Search

Power system protection and asset management present persistent technical challenges, particularly in the context of the smart grid and renewable energy sectors. This paper aims to address these challenges by providing a comprehensive assessment of machine learning applications for effective asset management in power systems. The study focuses on the increasing demand for energy production while maintaining environmental sustainability and efficiency. By harnessing the power of modern technologies such as artificial intelligence (AI), machine learning (ML), and deep learning (DL), this research explores how ML techniques can be leveraged as powerful tools for the power industry. By showcasing practical applications and success stories, this paper demonstrates the growing acceptance of machine learning as a significant technology for current and future business needs in the power sector. Additionally, the study examines the barriers and difficulties of large‐scale ML deployment in practical settings while exploring potential opportunities for these tactics. Through this overview, insights into the transformative potential of ML in shaping the future of power system asset management are provided. [ABSTRACT FROM AUTHOR]

Due to intermittent renewable energy and fluctuating load demand, distribution networks with renewable distributed generation (DG) installations are more likely to suffer voltage issues and significant power losses. The performance of conservation voltage regulation (CVR) schemes may be adversely affected by the undesirable voltage profile at specific nodes. This paper aims to reduce power losses in CVR‐implemented networks by optimally planning new renewable DGs without changing the existing ones. A scenario‐based optimal renewable DG planning model is proposed with a novel scenario formation method. The uncertainties of load demand and renewables are captured jointly and formed into a finite number of scenarios based on a multivariate Gaussian mixture model (MultiGMM). The locations and capacities of different types of new renewable DGs are optimally planned for CVR performance improvements on power loss saving by aggregating the operation status and probabilities of the scenarios using mixed‐integer non‐linear programming (MINLP). A time‐series simulation is formulated for accuracy verification. The results of case studies show that the proposed model can significantly reduce power losses, active load demand, and reactive load demand. The accuracy of the planning results can be guaranteed with fewer scenarios compared to a widely used classical scenario‐based planning method. [ABSTRACT FROM AUTHOR]

Contemporary distribution networks can be seen with diverse dispatchable and non‐dispatchable energy resources. The coordinated scheduling of these dispatchable resources with non‐dispatchable resources can provide several techno‐economic and social benefits. Since battery energy storage systems (BESSs) and microturbine units are capital intensive. A thorough investigation of their coordinated scheduling on a purely economic basis will be an interesting and challenging task while considering dynamic electricity price and uncertainty of renewable power generation and load demand. This paper proposes a new methodology for optimal coordinated scheduling of BESSs and microturbine units considering existing renewable energy resources and dynamic electricity price to maximize daily profit function of the utility. In this study, a recently explored modified African buffalo optimization algorithm is employed. The key attributes of the proposed methodology are comprised of mean price‐based adaptive scheduling embedded within a decision mechanism system to maximize arbitrage benefits. Decision mechanism system keeps a track of system states as a‐priori thus guides the artificial intelligence‐based solution technique for sequential optimization. This may also reduce the computational burden of complex real‐life engineering optimization problems. Further, a novel concept of fictitious charges in coordination with BESS management algorithm is proposed to restrict the counterproductive operational management of BESSs. The application results investigated and compared on a benchmark 33‐bus test distribution system highlights the importance of the proposed methodology. [ABSTRACT FROM AUTHOR]

This study proposes an optimal approach for enhancing the quality of a weak electrical power distribution system by utilizing thyristor‐controlled series capacitors (TCSC). The proposed approach considers various system components, including shunt capacitors and distributed generators that rely on renewable energy sources, which are characterized by uncertainty due to changing weather conditions. The proposed system maximizes the benefits of these components by updating its parameters to improve the distribution power system's quality under all operating conditions, which change due to the varying energy generated by the distributed generators and load changes. The optimal TCSC values are calculated for all operating conditions using the Hyper‐Spherical Search with graded objective functions management strategy (HSSA‐GOFMS) algorithm. The GOFMS converts all restrictions into objective functions with limits and prioritizes the search for objectives according to a sequential system of priorities. This strategy prevents the possibility of non‐convergence. The results demonstrate the effectiveness of the proposed approach in improving the performance quality of a weak electrical distribution system. The proposed strategy achieves the target voltage level limits within nominal values, even with the interference of the system with generators distributed at variable power levels. Furthermore, the proposed approach enhances the performance quality of the system under a wide range of load changes. [ABSTRACT FROM AUTHOR]

For the past few years, the addition of renewable energy sources has been on the rise, but the unregulated addition of these sources can cause severe harm to the grid. Therefore, it is necessary to have a predefined limit for a grid, beyond which no further addition of renewables should be allowed without reinforcement. That limit is called the hosting capacity (HC), which is addressed in the literature by search‐based methods with heavy computational burdens. This manuscript first presents a framework to find the HC for a grid. The same framework is then applied to 503 different simulated but realistic LV distribution feeders in Finland to find their HCs and the most effective network parameters in defining a network specific HC. Next, different machine learning models, that is, Decision Tree (DT), Random Forest (RF), Linear Regression (LR), K nearest neighbours (KNN), Logistic Regression, and Support Vector Machine (SVM) are implemented on the generated data. For the classification case, the accuracy values for logistic regression, KNN, and SVM were 0.89, 0.84, and 0.81, respectively. The findings demonstrate that the developed machine learning based technique will enable distribution network operators in finding the HC without applying any deterministic or probabilistic approaches. [ABSTRACT FROM AUTHOR]