Skip to content
Log in

Similar papers

Conference Jul 2026

Reinforcement Learning-Based Building Energy Optimization for Flexibility Markets

The rapid growth of energy markets and demand-side response programs has created a significant need for intelligent building-level control strategies to capture high volatility energy consumption in response to price signals and grid conditions. This paper presents a reinforcement learning (RL)-based building energy management framework that models commercial buildings as active, grid-interactive assets capable of providing real-time flexibility while maintaining occupant comfort. The proposed approach integrates historical and real-time data from IoT sensors, HVAC systems, and weather forecasts to build an adaptive environment for RL agents. The RL model is applied to learn optimal control policies that minimize operational energy cost in response to flexibility markets through load shifting, peak shaving, and short-term demand response actions. The framework also incorporates a forecasting module to predict 30-minute interval energy consumption using deep learning, enabling proactive decision-making under uncertainty. Results from simulation experiments demonstrate that the RL agents achieve significant cost savings compared to rule-based control strategies and offer a reliable, automated control to unlock underlying flexibility within building systems. The paper discusses problem formulation, algorithmic development, simulation workflows, comparative metrics, and practical deployment considerations for Saudi Arabia's smart city initiatives.

Abdulaziz Almalaq · 0 citations
Aug 2026

Collaborative optimization of car-following control and energy management for PHEBs based on deep reinforcement learning

To address the synergistic optimization of car-following control and energy management, this study proposes a collaborative optimization framework based on deep reinforcement learning (DRL). At the car-following control level, a predictive cruise control (PCC) model is developed using the twin delayed deep deterministic policy gradient (TD3) algorithm, which incorporates safety and passenger comfort and power demand into the reward function. At the energy management level, a TD3-based energy management strategy (EMS) is formulated, incorporating constraints on battery state of health (SOH) degradation, temperature violations, state of charge (SOC) fluctuations, and comprehensive operating costs. Simulation results demonstrate that, compared to the traditional hierarchical optimization framework, the proposed strategy achieves significant improvements in terms of mean absolute jerk, root-mean-square (RMS) value of acceleration, power demand, battery SOH degradation, battery temperature violation, and comprehensive operating cost, with optimization rates of 32.7%, 49.55%, 5.39%, 28.30%, 91.76%, and 22.82%, respectively. Furthermore, generalization validation indicates that the proposed framework maintains strong robustness and adaptability under unknown driving conditions.

Chengrui Zhang, Fei Ju, Sichen Gao et al. · 0 citations
Open access Jul 2026

AI-driven optimization: revolutionizing energy efficiency in modern buildings.

Rising global energy demand and increasing decarbonization requirements have intensified the need for intelligent building energy management capable of handling nonlinear dynamics and multi-objective operational trade-offs. Conventional discrete-time and simulation-dependent control strategies often struggle to maintain temporal continuity, adaptive responsiveness, and consistent performance across heterogeneous building environments. Addressing these limitations, NODE-RL-BEM (Neural Ordinary Differential Equation Reinforcement Learning for Building Energy Management) introduces a unified continuous-time optimization paradigm that jointly models system dynamics and learns adaptive control policies. The approach integrates heterogeneous operational data, temporal state embeddings, neural differential equation modeling, and multi-objective reinforcement learning within a cohesive architecture designed for predictive and responsive energy optimization. Performance evaluation conducted on the ASHRAE Great Energy Predictor III dataset and the Intelligent Indoor Environment Dataset demonstrates the effectiveness of the proposed framework, achieving 42-48% energy savings, maintaining comfort violations below 0.5%, and improving indoor air quality by 28-35%. The framework further achieves a generalization score of 0.91 across diverse building operational scenarios, confirming strong transferability and stability. Continuous-time dynamics learning improves predictive fidelity and ensures smooth state evolution, while adaptive reinforcement learning enables robust decision-making under dynamic environmental and occupancy variations. Scalable applicability to multi-zone building environments highlights practical deployment feasibility. This work establishes a novel continuous-time dynamic-policy learning paradigm that integrates predictive modeling with real-time adaptive control, advancing data-driven intelligent building operation toward sustainable and autonomous energy management.

