlinear and varying load current and computing the exact compensation required in a power distribution system. The proposed FNNAR analysis involves two steps: The first step is designed to estimate the fundamental current in terms of polynomial or trigonometric functional expansion terms; while, the second step involves computations based on the weighted sum of the delayed output terms. An activation function is additionally incorporated to account for the nonlinearity and sudden variations of load current. Both the FNNAR models are developed and their parameters computed in an adaptive manner from the input–output data. The simulation results on a single-phase 110 V, 50 Hz system power distribution system are validated by a scaled down experimental model showing hardware results depicting load compensation. Adequate comparison of the two developed models is also discussed in the paper with two advanced variants of conventional algorithms viz. Least means square algorithm and second order generalized integrator based filtering technique.

The integration of distributed generators (DGs) into distribution networks has the potential to decrease network power losses, provided that DGs of suitable capacity are strategically positioned. In this regard, this paper proposes an optimal combination of a novel analytical and meta-heuristic method for the appropriate placement and sizing of dispatchable and renewable generators in an active distribution network with a preset power exchange contract with the main grid. A fuzzy framework embedded in a mixed-discrete grey wolf optimizer is adopted to find the accurate locations and capacities of renewable DGs, whereas a novel distributed zero bus technique is orchestrated to get the proper sizes of the required number of dispatchable biomass generators simultaneously. The proposed planning problem takes care of the intermittent attributes of renewable sources using the worst-case realization approach. The trade-off among multi-objectives, such as reduction in active power loss, improvement in node voltage profile, and curtailment in annualized DG costs, is achieved using fuzzy max-min composition. The economic viability of the obtained solutions is evaluated by a cost-benefit analysis. The efficacy of the suggested strategy is tested on a 69-bus distribution network. Additionally, the outcomes are compared with the already existing solutions in the literature.