Grid Optimization Research Articles

Research on Multi-Objective Reactive Power Optimization of Distribution Grid with Photovoltaics

With the introduction of large distributed photovoltaic (PV) power and electric vehicles, the inherent volatility of their output makes it difficult for traditional power grid structures and reactive power optimization methods to meet the needs of the safe operation of distribution grids and economic benefits. Therefore, a multi-objective reactive power optimization method for a distributed grid is proposed under a distributed PV power generation scenario. Aiming at the two objectives of the network loss and voltage fluctuation rate, the improved multi-objective particle swarm optimization algorithm is used to solve the model under the condition that the output of each device does not exceed the constraint and the optimal solution that can reduce the distribution grid loss and improve the voltage stability of the distribution grid is obtained. The simulation was conducted on the IEEE 33-node and 113-node distribution networks to verify the proposed method’s feasibility.

A Review of Synergies Between Advanced Grid Integration Strategies and Carbon Market for Wind Energy Development

The integration of wind energy into power systems is essential for achieving global decarbonization goals but poses significant challenges, including transmission losses, grid instability, and risks of wind farm disconnection during contingencies. This review focuses on advanced grid stability technologies, optimization strategies, and carbon trading mechanisms, proposing a synergistic framework to address these issues. By enhancing transmission efficiency and maintaining grid stability, these solutions reduce energy losses, contribute to carbon reduction, and create economic incentives through carbon credits. Moreover, optimization models enable wind farms to remain operational during severe faults, ensuring their active participation in carbon markets. This review connects recent technical advancements with economic and policy frameworks, offering a comprehensive pathway to achieving sustainable and stable power systems while maximizing the economic potential of wind energy.

Hierarchical design method for hybrid equipment used in friction stir welding

The proposed hybrid equipment for friction stir welding (FSW) focuses on balancing heavy load capacity and compact size. The kinematics and static compliance of the chosen mechanism are modeled and analyzed. Comprehensive objectives, including kinematic performance, static compliance, and overall machine quality, are considered. On this basis, a scale optimization domain and a compliance-constrained domain are established. A multi-layer grid optimization strategy is then implemented, progressing from the subsystem level to the entire machine to refine the design. Optimal parameters are identified through a detailed search in the size domain. This leads to the definition of three-dimensional design parameters for creating a digital prototype of the hybrid equipment, resulting in a structured hierarchical design method and process.

OPTO: Automated Optimization for Solid-State NMR Spectroscopy.

NMR spectroscopy presents boundless opportunities for understanding the structure, dynamics, and function for a broad range of scientific applications. Solid-state NMR (SSNMR), in particular, provides novel insights into biological and material systems that are not amenable to other approaches. However, a major bottleneck is the extent of user training and the difficulty of obtaining reproducible, high-quality experimental results, especially for the sophisticated multidimensional pulse sequences that are essential to provide site-resolved measurements in large biomolecules. Here, we present OPTO, a software operating environment that addresses these challenges and enhances the performance of many types of commonly utilized SSNMR experiments. OPTO is compatible with Varian OpenVnmrJ and Bruker Topspin, with a front-end graphical user interface that presents the instrument operator with access to powerful underlying optimization algorithms, including simplex and grid searches of the dozens of parameter settings required for optimal performance. Therefore, OPTO efficiently leverages instrument time and enables instrument operators to find optimal experimental conditions reliably. We demonstrate examples including improvements in (1) resolution, with an automated, global search of 21 shimming parameters to achieve a 12 parts per billion line width; (2) sensitivity, with searches and refinements of several cross-polarization conditions dependent on 16 parameters in triple resonance experiments; and (3) robustness, with results from protein samples on several spectrometers operating at different magnetic field strengths and magic-angle spinning rates.