Hamoud H. Alshammari · 0 citations
Open access Aug 2026

AI-Based Digital Twins for Renewable Energy Grid Optimization: A Framework for Real-Time Forecasting, State Estimation and Adaptive Dispatch

Growing reliance on distributed, weather-dependent renewable sources brings sizable variability, uncertainty, and stability problems to today's power grids. This work proposes an AI-driven digital twin (DT) framework that pairs a real-time physical-data grid model with a layered AI optimisation engine made up of an LSTM/Transformer forecaster, a graph neural network (GNN) state estimator, and a deep reinforcement learning (DRL) dispatch controller. The twin continually mirrors the physical grid, supporting scenario simulation, predictive fault detection, and closed-loop control recommendations before any action is applied to the real system. A case study on a modified IEEE 33-bus feeder with solar PV and wind generation shows the framework raises renewable utilisation by 19 points, cuts curtailment by 28 points, and lowers fault response time by 68.2% versus a conventional energy-management system. These results suggest AI-enhanced digital twins are a scalable route to self-optimising, resilient renewable-heavy grids.

Dr. G. Sripriya, V. S. Guhan, A. S. Nandha Kisore · 0 citations
Jul 2026

A deep reinforcement learning framework for hybrid electric vehicle energy management: Integrating real-world data augmentation and multi-scale perception

To solve the generalization bottleneck and environmental perception limitation of deep reinforcement learning (DRL) in the charge-sustaining (CS) stage of hybrid electric vehicles, this paper proposes a novel adaptive hierarchical energy management strategy, which combines real-world data enhancement and multi-scale perception. Aiming at the problems of over-fitting and short-sighted decision-making commonly existing in traditional strategies, this study constructed a fresh enhanced training set covering all-round driving cycles based on real driving data, thus breaking through the training restrictions brought by standard driving cycles. In addition, the historical average speed window is introduced as the enhanced state, which enables the strategy to capture the macro traffic flow trend. In terms of control architecture, a bi-level coupled mechanism based on soft actor-critic (SAC) and equivalent consumption minimization strategy (ECMS) is designed. Experimental results indicate that across multi-type driving cycle tests, the final SOC deviation of the agent based on the augmented training set is maintained within ±2%. Under identical operating conditions, the augmented agent incorporating a 30 s average velocity observation achieves a 43.55% reduction in the standard deviation of SOC fluctuations compared to its counterpart without such observation. While ensuring battery SOC robustness, the proposed strategy improves fuel economy by 5.2% on average across various driving cycles compared to adaptive ECMS (AECMS).

Yushan Li, Lianbo Zhao, Fanyu Meng et al. · 0 citations
Open access Aug 2026

Deep Learning-Supported Hybrid Renewable Energy System Optimization

Energy system optimization seeks to utilize multiple energy sources efficiently under technical, economic, and environmental constraints. The increasing integration of renewable energy and the need for sustainable operation have made the optimal planning and management of hybrid energy systems crucial. Classical optimization methods, including Linear Programming, Nonlinear Programming, and simulation-based models, often face limitations when addressing high-dimensional and nonlinear problems. This study introduces a deep learning–based surrogate modeling framework for sizing the components of hybrid renewable energy systems. Initially, Particle Swarm Optimization (PSO) is employed to determine the optimal component sizes for a large number of synthetically generated hourly solar irradiance and load profiles. These optimal solutions are then used as target labels. The associated annual time-series data are transformed into multi-channel Data Map (DMAP) images, which serve as inputs for convolutional neural networks (CNNs). After training, the CNN models are capable of directly estimating the required number of photovoltaic (PV) panels, inverter capacity, and battery units from the DMAP images, eliminating the need to perform the iterative PSO optimization during the prediction stage. Various convolutional neural network architectures, including ResNet, DenseNet121, RegNet, ConvNeXt, EfficientNet, SqueezeNet, MobileNet, and InceptionV3, were evaluated for this multi-output regression task. The results indicate that ResNet and DenseNet121 achieve the best performance, while ConvNeXt provides strong results with a modern architectural design. Among the evaluated models, DenseNet121 achieved coefficients of determination (R2) of 0.934, 0.988, and 0.947 for predicting the sizes of the PV array, inverter, and battery bank, respectively. These results correspond to an average prediction accuracy of approximately 90.6%. ResNet produced similar performance, with its highest R2 value reaching 0.983 for inverter sizing. Lightweight networks such as SqueezeNet and MobileNet demonstrate notable effectiveness for resource-constrained systems, whereas InceptionV3 underperforms in leveraging its multi-scale architecture. These results demonstrate that, once the models have been trained, deep learning–based surrogate models can generate sizing decisions comparable to those obtained using PSO with only a fraction of the computational effort. As a result, they provide a fast and practical alternative to conventional iterative optimization methods for component sizing in smart grid and sustainable energy planning applications.

Yasemin Alakoç Bozkurt, Cemil Altın, Talip Çay · 0 citations