Traditional methods of asset management in electric power systems rely upon fixed schedules and reactive measurements, leading to challenges in the transparent prioritization of maintenance under evolving operating conditions and incomplete data. In this paper, we introduce a new, fully integrated artificial intelligence (AI)-driven approach for enhancing the resilience and reliability of open-source asset management tools to support improved performance and decisions in electric power system operations. This methodology addresses and overcomes several significant challenges, including data heterogeneity, algorithmic limitations, and inflexible decision-making, through a three-module workflow. The data fidelity module provides a domain-aware pipeline for identifying structural (missing) values from explicit missingness using sophisticated imputation methods, including Multiple Imputation Chain Equations (MICE) and Generative Adversarial Network (GAN)-based hybrids. The characterization module employs seven complementary weighting strategies, including PCA, Autoencoder, GA-based optimization, SHAP, Decision-Tree Importance, and Entropy Weighting, to achieve objective feature weight assignment, thereby eliminating the need for subjective manual rules. The optimization module enhanced the action space through multi-objective optimization, balancing reliability maximization and cost minimization. A synthetic dataset of 100 power transformers was used to validate that the MICE achieved better imputation than other methods. The optimized weighting framework successfully categorizes Health Index values into five condition levels, while the multi-objective maintenance policy optimization generates decisions that align with real-world asset management practices. The proposed framework provides the Transmission and Distribution System Operators (TSOs/DSOs) with an adaptable, industry-oriented decision-support workflow system for enhancing reliability, optimizing maintenance expenses, and improving asset management policies for critical power infrastructure.

Conventional capacity expansion planning (CEP) relies on a perfect-foresight planning horizon and linear investment optimisation, which fail to capture the non-linear dynamics of electricity markets. In the Nordics, hydro-related weather variability plays a critical role in maintaining the robustness of the power system. This paper addresses the intra-year perfect-foresight limitation in current CEP models, focusing on hydro-dominated power systems with substantial hydro reservoir capacity, using Sweden's decarbonisation pathway toward 2050 as a case study. Our approach provides a robust long-term CEP framework by leveraging short-term price forecasts to guide storage dispatch decisions. The proposed CEP model has been historically validated and captures the dynamics of seasonal storage hydro reservoirs, achieving deviations of less than €1/MWh in annual average prices across all Swedish bidding zones. A comparative analysis between the proposed and conventional CEP models (cGrid and GenX), together with the Ten-Year Network Development Plan (TYNDP 2024), reveals a broad alignment in capacity expansion and dispatch under an average weather year. However, in a problematic weather year, with correlated low wind output and reduced hydro inflows, significant divergences emerge, with half-year price averages differing by up to ±€40/MWh. These discrepancies are mainly driven by contrasting approaches to hydro reservoir modelling. Notably, the proposed CEP model recommends a 37.5 % increase in firm nuclear capacity to mitigate supply shortages, whereas the conventional CEP suggests a 4.4 % reduction, thereby increasing reliance on weather-dependent resources. These findings underscore the limitations of perfect-foresight CEP in power systems with substantial seasonal storage resources.

This paper introduces a novel hybrid optimization algorithm, Adaptive Hybrid PSO-Embedded GA (AHPEGA), which dynamically adapts to optimization performance by integrating Particle Swarm Optimization (PSO) and Genetic Algorithms (GA). The primary objective is to enhance the neuroevolutionary training of multilayer perceptron-based controllers (MLPCs) through the joint optimization of model parameters and structural hyperparameters. Traditional training methods frequently encounter issues such as premature convergence and limited generalization. AHPEGA addresses these limitations through an adaptive training strategy that dynamically adjusts parameters during the evolutionary process, thereby improving convergence speed and solution quality. By effectively reducing entrapment in local minima and balancing exploration and exploitation, AHPEGA improves the quality of neural controller design. The algorithm's performance is evaluated against conventional optimization methods, demonstrating significant improvements in accuracy, convergence speed, and consistency across multiple runs. The practical applicability of the proposed method is demonstrated through simulation in the context of a VSC-based islanded microgrid (MG), where ensuring reliable and effective control under variable operating conditions is critical. This highlights AHPEGA's capability to optimize intelligent control strategies in MG systems, particularly under dynamic and uncertain conditions, reinforcing its practical value in real-world energy environments.

This Study introduces an open-source toolbox for asset management in power systems developed under the European ATTEST project. This paper focuses on presenting an open-source toolbox for Transmission and Distribution System Operators (TSOs and DSOs) to improve the reliability and efficiency of power networks, including a solution to the difficulties faced by the power industry, such as the aging infrastructure and the growing need for renewable energy integration. The toolbox uses predictive analytics and machine learning to evaluate the health of assets, enhance maintenance plans, and guarantee efficient resource distribution. It evaluates the condition of power grid assets through clustering (K-means, SOM) and reinforcement learning (Q-learning), providing actionable insights for improving asset management. This approach allows TSOs and DSOs to adopt proactive maintenance strategies, reducing the risk of failures, minimizing downtime, and extending the lifespan of critical infrastructure. The toolbox provides actionable insights for planning maintenance strategies and optimizing resource allocation. Scalability tests were conducted using a synthetic power grid of 600 transformers alongside real-world data from five European electrical companies. Due to space constraints, only the results from 92 transformers. This research contributes to achieving sustainable power systems and supporting the energy transition by focusing on intelligent asset management.