Smart Grid Optimization Through Machine Learning and Deep Learning

Smart Grid Optimization Through Machine Learning and Deep Learning

Smart Grid Load Forecasting Models Using Recurrent Neural Network and Long Short-Term Memory

This paper presents two methods for managing electrical energy consumption and demand, with the objective of developing reliable and accurate forecasting models for smart electrical network energy consumption and optimization. The first method utilizes a recurrent neural network (RNN), while the second employs long short-term memory (LSTM) techniques. This approach builds upon previous studies that have explored the use of machine learning models for energy forecasting, but often with limited performance or the inability to capture long-term dependencies in the data. The study utilizes the Global Energy Forecast 2012 database - for the period from 2004 to 2008, with a focus on electricity consumption - to validate the performance of the proposed models. The R-squared (R2) score is used as the primary evaluation metric, with the LSTM model achieving a remarkable 90% R2 score, outperforming the RNN model's 80% R2 score. This is a significant improvement over previous studies, which have typically reported R2 scores in the range of 70-80% for energy forecasting models. Furthermore, the LSTM model demonstrates superior error rate performance, with a Mean Squared Error (MSE) of 4.345%, compared to the RNN model's 16.644% MSE. This highlights the ability of LSTM models to capture long-term dependencies in the data, which is crucial for accurate energy consumption forecasting, a limitation often observed in traditional RNN-based approaches. The findings of this study highlight the superior performance of the LSTM-based approach in accurately predicting energy consumption in smart grids, a crucial aspect for optimizing energy management and distribution. This contribution is particularly significant, as it showcases the advantages of LSTM models over traditional RNN techniques in the context of energy forecasting, providing valuable insights for researchers and practitioners in the field of smart grid optimization, where accurate forecasting is essential for efficient energy management and distribution.

Research on Digitalization and Efficiency Optimization of Power Grid Based on Big Data Analysis

With the increasing penetration of distributed intermittent energy into distribution networks, the self-healing problem of distribution networks faces significant challenges. The load level and demand response must be considered as critical factors affecting fault recovery. This paper proposes a fault recovery strategy that combines islanding division and network reconstruction. First, a distribution network model with a distributed energy storage system is established. To optimize the use of distributed energy resources, controllable loads that can respond to demand are prioritized, and high-priority loads are included in the islanded network after a fault. Based on the islanding division results, the remaining non-faulty power loss areas are restored through main network reconstruction. The improved whale optimization algorithm is employed to solve the problem. Simulation results demonstrate that load demand response is closely linked to the islanding process, and an optimal fault recovery strategy can be achieved by utilizing the distributed energy storage system and the main network.

Big data analytics for seasonal crop patterns: integrating machine learning techniques

This study addresses the challenge of predicting rice growing season lengths, crucial for agricultural planning in tropical regions. Climate variability and season timing create uncertainties in decision-making, and while machine learning is widely used in agriculture, a gap persists in integrating spatial-temporal data for accurate season length prediction and region-specific pattern analysis influenced by rainfall. Using a combination of Random Forest algorithms with hyperparameter optimization (grid search), and clustering techniques such as PCA, K-Means, and Hierarchical Clustering, this study analyzes key features such as the start of the season (SOS), end of the season (EOS), and their significance indicators (sig_sos and sig_eos). The findings reveal a strong correlation (0.98) between SOS and EOS, with an optimal growing season ranging from day 93 to day 207 (113.82 days). The Random Forest model, optimized with Grid Search, achieved a MSE of 28.9474 and an R2 of 0.8636, showing an outstanding predictive result. SHAP and LIME analyses identified sos and eos as the most influential predictors, while cluster analysis highlighted three distinct growing season groups characterized by variations in rainfall and seasonal stability. These results underscore the importance of understanding localized agricultural conditions and provide actionable insights for optimizing planting schedules, resource allocation, and climate adaptation strategies. By integrating advanced machine learning techniques with spatial-temporal data, this study establishes a foundation for improving agricultural resilience and sustainability in the face of climate variability

Optimized Power Distribution using Machine Learning for Load Forecasting, Fault Detection, and Voltage Regulation

