Optimal control of a heat pump-based energy system for space heating and hot water provision in buildings: Results from a field test

Energy and Buildings, Год журнала: 2024, Номер 310, С. 114116 - 114116

Опубликована: Март 27, 2024

Язык: Английский

---

Interconnected Smart Transactive Microgrids—A Survey on Trading, Energy Management Systems, and Optimisation Approaches

Journal of Sensor and Actuator Networks, Год журнала: 2024, Номер 13(2), С. 20 - 20

Опубликована: Март 1, 2024

The deployment of isolated microgrids has witnessed exponential growth globally, especially in the light prevailing challenges faced by many larger power grids. However, these remain separate entities, thus limiting their potential to significantly impact and improve stability, efficiency, reliability broader electrical system. Thus, address this gap, concept interconnected smart transactive (ISTMGs) arisen, facilitating interconnection microgrids, each with its unique attributes aimed at enhancing performance grid Furthermore, ISTMGs are expected create more robust resilient energy networks that enable innovative efficient mechanisms for trading sharing between individual centralized grid. This paradigm shift sparked a surge research developing effective ISTMG mechanisms. paper, we present review current state-of-the-art focus on trading, management systems (EMS), optimization techniques ISTMGs. We discuss various types architectures, platforms, stakeholders involved proceed elucidate suitable applications EMS within such frameworks, emphasizing utility domains. includes an examination tools methodologies deploying Subsequently, conduct analysis constraints, delineate prospects future advance establishment utilization

Язык: Английский

---

A frequency cooperative control strategy for multimicrogrids with EVs based on improved evolutionary-deep reinforcement learning

International Journal of Electrical Power & Energy Systems, Год журнала: 2024, Номер 159, С. 109991 - 109991

Опубликована: Апрель 29, 2024

The construction of microgrid has fully promoted the large-scale access distributed power and rapid development electric vehicles (EVs). And system faces operational issues such as strong random disturbances from sources loads, well unexpected events during operation. These can lead to unstable frequency, excessive discharge EVs, increased control costs. To solve issues, a frequency cooperative strategy for multimicrogrids with EVs based on improved evolutionary-deep reinforcement learning (EDRL) is proposed in this paper. First, consider effect vehicle-to-grid (V2G) process minimum time required charge an EV, impact output distribution micro turbines (MTs) regulation cost, coupling relationship between generator terminal voltage control, multimicrogrid comprehensive model constructed. Second, order deal engineering tasks deceptive rewards sparse integrated evolutionary algorithms deep are combined effectively assist training jump out local optimal solutions approach strategies. Meanwhile, algorithm novelty search intelligent partition strategy, so that better convergence characteristics reduce cost information transmission computational complexity under premise ensuring effect. Furthermore, state space, action space reward function controller defined. Finally, simulation results show ability coordinated adjustment MT unit unnecessary discharging while requirements each submicrogrid, which far superior PID fuzzy traditional control.

Язык: Английский

---

A multilayer voltage intelligent control strategy for distribution networks with V2G and power energy Production-Consumption units

International Journal of Electrical Power & Energy Systems, Год журнала: 2024, Номер 159, С. 110055 - 110055

Опубликована: Июнь 1, 2024

The large-scale application of distributed power generation and the prospects vehicle to grid (V2G) technology lead unstable operating voltage distribution network, but also new possible approaches regulation system. Therefore, a multilayer intelligent control strategy is proposed for network with V2G energy production-consumption units (PECUs). First, model PECU includes facilities such as electric (EV) charging stations (CSs), urban loads, energy. basic unit that can accept scheduling, provide flexible resources system, enhance system capabilities. Second, based on impact process demand EV users, CS frequency modulation controllable boundary evaluation sum response were obtained, which enable CSs participate in regulation. Furthermore, address complex engineering tasks hidden rewards comprehensive control, an evolutionary deep reinforcement learning (EDRL) algorithm applied improved novelty search operations. This further enhances scalability agent design structure network. Finally, simulation results show models fully utilize capabilities while satisfying internal supply balance PECUs. has best coordinated ability, compared particle swarm optimization (PSO), losses are reduced by 23.93 %, deviations significantly lowered 71.95 it ensures demands reducing discharge EVs 6 times.

Язык: Английский

---

Review of Computational Intelligence Approaches for Microgrid Energy Management

IEEE Access, Год журнала: 2024, Номер 12, С. 123294 - 123321

Опубликована: Янв. 1, 2024

This research investigates implementing and optimizing microgrid energy management systems (EMS) utilizing artificial intelligence (AI). Inspired by the need for efficient resource utilization limitations of traditional control methods, it addresses essential aspects design, such as cost-effectiveness, system capacity, power generation mix, customer satisfaction. The primary goals are to optimize management, techniques, AI applications in microgrids. study critically examines classification systems, various EMS applications, their associated challenges. Additionally, discusses different optimization techniques relevant EMS, highlighting benefits, emphasizes importance hybrid demand-side storage addressing intermittency renewable sources. unsupervised learning (USL), supervised (SL), semi-supervised (SSL), extensively analyzed relation specific applications. explores AI-based hierarchical controls at primary, secondary, tertiary levels. Furthermore, methods like deep load forecasting reinforcement optimal emphasized substantial contributions enhancing reliability efficiency. concludes that integrating distributed resources (DER) using advanced algorithms can lead significant financial benefits improved sustainability operations. Over 200 papers were referenced this study.

