Reinforcement Learning-Based Energy Management for Hybrid Power Systems:State-of-the-Art Survey,Review,and Perspectives

The new energy vehicle plays a crucial role in green transportation,and the energy management strategy of hybrid power systems is essential for ensuring energy-efficient driving.This paper presents a state-of-the-art survey and review of reinforcement learning-based energy management strategies for hybrid power systems.Additionally,it envisions the outlook for autonomous intelligent hybrid electric vehicles,with reinforcement learning as the foundational technology.First of all,to provide a macro view of historical development,the brief history of deep learning,reinforcement learning,and deep reinforcement learning is presented in the form of a timeline.Then,the comprehensive survey and review are conducted by collecting papers from mainstream academic databases.Enumerating most of the contributions based on three main directions—algorithm innovation,powertrain innovation,and environment innovation—provides an objective review of the research status.Finally,to advance the application of reinforcement learning in autonomous intelligent hybrid electric vehicles,future research plans positioned as“Alpha HEV”are envisioned,integrating Autopilot and energy-saving control.

Role of outdoor trees on pedestrian wind and thermal conditions around a pre-education building for sustainable energy management

Finding sustainable energy resources is essential to face the increasing energy demand.Trees are an important part of ancient architecture but are becoming rare in urban areas.Trees can control and tune the pedestrian-level wind velocity and thermal condition.In this study,a numerical investigation is employed to assess the role of trees planted in the windward direction of the building complex on the thermal and pedestrian wind velocity conditions around/inside a pre-education building located in the center of the complex.Compared to the previous studies(which considered only outside buildings),this work considers the effects of trees on microclimate change both inside/outside buildings.Effects of different parameters including the leaf area density and number of trees,number of rows,far-field velocity magnitude,and thermal condition around the main building are assessed.The results show that the flow velocity in the spacing between the first-row buildings is reduced by 30%-40% when the one-row trees with 2 m height are planted 15 m farther than the buildings.Furthermore,two rows of trees are more effective in higher velocities and reduce the maximum velocity by about 50%.The investigation shows that trees also could reduce the temperature by about 1℃around the building.

Regenerative Braking Energy Recovery System of Metro Train Based on Dual-Mode Power Management

In order to fully utilize the regenerative braking energy of metro trains and stabilize the metro DC traction busbar voltage,a hybrid regenerative braking energy recovery system with a dual-mode power management strategy is proposed.Firstly,the construction of the hybrid regenerative braking energy recovery system is explained.Then,based on the power demand of low-voltage load in metro stations,a dual-mode power management strategy is proposed to allocate the reference power of each system according to the different working conditions,and the control methods of each system are set.Finally,the correctness and effectiveness of the dual-mode strategy are verified through simulation,and the proposed braking energy utilization scheme is compared with other singleform utilization schemes.The results illustrate that the hybrid system with the dual-mode strategy can effectively recycle the regenerative braking energy of metro train and inhibit the busbar voltage fluctuation;the proposed braking energy utilization scheme has certain advantages on energy recovery and DC bus voltage stabilization compared with other single-form schemes;the proposed power management strategy can correctly allocate the reference power of each system with a lower construction cost.

Smart Energy Management System Using Machine Learning

Energy management is an inspiring domain in developing of renewable energy sources.However,the growth of decentralized energy production is revealing an increased complexity for power grid managers,inferring more quality and reliability to regulate electricity flows and less imbalance between electricity production and demand.The major objective of an energy management system is to achieve optimum energy procurement and utilization throughout the organization,minimize energy costs without affecting production,and minimize environmental effects.Modern energy management is an essential and complex subject because of the excessive consumption in residential buildings,which necessitates energy optimization and increased user comfort.To address the issue of energy management,many researchers have developed various frameworks;while the objective of each framework was to sustain a balance between user comfort and energy consumption,this problem hasn’t been fully solved because of how difficult it is to solve it.An inclusive and Intelligent Energy Management System(IEMS)aims to provide overall energy efficiency regarding increased power generation,increase flexibility,increase renewable generation systems,improve energy consumption,reduce carbon dioxide emissions,improve stability,and reduce energy costs.Machine Learning(ML)is an emerging approach that may be beneficial to predict energy efficiency in a better way with the assistance of the Internet of Energy(IoE)network.The IoE network is playing a vital role in the energy sector for collecting effective data and usage,resulting in smart resource management.In this research work,an IEMS is proposed for Smart Cities(SC)using the ML technique to better resolve the energy management problem.The proposed system minimized the energy consumption with its intelligent nature and provided better outcomes than the previous approaches in terms of 92.11% accuracy,and 7.89% miss-rate.

