Special Issue on “Modeling, Simulation and Control of Energy Systems”

1. Introduction

Clean energy systems are at the forefront of current discussions on the protection of our environment against global warming. Decarbonization strategies along with novel energy systems have diverted research groups to the fields of renewables, alternative fuels, energy efficiency improvement, autonomous systems design, waste management and energy policy regulations. To this end, frameworks for the modeling, simulation and control of energy systems have become essential tools for identifying the complex dynamics of these strategies and predicting system performance under stochastic disturbances and system variations (either short-term or long-term). Such energy systems may range from TRL (technology readiness level) 1 to 9, and crucial efforts have been devoted to their efficient scale-up and commercialization. Additional challenges are the advanced modeling and process control techniques that have advanced from pilot- to industrial-scale applications. This Special Issue entitled “Modeling, Simulation and Control in Energy Systems” collected high-quality studies that address challenges facing the areas of energy production, management, utilization and storage. Three core topics were identified: (i) power system optimization, grid management and renewable integration; (ii) intelligent control, prediction and data-driven modeling in energy systems; and (iii) mechanical, thermal and fluid system modeling and engineering design. Overall, nineteen (19) studies were published and are available online at: https://www.mdpi.com/journal/processes/special_issues/HI863SV90Y. The list of contributions refers to articles numbered 1–19.

2. An Overview of the Published Papers

2.1. Power System Optimization, Grid Management and Renewable Integration

Articles related to this topic focused on the improvement in performance, economics, reliability, and sustainability of modern power systems through optimization, forecasting, coordination and intelligent control. The methods that were analyzed involved (among others) intelligent optimization algorithms (PSO, ALO, GA), economic modeling, risk assessment, multi-energy coupling, cloud–edge computing and renewable integration.

Serrano et al. (List of Contributions 1) proposed a hybrid optimization method using the Greedy Randomized Adaptive Search Procedure (GRASP) and Tabu Search to improve capacity management in smart grids through the optimal siting of distributed generators and coordinated load shedding. Applied on a real 1806-bus network, this approach enhances voltage levels, reduces losses and strengthens reliability by enabling islanded operation and more efficient load restoration. The results showed that combining DG allocation with demand-side management outperformed strategies using only one technique, providing a promising tool for modern grid planning and performance improvement. Wang et al. (List of Contributions 2) discussed a pricing strategy tailored for electricity prosumers who both generate and consume energy. Users were clustered by demand preference, price sensitivity and risk tolerance, enabling customized packages. Two pricing schemes were proposed: fluctuating time-of-use pricing and discount-based purchase–sale packages. A hybrid Self-Adaptive Weight and Reverse Learning Particle Swarm Optimization (SAW and RL-PSO) algorithm was used to determine optimal pricing while accounting for market uncertainty via CvaR (Conditional Value at Risk) risk modeling. The results showed reduced prosumer costs, improved load balancing and increased retailer revenue, improving renewable energy consumption and market stability. In a study by Li et al. (List of Contributions 3), a robust economic dispatch strategy that integrates virtual power plants (VPPs) to manage renewable-energy uncertainty in modern grids was developed. Renewable output variability was modeled using interval-based prediction errors, and a mixed-integer nonlinear VPP optimization problem was solved using an improved Ant Lion Optimizer. Testing on an IEEE-30 bus system showed considerably reduced computation time. Chen et al. (List of Contributions 4) developed a linear two-port model of pipeline heat transfer (derived from an implicit upwind difference scheme) to compress partial differential equation (PDE) dynamics into offline coefficient matrices. Embedded in an electro-thermal coupling system (EDETCS), the approach preserves accuracy while drastically speeding computation. Case studies showed lower operating costs and reduced wind curtailment, improving system flexibility and economic performance. In List of Contributions 5, an adaptive cloud–edge scheduling method that mines edge-side computing resources to relieve cloud master station load for power distribution security tasks is proposed. The authors modeled multi-stage relay security protection, formulated a Karush–Kuhn–Tucker (KKT)-solved linear program to minimize the maximum expected cloud pressure and validated the approach with software simulations. The results showed up to ~35% reductions in peak cloud master pressure and improved resource utilization across varied network scenarios. Zhang et al. (List of Contributions 6) proposed a dual-sequence Monte Carlo method that integrates electrical–thermal cable modeling, emergency-rating behavior and insulation aging to assess underground cable reliability. By introducing design-risk and aging-risk coefficients, the framework quantifies how long- and short-term emergency loading affects failure probability and network performance. Applied to an enhanced IEEE 14-bus network system, the method showed that emergency ratings cut expected unserved energy by ~59% but also accelerated aging on critical cables, providing a balanced tool for risk-aware planning and asset management. Zhang et al. (List of Contributions 7) developed a multi-objective dispatch strategy for park-level integrated energy systems by tightly coupling P2G, CCS and CHP units to enhance flexibility and reduce carbon emissions under uncertain wind and solar output. A reward–penalty ladder carbon-trading mechanism was introduced, and renewable scenarios were generated via kernel density estimation and Frank Copula modeling. Using an NSDBO solver, the simulations showed significant gains in renewable utilization, carbon emissions and total costs decreases. Finally, in (List of Contributions 8), the authors developed an optimization model for selecting grid loss-reduction strategies under China’s transmission and distribution pricing reforms. The model minimizes life-cycle costs by combining investment expenses, direct power-loss costs and carbon-emission penalties, while enforcing voltage, branch-flow, and budget constraints. Using an adaptive genetic algorithm, the method efficiently found optimal strategy combinations among many possibilities.