In this research, a new approach called Particle Swarm Optimization–Proportional–Integral (PSO-PI) is proposed for optimizing the gains of current and voltage controllers as well as the LCL filter parameters in voltage source converter (VSC)-based islanded microgrids. The control problem is framed as an optimization task, where PSO optimally tunes parameters. Unlike conventional offline methods, this study employs a simulation-based online optimization framework, integrating PSO within SIMULINK environment for dynamic, iterative adjustments, enhancing adaptability and efficiency. Well-founded mathematical models define parameter bounds, ensuring a unity ratio between converter-side and coupling inductances, and setting the switching-to-resonance frequency ratio by considering converter-side and output ripple currents. The objective function is formulated to improve the tracking performance of the outer voltage and inner current control loops with respect to their reference signals while minimizing total harmonic distortion (THD) and maintaining an optimal balance between filtering effectiveness and system performance. The PSO-PI approach achieves a more compact LCL filter while complying with IEEE-519 standards and outperforms the conventional method (CM). Simulations validate its effectiveness under various disturbances, including load changes, faults, and parameter variations, demonstrating improved damping and robustness in VSC-based islanded microgrids. Notably, improved transient response is achieved, with a 61.80% reduction in settling time and a 51.34% decrease in overshoot. Integrating PSO within the SIMULINK framework enables dynamic fine-tuning of VSC parameters through simulation-driven optimization, highlighting its potential as a robust microgrid control strategy.

There is an unprecedented need to expand thetoolbox of solutions to boost the scalability of clean powerand energy systems. Amidst this challenge, nuclear energy isincreasingly recognized as an important player in the pathtoward deep decarbonization of the global energy mix. Thispaper presents a technology review of small modular reactor(SMR) concepts currently under development and deployment.Both conventional and next-generation reactor technologies areevaluated with a focus on potential power system integrationbenefits, added system values, and provision of system-bearingservices. Nevertheless, there are currently some uncertainties inthe techno-economic competitiveness of SMRs, whether they canleverage economics of mass production over their inherent lack ofeconomics of scale. To address this challenge, our paper providessome basic cost analysis of SMRs, taking into account theirexpected learning curves to evaluate the cost of future deploymentsand give insights into their economic competitiveness andpotential role as a disruptive solution in the energy transition.

Power grids have become a key component in the energy transition and decarbonization of industries due to the increasing electrification of different loads, such as heating. This paper presents the overview and validation of a new method for electricity retailers or Virtual Power Plant operators to match their production portfolio dominated by Renewable Energy Resources (RES) with electrical demand of thermal loads (heat pumps) via a price control mechanism. We propose a novel three-step framework to exploit the flexibility of Thermostatically controlled loads (TCLs) by (1) simulating a pool of households, (2) estimating the governing physical parameters of the aggregated households, and (3) controlling the heat pump electric power via a dynamic price policy; for parameter estimation and price policy a Differential Evolution (DE) optimization algorithm is used. The proposed method performs well for a large number of parameters and reduced training data (errors around 0.4% and 0.55% on the average load power and standard deviation) and effectively controls the loads through a dynamic price policy that reduces the total price for the household owner or customer compared to a tariff without demand response (DR) (reduction of up to 53.63% on average per house), respecting the technical constraints of the grid.

Few-Shot Short-Term Wind Power Forecasting (FS-STWPF) is designed to develop accurate short-term wind power forecasting models with limited training data, reducing the losses suffered by wind farms and power systems due to the data scarcity. Based on the idea of extracting valuable knowledge from the source wind farms and then applying it to the target wind farm, a novel Meta-Learning approach (WG-Reptile) has been proposed in this paper. Building on the existing Reptile algorithm, two specific designs have been made in WG-Reptile for FS-STWPF: (1) Within-Task Samples Assignment method based on Operational Scenario (WTSAOS) has been proposed to improve the adaptability of the models to changing conditions. (2) Gradients Conflict Attenuation method based on Cosine Similarity (GCACS) has been proposed to enhance the effect of knowledge fusion from different source wind farms. Two open wind power forecasting datasets and three deep learning models have been used to implement 24-h-ahead FS-STWPF experiments with different amounts of training data. The results illustrate that the proposed WG-Reptile is able to outperform the other few-shot learning approaches. Intuitively, with only 30-day training data, the accuracy of the proposed WG-Reptile can be equivalent to the conventional supervised learning approaches trained on 6-month.