Reinforcement Learning Model for Energy System Management to Ensure Energy Efficiency and Comfort in Buildings

This article focuses on the challenges ofmodeling energy supply systems for buildings,encompassing both methods and tools for simulating thermal regimes and engineering systems within buildings.Enhancing the comfort of living or working in buildings often necessitates increased consumption of energy and material,such as for thermal upgrades,which consequently incurs additional economic costs.It is crucial to acknowledge that such improvements do not always lead to a decrease in total pollutant emissions,considering emissions across all stages of production and usage of energy and materials aimed at boosting energy efficiency and comfort in buildings.In addition,it explores the methods and mechanisms for modeling the operating modes of electric boilers used to collectively improve energy efficiency and indoor climatic conditions.Using the developed mathematical models,the study examines the dynamic states of building energy supply systems and provides recommendations for improving their efficiency.These dynamic models are executed in software environments such as MATLAB/Simscape and Python,where the component detailing schemes for various types of controllers are demonstrated.Additionally,controllers based on reinforcement learning(RL)displayed more adaptive load level management.These RL-based controllers can lower instantaneous power usage by up to 35%,reduce absolute deviations from a comfortable temperature nearly by half,and cut down energy consumption by approximately 1%while maintaining comfort.When the energy source produces a constant energy amount,the RL-based heat controllermore effectively maintains the temperature within the set range,preventing overheating.In conclusion,the introduced energydynamic building model and its software implementation offer a versatile tool for researchers,enabling the simulation of various energy supply systems to achieve optimal energy efficiency and indoor climate control in buildings.

Enhanced Deep Reinforcement Learning Strategy for Energy Management in Plug-in Hybrid Electric Vehicles with Entropy Regularization and Prioritized Experience Replay

Plug-in Hybrid Electric Vehicles(PHEVs)represent an innovative breed of transportation,harnessing diverse power sources for enhanced performance.Energy management strategies(EMSs)that coordinate and control different energy sources is a critical component of PHEV control technology,directly impacting overall vehicle performance.This study proposes an improved deep reinforcement learning(DRL)-based EMSthat optimizes realtime energy allocation and coordinates the operation of multiple power sources.Conventional DRL algorithms struggle to effectively explore all possible state-action combinations within high-dimensional state and action spaces.They often fail to strike an optimal balance between exploration and exploitation,and their assumption of a static environment limits their ability to adapt to changing conditions.Moreover,these algorithms suffer from low sample efficiency.Collectively,these factors contribute to convergence difficulties,low learning efficiency,and instability.To address these challenges,the Deep Deterministic Policy Gradient(DDPG)algorithm is enhanced using entropy regularization and a summation tree-based Prioritized Experience Replay(PER)method,aiming to improve exploration performance and learning efficiency from experience samples.Additionally,the correspondingMarkovDecision Process(MDP)is established.Finally,an EMSbased on the improvedDRLmodel is presented.Comparative simulation experiments are conducted against rule-based,optimization-based,andDRL-based EMSs.The proposed strategy exhibitsminimal deviation fromthe optimal solution obtained by the dynamic programming(DP)strategy that requires global information.In the typical driving scenarios based onWorld Light Vehicle Test Cycle(WLTC)and New European Driving Cycle(NEDC),the proposed method achieved a fuel consumption of 2698.65 g and an Equivalent Fuel Consumption(EFC)of 2696.77 g.Compared to the DP strategy baseline,the proposed method improved the fuel efficiency variances(FEV)by 18.13%,15.1%,and 8.37%over the Deep QNetwork(DQN),Double DRL(DDRL),and original DDPG methods,respectively.The observational outcomes demonstrate that the proposed EMS based on improved DRL framework possesses good real-time performance,stability,and reliability,effectively optimizing vehicle economy and fuel consumption.