2.2. Intelligent Control, Prediction and Data-Driven Modeling in Energy Systems

In this field, relevant articles were identified using advanced control techniques, machine learning, hybrid ML–numerical models, or deep learning to predict system behavior, support energy management, or improve decision-making. All articles shared a focus on modeling accuracy, feature selection, AI-based forecasting and real-time control decision-making.

Kaiser et al. (List of Contributions 9) conducted a review that examined the limitations of conventional CFD modeling for PEM fuel cells and highlighted how machine learning can enhance parameter estimation, mass-transfer prediction and electrochemical modeling. The authors analyzed the current challenges in water and thermal management, durability and material characterization, showing how hybrid machine learning (ML) CFD approaches can improve reliability and reduce computational costs. In another interesting study (List of Contributions 10), an indirect primary-side control strategy for wireless power transfer electrical vehicle (WPT-EV) chargers was constructed. This system holds battery charging current constant despite wireless-link latency and coil misalignment. A PI controller was paired with an adaptive hill-climbing algorithm that updated an efficiency-based mapping only when new battery measurements arrive. The simulation and laboratory tests showed tight current tracking, smooth phase-shift actuation and robust performance efficiency. Moreover, in (List of Contributions 11), the authors developed a robust hybrid method combining improved Complementary Ensemble Empirical Mode Decomposition for denoising/decomposition, Hilbert Transform for instantaneous features and a PSO algorithm for classification. Validated on real smart-meter and fault-recorder data, the approach achieved >95% identification accuracy under heavy noise. The method is computationally efficient and well-suited for real-time grid monitoring and multi-disturbance scenarios. Tong et al. (List of Contributions 12) proposed a hybrid framework for accurate transformer top-oil temperature prediction by integrating Variational Mode Decomposition (VMD) decomposition, Kernel principal component analysis, and a time-aware Shapley Additive Explanations–Multilayer Perceptron (SHAP–MLP) method for dynamic feature selection. A Time-Series Forecasting System (SOFTS) deep-learning model, optimized with hierarchical Bayesian search, captures multiscale dependencies and improves robustness under fluctuating operating conditions. Finally, Huang et al. (List of Contributions 13) developed a convolutional neural network–bidirectional gated recurrent unit hybrid to predict NOx concentration and boiler thermal efficiency from 3000 operational samples of a 300 MW boiler. It couples IFDA (an improved flow-direction algorithm) for single-objective tuning and NSGA-II for multi-objective optimization of airflows and O₂ to reduce NOx while raising efficiency. The results show the CNN-IFDA-BiGRU model substantially lowers prediction error, and the non-dominated sorting genetic algorithm (NSGA-II) yields ~5.01% NOx reduction and a 0.32% efficiency gain on average.

2.3. Mechanical, Thermal and Fluid System Modeling and Engineering Design

Ending this editorial review, the studies included this last topic involved the simulation, analysis and optimization of mechanical, thermal, combustion and multiphase fluid systems. Other areas studied included hardware implementation, control dynamics, thermal/fluid behavior and component-level performance analysis.