The optimization of electric power distribution systems is crucial for enhancing efficiency, reliability, and sustainability in modern power networks. Traditional optimization methods often struggle to handle the complexity and variability of large-scale power distribution grids. With the advent of ML, new opportunities have emerged to address these challenges more effectively. This paper explores the application of machine learning algorithms in optimizing power distribution systems, focusing on load forecasting, fault detection, voltage regulation, and network reconfiguration. By employing supervised, unsupervised, and reinforcement learning techniques, ML models can process vast amounts of real-time data, identify patterns, and make accurate predictions for system performance enhancement. This study presents a comprehensive review of recent advancements in ML-based optimization techniques, emphasizing their ability to improve the accuracy of load demand forecasts and reduce energy losses. Moreover, it discusses the integration of smart grid technology with ML models to enable adaptive control strategies that can respond to dynamic power demands. Various case studies and simulation results are included to demonstrate the practical benefits of machine learning applications in electric power distribution. The findings suggest that incorporating machine learning into the power distribution framework can significantly boost operational efficiency, reduce downtime, and facilitate the transition to a more intelligent and sustainable power grid. This paper concludes with a discussion of the challenges and future prospects of ML in electrical grid optimization, such as scalability, data privacy, and the need for real-time computation.

Advanced integration of IoT and AI algorithms for comprehensive smart meter data analysis in smart grids

Abstract Smart grids utilize advanced communication, information, and control technologies to build a more friendly power supply system. However, the development of technology leads to the increasing complexity of smart meter (SM) data in the power grid, and analyzing SM data becomes difficult. Therefore, in order to realize the effective analysis and forecasting of users’ electricity consumption behavior, an SM data analysis technique integrating the Internet of Things and artificial intelligence is proposed, which contains three major techniques: meter reading systems, load clustering, and load forecasting, and at last the technique is validated. Finally, the technology is validated. The experimental results showed that the average value of the Davis-Balding index for the proposed method combining recursive graphs and convolutional autoencoders was 1.87, which was significantly lower than the average value of 2.35 for other methods. The average value of the Dunn index was 0.086, which was higher than the 0.065 for the comparison method. In the load forecasting model, the model that employs multi-task learning and a gated cycle unit demonstrates superior performance across all comparison schemes, with the lowest average absolute percentage error of 12.90%. This is significantly lower than the comparison model’s 15.35%, which highlights the enhanced efficacy of the proposed model. The study realizes the effective analysis and prediction of consumers’ electricity consumption behavior and provides new ideas and solutions for the optimization and management of smart grids.

A Fast Two-Dimensional Direction-of-Arrival Estimator Using Array Manifold Matrix Learning

Sparsity-based methods for two-dimensional (2D) direction-of-arrival (DOA) estimation often suffer from high computational complexity due to the large array manifold dictionaries. This paper proposes a fast 2D DOA estimator using array manifold matrix learning, where source-associated grid points are progressively selected from the set of predefined angular grids based on marginal likelihood maximization in the sparse Bayesian learning framework. This grid selection reduces the size of the manifold dictionary matrix, avoiding large-scale matrix inversion and resulting in reduced complexity. To overcome grid mismatch errors, grid optimization is established based on the marginal likelihood, with a dichotomizing-based solver provided that is applicable to arbitrary planar arrays. For uniform rectangular arrays, we present a 2D zoom fast Fourier transform as an alternative to the dichotomizing-based solver by transforming the manifold vector in a specific form, thus accelerating the computation without compromising accuracy. Simulation results verify the superior performance of the proposed methods in terms of estimation accuracy, computational efficiency, and angle resolution compared to state-of-the-art methods for 2D DOA estimation.

A photovoltaic power estimation model based on the improved local cloud occlusion index algorithm considering Sun block luminance

Abstract Accurate estimation of photovoltaic (PV) power generation is of great significance to the operation and optimization of the power grid. However, due to the rapid movement of clouds in a short period of time, the changes in PV power are intermittent and fluctuating. Nowadays, sky images have been widely used for PV power estimation, which can effectively extract overall or local cloud information. However, Sun halos are often mistaken for being part of clouds and lead to incorrect cloud coverage calculations. To solve this problem, this paper proposes a method to calculate the local cloud occlusion index based on an improved two-dimensional Gaussian function. In addition, a new parameter, Sun block luminance, is introduced to quantify the lighting conditions in the solar area. On the other hand, this paper constructs a novel hybrid model, which introduces asymmetric convolution and SE attention module to enhance the extraction ability of sky image features. The test results show that, under sunny and cloudy conditions, the RMSE, MAE and R 2 values of the proposed model are 0.562, 0.462 and 0.989 on sunny days, 3.119, 2.248 and 0.774 on cloudy days, respectively, proving that the proposed model is superior to benchmark models.