Язык: Английский

---

Coordinated energy management strategy for multi-energy hub with thermo-electrochemical effect based power-to-ammonia: A multi-agent deep reinforcement learning enabled approach

Renewable Energy, Год журнала: 2023, Номер 214, С. 216 - 232

Опубликована: Июнь 10, 2023

Язык: Английский

---

Multi-objective optimization of turbine blade profiles based on multi-agent reinforcement learning

Energy Conversion and Management, Год журнала: 2023, Номер 297, С. 117637 - 117637

Опубликована: Окт. 11, 2023

Язык: Английский

---

Towards a blockchain and machine learning-based framework for decentralised energy management

Energy and Buildings, Год журнала: 2023, Номер 303, С. 113757 - 113757

Опубликована: Ноя. 20, 2023

In most domestic buildings, gas and electricity are supplied by energy utility companies through centralised systems. This often results in a high burden on central management systems has adverse effects prices. Blockchain-based peer-to-peer trading platforms can deliver strategic operation of decentralised multi-energy network among multiple buildings to reduce global greenhouse emissions address climate change issues. However, prevailing blockchain-based focused system implementation for while lacking predictive control scheduling optimisation. Therefore, this paper presents an integrated blockchain machine learning-based framework forms (i.e., heat electricity) allocation transmission, buildings. Machine learning is harnessed predict day-ahead generation consumption patterns prosumers consumers within the network. The proposed will establish optimal automated users transactions. approach focuses energy-matching from both supply demand sides encouraging direct between consumers. security fairness also be enhanced using smart contracts strictly execute bill payment rules. A case study 4 real-life introduced determine economic technical potential framework. comparison approaches, key benefit improved computational load/failure single point, strategy, workload, capital cost energy. Findings suggest that costs reduced 7.60%-25.41% prosumer fall 5.40%-17.63% consumer practical applications, involve larger number community decentralise trading, thus significantly contributing reduction enhancing environmental sustainability.

Язык: Английский

---

Energy optimization management of microgrid using improved soft actor-critic algorithm

International Journal of Renewable Energy Development, Год журнала: 2024, Номер 13(2), С. 329 - 339

Опубликована: Фев. 20, 2024

To tackle the challenges associated with variability and uncertainty in distributed power generation, as well complexities of solving high-dimensional energy management mathematical models mi-crogrid optimization, a microgrid optimization method is proposed based on an improved soft actor-critic algorithm. In method, algorithm employs entropy-based objective function to encourage target exploration without assigning signifi-cantly higher probabilities any part action space, which can simplify analysis process generation while effectively mitigating convergence fragility issues model management. The effectiveness validated through case study op-timization results revealed increase 51.20%, 52.38%, 13.43%, 16.50%, 58.26%, 36.33% total profits compared Deep Q-network algorithm, state-action-reward-state-action proximal policy ant-colony strategy genetic fuzzy inference system, theoretical retailer stragety, respectively. Additionally, com-pared other methods strategies, learn more optimal behaviors anticipate fluctuations electricity prices demand.

Язык: Английский

---

A Novel MADRL With Spatial-Temporal Pattern Capturing Ability for Robust Decentralized Control of Multiple Microgrids Under Anomalous Measurements

IEEE Transactions on Sustainable Energy, Год журнала: 2024, Номер 15(3), С. 1872 - 1884

Опубликована: Апрель 8, 2024

The anomalous measurements pose significant challenges for the secure and economical operation of multiple microgrids (MMGs). However, existing works still cannot effectively address this problem. Therefore, paper proposes a robust decentralized multi-agent deep reinforcement learning (MADRL) control approach by developing novel actor network on basis centralized training execution framework (CTDEF). To achieve MMG systems against data, extracts variation patterns from both temporal spatial perspectives. From perspective, are first cast to graph, multi-head graph attention (MGAT) is employed extract structural correlations among these measurements. feature extracted MGAT along with state historical actions processed two recurrent networks obtain trajectory history each MG. confederate image technology developed therein agent infer intentions other agents in order better history. more fully express between nodes decision agents, node cognition regularizer mutual information-based regularization term designed optimizing network, respectively. By integrating perspectives, proposed achieves greater robustness outliers than approaches that consider only one perspective. experimental results confirm effectiveness approach.

Язык: Английский

---