Physics-Informed Neural Networks (PINNs) have emerged as a powerful tool for modeling and controlling complex energy systems by embedding physical laws into deep learning architectures. This review paper highlights the application of PINNs in grid-connected inverter systems (GCISs), categorizing them by key tasks: parameter estimation, state estimation, control strategies, fault diagnosis and detection, and system identification. Particular focus is given to the use of PINNs in enabling accurate parameter estimation for aging and degradation monitoring. Studies show that PINN-based approaches can outperform purely data-driven models and traditional methods in both computational efficiency and accuracy. However, challenges remain, mainly related to high training costs and limited uncertainty quantification. To address these, emerging strategies such as advanced PINN frameworks are explored. The paper also explores emerging solutions and outlines future research directions to support the integration of PINNs into practical inverter design and operation.

The control of closed architecture industrial optical tweezers systems presents substantial challenges due to the limited knowledge of inner controller configurations and control gain structures. Conventional methods, such as the computed-torque approach, often prove inadequate for closed architecture robots because the torque terminal is inaccessible to end-users. This creates a disconnect between advanced control algorithms and practical industrial needs. To tackle these challenges, we propose a novel approach utilizing adaptive neural network-based control methodology. Our method employs neural networks to approximate the low-level controller, thereby bypassing the constraints of traditional closed-loop architectures. By integrating an external feedback loop that operates independently of the inner control architecture, our approach offers flexibility for design adjustments and facilitates advanced control actions. Furthermore, we conduct a comprehensive stability analysis of external feedback control systems interacting with a fixed inner loop. In summary, this study represents a pioneering effort in addressing the complexities of industrial optical tweezers systems with closed uncertain architectures, providing an innovative solution aimed at enhancing control capabilities for applications in industrial settings.

In this paper, an intelligent integrated approach is proposed to control the reactive power and restore the voltage balance in a three-phase power system using particle swarm optimization (PSO), Gaussian process regression (GPR), and support vector machine (SVM). The PSO algorithm is used in offline mode to determine the optimal set of firing angles for the thyristor-controlled-reactor (TCR) compensator according to the smallest fitness value required for voltage balancing. The optimum firing angles are then used to train the GPR and SVM regression models. The GPR and SVM models are finally used as a real-time controller to retrieve the voltage balance in online mode. A simulation model and experimental setup of the electrical power system are built. The modeled system consists of a 500 km long transmission line. The line is divided into three-pi sections to guarantee a real system response. Several simulation and practical case studies have been conducted to test and validate the capability of the proposed integrated approach in solving the voltage unbalance problem. The results have revealed the supreme ability of the proposed integrated approach to restore the voltage balance quickly (within 20 ms) and for a wide range of voltage unbalance factors (VUFs) (3.90–8.42%).

Transmission lines in the desert are exposed to the desert environment, which includes sandstorms as one of its hallmarks. A conductive layer develops with prolonged sand deposition and the presence of moisture, ambient humidity, and dew. The ensuing leakage current causes surface discharge, which limits the life of the insulator and interrupting the power supply. The locations of power lines in the Egyptian Sinai desert, where sandstorms are known to occur frequently, are exposed to such a risk. In order to estimate the danger of insulator failure, this paper studies the flow of leakage current on porcelain insulators that have been contaminated with sand. This work relies on accurate data collected and published in a prior study regarding Sinai, which mainly focused on contaminating sands grain sizes. Porcelain insulator is simulated using finite element method to determine the leakage current that results on its contaminated surface. The probabilistic characteristics of the insulators leakage current are derived using Monte Carlo technique, allowing for the risk assessment of insulator failure. This assessment can be used to justify the suitability of using this kind of insulator in Sinai.

The growing deployment of renewable energy systems calls for accurate parameter estimation and compact modeling of grid-connected inverter systems (GCISs). This paper presents a framework for parameter estimation and physics-structured model compression using Physics-Structured-Informed Neural Networks (Ψ-NNs). A teacher physics-informed neural network (PINN) estimates LC-filter parameters by combining measurements with governing equations. A regularized student learns the teacher mapping through physics-informed distillation. Hierarchical clustering identifies shared weight magnitudes, enabling structured reconstruction and fine-tuning. The framework is evaluated on mathematical-model and MATLAB/Simulink datasets. The teacher achieves a mean relative parameter error below $5\%$ across all cases, outperforming the extended Kalman filter, remaining competitive with least squares on mathematical-model data, and outperforming both baselines on all MATLAB/Simulink records. The reconstructed networks retain only 23–78 trainable base coefficients (0.15–0.50% of the student’s 15,600 parameters): 23–55 (0.15–0.35%) on the mathematical-model records and 62–78 (0.40–0.50%) on the MATLAB/Simulink records. These ratios count tied magnitudes only; after cluster assignments and signs are stored, the packed footprint is about 19–22% of the student model. The representation is value-tying rather than sparsity, so dense inference is not automatically faster. Reconstruction reduces the student data residual by nearly two orders of magnitude on degraded mathematical-model data, while remaining comparable on MATLAB/Simulink data except under the strongest combined artifacts. These results demonstrate the framework’s potential to combine robust parameter estimation with compact representations of GCIS dynamics, revealing a compression–accuracy trade-off under severe measurement degradation.