A Predictive Energy Management Strategies for Mining Dump Trucks

The plug-in hybrid vehicles(PHEV)technology can effectively address the issues of poor dynamics and higher energy consumption commonly found in traditional mining dump trucks.Meanwhile,plug-in hybrid electric trucks can achieve excellent fuel economy through efficient energy management strategies(EMS).Therefore,a series hybrid system is constructed based on a 100-ton mining dump truck in this paper.And inspired by the dynamic programming(DP)algorithm,a predictive equivalent consumption minimization strategy(P-ECMS)based on the DP optimization result is proposed.Based on the optimal control manifold and the SOC reference trajectory obtained by the DP algorithm,the P-ECMS strategy performs real-time stage parameter optimization to obtain the optimal equivalent factor(EF).Finally,applying the equivalent consumption minimization strategy(ECMS)realizes real-time control.The simulation results show that the equivalent fuel consumption of the P-ECMS strategy under the experimentally collected mining cycle conditions is 150.8 L/100 km,which is 10.9%less than that of the common CDCS strategy(169.3 L/100 km),and achieves 99.47%of the fuel saving effect of the DP strategy(150 L/100 km).

Research on Smart Energy Monitoring and Management System Based on Digital Twin Technology

Smart energy monitoring and management system lays a foundation for the application and development of smart energy. However, in recent years, the work efficiency of smart energy development enterprises has generally been low, and there is an urgent need to improve the application efficiency, resilience and sustainability of smart energy monitoring and management system. Digital twin technology provides a data-centric solution to improve smart energy monitoring and management system, bringing an opportunity to transform passive infrastructure assets into data-centric systems. This paper expounds on the concept and key technologies of digital twin, and designs a smart energy monitoring and management system based on digital twin technology, which has dual significance for promoting the development of smart energy field and promoting the application of digital twin.

A utility and easily fabricated dual-mode fiber film for efficient and comfortable thermal management

Nowadays, the global climate is constantly being destroyed and the fluctuations in ambient temperature are becoming more frequent. However, conventional single-mode thermal management strategies(heating or cooling) failed to resolve such dynamic temperature changes. Moreover, developing thermal management devices capable of accommodating these temperature variations while remaining simple to fabricate and durable has remained a formidable obstacle. To address these bottlenecks, we design and successfully fabricate a novel dual-mode hierarchical(DMH) composite film featuring a micronanofiber network structure, achieved through a straightforward two-step continuous electrospinning process. In cooling mode, it presents a high solar reflectivity of up to 97.7% and an excellent atmospheric transparent window(ATW) infrared emissivity of up to 98.9%. Noted that this DMH film could realize a cooling of 8.1 ℃ compared to the ambient temperature outdoors. In heating mode, it also exhibits a high solar absorptivity of 94.7% and heats up to 11.9 ℃ higher than black cotton fabric when utilized by individuals. In practical application scenarios, a seamless transition between efficient cooling and heating is achieved by simply flipping the film. More importantly, the DMH film combining the benefits of composites demonstrates portability, durability, and easy-cleaning, promising to achieve large-scale production and use of thermally managed textiles in the future. The energy savings offered by film applications provide a viable solution for the early realization of carbon neutrality.

Survey on AI and Machine Learning Techniques for Microgrid Energy Management Systems

In the era of an energy revolution,grid decentralization has emerged as a viable solution to meet the increasing global energy demand by incorporating renewables at the distributed level.Microgrids are considered a driving component for accelerating grid decentralization.To optimally utilize the available resources and address potential challenges,there is a need to have an intelligent and reliable energy management system(EMS)for the microgrid.The artificial intelligence field has the potential to address the problems in EMS and can provide resilient,efficient,reliable,and scalable solutions.This paper presents an overview of existing conventional and AI-based techniques for energy management systems in microgrids.We analyze EMS methods for centralized,decentralized,and distributed microgrids separately.Then,we summarize machine learning techniques such as ANNs,federated learning,LSTMs,RNNs,and reinforcement learning for EMS objectives such as economic dispatch,optimal power flow,and scheduling.With the incorporation of AI,microgrids can achieve greater performance efficiency and more reliability for managing a large number of energy resources.However,challenges such as data privacy,security,scalability,explainability,etc.,need to be addressed.To conclude,the authors state the possible future research directions to explore AI-based EMS's potential in real-world applications.