AI-powered Smart Grids: Energy Optimization

This paper aims at explaining how AI can augment smart grids toward improved energy management. Mainly, the research is concerned with using artificial intelligence together in order to optimize energy distribution, minimize the cost of running organizations and finally establish means to harness sustainable energy resources. Four AI algorithms are utilized in this study: During the smart grid, load forecasting, fault detection, and energy optimization, ANN, SVM, GA, and RL approaches can be used. Based on the experimental findings, it is shown that the ANN achieved a 92% forecast precision; on the other hand, the utility of both the SVM and GA algorithms in energy optimization ranged from 18–23%. The RL algorithm returned the best result and cut wastage by 32% and enhanced load balance efficiency. These results demonstrate that AI strategy performs better than the conventional energy management systems, particularly for accommodating renewable generation and dynamic grid loads. By optimising the utilisation of the grid, and utilising artificial intelligence, energy losses are kept to an absolute minimum, thus supplementing sustainable efforts. The study establishes that smart grid using AI technology is the innovative way of enhancing energy management and thus sustainable smart infrastructures. In future work, issues of data security and scalability will be targeted in order to enhance the application of AI in smart grids.

Freeform mirror design for beam shaping by a combined approach using Poisson-based grid optimization and a discrete cosine transform

In this study, we have designed freeform mirrors capable of transforming uniform irradiance distribution into complicated irradiance distribution. The design methodology incorporates two key algorithms: Poisson-based grid optimization and a discrete cosine transform (DCT). The grids on the target plane were optimized using Poisson-based grid optimization while maintaining uniformity on the incident plane. By obtaining grids for both input and target planes, we derived the normal vector field and utilized the DCT to calculate the sags of the freeform surface. Several examples have been devised to substantiate the validity of the approach. For Case 1 and Case 2, the results demonstrate that these freeform surfaces produce high-contrast and complicated irradiance maps with a notable proportion exceeding 4:1 in terms of irradiation intensity ratio between the graphic area and the background. In Case 2, when compared to the prescribed irradiance distribution, the normalized cross correlation values for generated irradiance distributions by the freeform mirror, both with and without sag error, are 94% and 96%, respectively. Furthermore, Cases 3 and 4 demonstrate that incident beams with complex irradiance distributions can be transformed into uniform irradiance profiles by the freeform mirror designed using the proposed method, achieving uniformity levels of 93% and 95%, respectively. It is noteworthy that the incident beam aperture in Case 4 exhibits an elliptical shape.

An enhanced guided stochastic search with repair deceleration mechanism for very high-dimensional optimization problems of steel double-layer grids

Finding reasonably good solutions using a fewer number of objective function evaluations has long been recognized as a good attribute of an optimization algorithm. This becomes more important, especially when dealing with very high-dimensional optimization problems, since contemporary algorithms often need a high number of iterations to converge. Furthermore, the excessive computational effort required to handle the large number of design variables involved in the optimization of large-scale steel double-layer grids with complex configurations is perceived as the main challenge for contemporary structural optimization techniques. This paper aims to enhance the convergence properties of the standard guided stochastic search (GSS) algorithm to handle computationally expensive and very high-dimensional optimization problems of steel double-layer grids. To this end, a repair deceleration mechanism (RDM) is proposed, and its efficiency is evaluated through challenging test examples of steel double-layer grids. First, parameter tuning based on rigorous analyses of two preliminary test instances is performed. Next, the usefulness of the proposed RDM is further investigated through two very high-dimensional instances of steel double-layer grids, namely a 21,212-member free-form double-layer grid, and a 25,514-member double-layer multi-dome, with 21,212 and 25,514 design variables, respectively. The obtained numerical results indicate that the proposed RDM can significantly enhance the convergence rate of the GSS algorithm, rendering it an efficient tool to handle very high-dimensional sizing optimization problems.

