Electrical Engineering最新文献

In the realm of economics, agriculture holds supreme importance. The Internet of Things (IoT) is now pivotal in agriculture, aiding farmers in monitoring crop yield. Smart meters and control methods streamline agricultural operations, managing intelligent equipment, bidirectional communication, and user interaction. Data from sensors capturing soil and environmental parameters like moisture, humidity and temperature are integrated into Neural Networks for predictive analysis. Water scarcity, irrigation, and electrical power utilization creates impact on global crop growth and quality. This paper introduces an IoT-enabled product for coconut farming, enabling real-time monitoring and control of irrigation, energy usage, and power quality. The Smart Agriculture Irrigation Management System (AIMS) monitors valves, pumps, water levels, soil, and environmental conditions autonomously. Users can implement automated or manual decision-making processes. Additionally, a Smart Agriculture Energy Management System with integrated Smart Agriculture Energy Meter monitors power consumption, Power Quality, anomalies, and disturbances, notifying farmers via cloud services with predicted values. Implemented in a coconut farm in Sirumalai, Tamil Nadu, India, the system aims to reduce manual stress, enhancing productivity, yield, and water saving by over 30%. Predicted energy consumption patterns and tariffs help farmers avoid excessive costs, resulting in around 40% energy savings, facilitated by the superior performance of RNN-LSTM model over traditional methods.

Modern energy management systems play a crucial role in integrating multiple renewable energy sources into electricity grids, enabling a balanced supply–demand relationship while promoting eco-friendly energy consumption. Among these renewables, wind energy, with its environmental and economic advantages, poses challenges due to its inherent variability, demanding accurate prediction models for seamless integration. This paper presents an innovative hybrid deep learning model that integrates a gated recurrent unit (GRU)-based attention mechanism neural network for wind power generation forecast. The developed model’s performance is compared against six other models, comprising four deep learning approaches—long short-term memory (LSTM), 1D convolutional neural network, convolutional neural short-term memory (CNN-LSTM), and convolutional long short-term memory (ConvLSTM)—as well as two machine learning models—random forest and support vector regression. The proposed GRU-based attention model demonstrates superior performance, particularly in 1-step to 3-step ahead predictions, with mean absolute error values of 59.45, 114.95, and 176.06, root mean square error values of 109.03, 201.83, and 296.55, normalized root mean square error values of 0.080, 0.148, and 0.218, and coefficient of determination (R2) values of 0.992, 0.975, and 0.948, for forecast horizons of 10, 20, and 30 min, respectively. These results underscore the robust predictive capability of the proposed algorithm. Significantly, this research constitutes the first application of the hybrid GRU-based attention model to the Yalova wind turbine dataset, achieving better accuracy when compared to prior studies utilizing the same data.

To effectively prevent cascading failures during the economic dispatch of AC/DC systems, this paper proposes a preventive control strategy that considers both safety and economic factors and establishes a nonlinear bilayer optimization model. First, based on the action characteristics of relay protection, a mathematical method for determining cascading trips is presented. Second, using the active power output of units as a control variable, a static security margin index that measures the safety level of AC/DC systems is established. An economic indicator is then given by comprehensively considering network loss, unit power supply cost, and environmental governance cost. To address the issues of the sparrow search algorithm falling into local optima and insufficient convergence accuracy, an improved sparrow search algorithm is proposed by integrating improved Circle chaos, spiral search, the Levy flight strategy, and mutation perturbation. Standard test functions are used for comparative analysis with other algorithms to demonstrate the effectiveness of the proposed algorithm. Next, this paper employs this algorithm to solve the aforementioned bilayer preventive control model using a Jacobian matrix preconditioning method combined with sparse storage technology in power flow calculations to improve computational efficiency. Finally, the improved IEEE 39-bus system is used for simulation analysis of the proposed algorithm and model, verifying the feasibility of the proposed strategy.

The problem of charging unmanned aerial vehicles (UAVs) is an important application area for wireless power transmission (WPT). However, the flight characteristics of UAVs make it difficult for the traditional single-coil coupling mechanism to ensure alignment between the transmitting and receiving ends. It also has poor anti-skewing performance and a cumbersome optimization process, which ultimately decreases the efficiency of charging UAVs. To address this problem, this paper takes the 2 × 2 square coil array as research object, analyzes the distribution characteristics of the axial magnetic field at a specific height, proposes a "cross" compensation structure, and adopts genetic algorithm to optimize the side length and turns of the compensation coil. The goal is to significantly improve the axial magnetic field uniformity in the charging area and reduce the degree of magnetic field leakage. Finally, the effectiveness and superiority of the proposed "crossover" compensation structure and its optimization method are demonstrated through the construction and testing of the system experimental platform, which significantly improves the anti-offset performance and transmission efficiency of the WPT system.

This paper proposes an approach for the single-ended and the double-ended traveling wave-based fault location algorithm using the empirical mode decomposition associated with the Teager energy operator to extract characteristic data from the faulted voltage signals of an overhead transmission line. The simulation of the power system uses the JMarti line model, with an ideally transposed transmission line, and it was carried out using the alternative transients program (ATP) software. Subsequently, the MATLAB software was used for extracting the traveling wave arrival times and to perform the single-ended and the double-ended fault location algorithms for all simulated scenarios in ATP. The numerical and graphical results prove that the proposed methodology with the Teager energy operator and the double-ended analysis can better extract the characteristic data of the voltage signals and estimate the fault location with good accuracy, with percentage error of 0.034% for the best results, depending only on the fault type and the sampling rate adopted.

The increasing complexity of modern power systems due to the integration of prosumers, renewable energy sources, and energy storage, has significantly complicated system organization and planning. Traditional centralized power plants are being replaced by decentralized structures, making the power flow more complex to predict. As a result, alternative methodologies for power system planning are imminent. This paper introduces a novel approach using a combination of artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models for forecasting system states. Here, ANN model predicts energy consumption, while the ANFIS model forecasts thermal and hydro power plant production as well as CO₂ emissions. The accuracy of these models results from leveraging the collective expertise of power system planning professionals, utilizing extensive databases containing hourly data from measurements in Serbian power systems. These datasets encompass hourly production data from various energy sources, energy consumption patterns, and relevant environmental parameters (such as temperature, wind speed, and solar irradiation). To underscore the effectiveness of the proposed ANN model, predictions of power consumption from ANN are compared with predictions from ARIMA (autoregressive integrated moving average) model. The developed forecasting models are employed to predict annual and daily energy consumption, seasonal variations in thermal and hydro production, and annual CO₂ emissions. The dependencies between power consumption/production and ambient parameters are visually depicted in three-dimensional representations. Model accuracy is evaluated through graphical, numerical, and error-based analyses across four distinct error metrics. By utilizing historical data and expert insights from previous production scheduling, these models enhance the precision of future production scheduling decisions. This approach minimizes human error, maximizes the utilization of human expertise, and establishes a framework for effectively planning large-scale power systems. The primary contribution of this research lies in the integration of artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) methodologies. This combined approach minimizes the errors inherent in each individual methodology while leveraging their respective advantages. Specifically, the consumption prediction error achieved is 5.64%. When ANFIS is utilized with a training database based on ANN consumption prediction, the prediction error for CO₂ emissions is 1.27%.