Provatas and Ipsakis (List of Contributions 14) present an educational yet practical framework for designing and simulating feedback controllers for an active vehicle suspension system. Using a quarter-car model, the authors compared P, PI, PID, genetic-algorithm-optimized PID and internal model controllers (IMCs) across disturbance and tracking scenarios. The results show that the optimal PID and IMC deliver superior comfort and stability, with the optimized PID offering best set-point tracking while IMCs demand less actuator effort. The work stands out for its step-by-step methodology, making advanced control concepts accessible for engineering students. In (List of Contributions 15), the authors analyzed how different blade tip clearances affect internal flow behavior, gas distribution and efficiency in gas–liquid multiphase pumps. Using Eulerian two-fluid CFD modeling with four clearance values (0–0.9 mm), the authors found that larger clearances increased tip leakage flow almost linearly, causing increased energy loss and a noticeable drop in head coefficient. Whereas small clearances led to uneven flow and localized gas accumulation, wider gaps improved uniformity but also reduced efficiency due to leakage. The findings guide optimal clearance selection in high-performance multiphase pump design. Korukcu (List of Contributions 16) introduced the Ozan Combustion Calculator (OCC), a MATLAB-based GUI designed to compute combustion parameters for 16 fuels among common alkanes and alcohols. The tool calculates adiabatic flame temperature, energy release and exergy destruction for both dry and moist air while visually displaying combustion equations—an improvement over earlier software. Validation against methods in the literature showed high accuracy. The study also reveals and offers valuable insights for combustion modeling and education. In (List of Contributions 17), a ferrofluid-based sealing system was used to increase the reliability of pressure-vessel safety valves. A composite structure was developed combining mechanical end-face sealing with a magnetic-fluid seal driven by an electromagnet. Using numerical modeling and through validation with experiments, key geometric parameters—seal gap, groove width, tooth height, tooth width and number of sealing stages—were optimized. The results showed that the sealing pressure first increased then declined with tooth height, groove width and stage number. In another interesting and applied research study (List of Contributions 18), PID, H∞, model predictive control (MPC) and multiple input–multiple output Model Reference Adaptive Control (MRAC) control strategies were compared in offshore-produced water treatment using pilot-scale plant data. Focusing on the multivariable control of separator water level and hydrocyclone pressure drop ratio, the study shows that MRAC provides superior tracking accuracy and adaptability, whereas MPC achieves the highest de-oiling efficiency with reduced control effort. The results highlight clear trade-offs among robustness, efficiency, complexity and industrial feasibility, offering practical guidance for selecting control architectures in offshore water treatment systems. Finally, the authors of (List of Contributions 19) developed a Space Vector Pulse-Width Modulation (SVPWM)-based variable-frequency inverter (31–300 Hz) implemented on a low-cost 8-bit microcontroller. By varying the number of switching periods per electrical cycle (930–1838) and linearizing computations, the design achieves precise timing with minimal computational load. Experimental and MATLAB/Simulink evaluations show the performance comparable to that of other implementations, demonstrating industrial-level capability using inexpensive hardware.

3. Conclusions

In conclusion, the papers collected in this Special Issue highlight the growing role of advanced modeling, simulation and intelligent control techniques in addressing the technical, economic and environmental challenges facing modern energy and engineering systems. From power system optimization and renewable integration to data-driven prediction, adaptive control and detailed multi-physics modeling, these contributions demonstrate how interdisciplinary approaches can enhance system efficiency, reliability and sustainability. Together, they provide valuable insights for researchers and practitioners and point toward future directions where digitalization, artificial intelligence and integrated energy solutions will play an increasingly central role.

• Intelligent Control, Prediction and Data-Driven Modeling in Energy Systems

Other notable articles on this topic were not included in this Special Issue [1,2,3,4,5,6,7,8,9,10].

• Power System Optimization, Grid Management and Renewable Integration

Other notable articles on this topic were not included in this Special Issue [11,12,13,14,15,16,17].

• Mechanical, Thermal and Fluid System Modeling and Engineering Design

Other notable articles on this topic and not included in this Special Issue include [18,19,20,21,22].