Deep weighted survival neural networks to survival risk prediction

Survival risk prediction models have become important tools for clinicians to improve cancer treatment decisions. In the medical field, using gene expression data to build deep survival neural network models significantly improves accurate survival prognosis. However, it still poses a challenge in building an efficient method to improve the accuracy of cancer-specific survival risk prediction, such as data noise problem. In order to solve the above problem, we propose a diversity reweighted deep survival neural network method with grid optimization (DRGONet) to improve the accuracy of cancer-specific survival risk prediction. Specifically, reweighting can be employed to adjust the weights assigned to each data point in the dataset based on their importance or relevance, thereby mitigating the impact of noisy or irrelevant data and improving model performance. Incorporating diversity into the goal of multiple learning models can help minimize bias and improve learning outcomes. Furthermore, hyperparameters can be optimized with grid optimization. Experimental results have demonstrated that our proposed approach has significant advantages (improved about 5%) in real-world medical scenarios, outperforming state-of-the-art comparison methods by a large margin. Our study highlights the significance of using DRGONet to overcome the limitations of building accurate survival prediction models. By implementing our technique in cancer research, we hope to reduce the suffering experienced by cancer patients and improve the effectiveness of treatment.

Construction of inherent geometric pattern for equal-area hexagonal discrete global grid systems

ABSTRACT Discrete Global Grid Systems (DGGSs) are hierarchical frameworks that provide seamless global coverage and enable the efficient processing and analysis of heterogeneous geospatial data. As cell resolution becomes finer, similarities in cell attributes, such as cell size and shape, increase. Understanding these shared characteristics enhances our knowledge of DGGSs structures, improving their reliability and accuracy across a range of applications. However, existing methods for quantifying these characteristics are limited, often emphasizing overall analysis and visualization rather than detailed assessment. This paper introduces a geometric pattern quantification method for equal-area hexagonal DGGSs by calculating inter-level attribute similarities to identify a reference level, segmenting the spatial distribution image at this reference level, and fitting the boundaries of these sub-images to create iso-feature lines. The results show that this method improves the accuracy and efficiency of cell attribute calculations, with enhancements of at least 5.6 times for study regions and 13.8 times for sample points at lower levels. This pattern provides a robust basis for grid selection and optimization in regional applications, promoting the broader adoption of DGGSs across various fields.

Systematic Review of Real-Time Data Analytics in the Energy Sector: Advancing Efficiency and Sustainability

This study systematically reviews the role of real-time analytics in enhancing efficiency and sustainability in the energy sector, focusing on its applications, challenges, and future opportunities. Leveraging advanced technologies like IoT, big data, AI, and machine learning, real-time analytics enables instantaneous data processing and actionable insights, transforming energy management. Key applications include predictive maintenance, dynamic grid optimization, and renewable energy integration, which collectively enhance operational efficiency and reduce energy waste. For example, real-time monitoring reduces transmission losses, balances energy supply and demand, and ensures the stability of renewable systems like solar and wind energy. Real-time analytics also supports sustainability goals by minimizing carbon emissions, optimizing fossil fuel efficiency, and empowering consumers through smart meters to adopt energy-efficient behaviors. However, the review identifies significant challenges, including high implementation costs, data security risks, and a lack of skilled professionals, which hinder widespread adoption, particularly in developing regions. The review emphasizes the need for investments, workforce training, and secure data systems to address these barriers. Emerging technologies such as blockchain, edge AI, and quantum computing present promising opportunities to overcome current limitations. Blockchain facilitates secure, decentralized energy transactions, while edge AI enhances scalability and reduces latency in distributed systems. Quantum computing, with its unparalleled processing power, could revolutionize energy.

Multi-charging Station access distribution network grid planning and optimization method based on MRFO algorithm

Abstract The traditional multi-charging station integration into the distribution network planning and optimization model is a regional optimization process with low overall efficiency, increasing the total optimization cost. Therefore, a multi-charging station access distribution network based on the MRFO (Manta Ray Foraging Optimization) algorithm is proposed. The MRFO calculation method was adopted to improve overall efficiency, and the design of the MRFO calculation multi-charging station access distribution network planning optimization model was completed. Optimization processing and multi-objective validation are used to achieve grid optimization. The test results show that the MRFO calculation and distribution network planning optimization method designed in this study ultimately achieve a relatively low total optimization cost, better optimization effect, and practical application value.

Wind power prediction based on deep learning models: The case of Adama wind farm

Wind power prediction based on deep learning models: The case of Adama wind farm