Distributionally robust optimization based chance-constrained energy management for hybrid energy powered cellular networks

Energy harvesting has been recognized as a promising technique with which to effectively reduce carbon emis-sions and electricity expenses of base stations.However,renewable energy is inherently stochastic and inter-mittent,imposing formidable challenges on reliably satisfying users'time-varying wireless traffic demands.In addition,the probability distribution of the renewable energy or users’wireless traffic demand is not always fully known in practice.In this paper,we minimize the total energy cost of a hybrid-energy-powered cellular network by jointly optimizing the energy sharing among base stations,the battery charging and discharging rates,and the energy purchased from the grid under the constraint of a limited battery size at each base station.In solving the formulated non-convex chance-constrained stochastic optimization problem,a new ambiguity set is built to characterize the uncertainties in the renewable energy and wireless traffic demands according to interval sets of the mean and covariance.Using this ambiguity set,the original optimization problem is transformed into a more tractable second-order cone programming problem by exploiting the distributionally robust optimization approach.Furthermore,a low-complexity distributionally robust chance-constrained energy management algo-rithm,which requires only interval sets of the mean and covariance of stochastic parameters,is proposed.The results of extensive simulation are presented to demonstrate that the proposed algorithm outperforms existing methods in terms of the computational complexity,energy cost,and reliability.

Efficient thermal management and all-season energy harvesting using adaptive radiative cooling and a thermoelectric power generator

Passive daytime radiative cooling(PDRC) is useful for thermal management because it allows an object to emit terrestrial heat into space without the use of additional energy.To produce sub-ambient temperatures under direct sunlight,PDRC materials are designed to reduce their absorption of solar energy and to enhance their long-wavelength infrared(LWIR) emissivity.In recent years,many photonic structures and polymer composites have been studied to improve the cooling system of buildings.However,in cold weather(i.e. during winter in cold climates),buildings need to be kept warm rather than cooled due to heat loss.To overcome this limitation,temperature-responsive radiative cooling is a promising alternative.In the present study,adaptive radiative cooling(ARC) film fabricated from a polydimethylsiloxane/hollow SiO_(2) microsphere/thermochromic pigment composite was investigated.We found that the ARC film absorbed solar radiation under cold conditions while exhibiting radiative cooling at ambient temperatures above 40℃.Thus,in outdoor experiments,the ARC film achieved sub-ambient temperatures and had a theoretical cooling power of 63.2 W/m~2 in hot weather.We also demonstrated that radiative cooling with an energy harvesting system could be used to improve the energy management of buildings,with the thermoelectric module continuously generating output power using the ARC film.Therefore,we believe that our proposed ARC film can be employed for efficient thermal management of buildings and all-season energy harvesting in the near future.

Reinforcement Learning-Based Electric Vehicles Energy Management Strategy with Battery Thermal Model

The promotion of electric vehicles(EVs)is restricted due to their short cruising range.It is desirable to design an effective energy management strategy to improve their energy efficiency.Most existing work concerning energy management strategies focused on hybrids rather than the EVs.The work focusing on the energy management strategy for EVs mainly uses the traditional optimization strategies,thereby limiting the advantages of energy economy.To this end,a novel energy management strategy that considered the impact of battery thermal effects was proposed with the help of reinforcement learning.The main idea was to first analyze the energy flow path of EVs,further formulize the energy management as an optimization problem,and finally propose an online strategy based on reinforcement learning to obtain the optimal strategy.Additionally,extensive simulation results have demonstrated that our strategy reduces energy consumption by at least 27.4%compared to the existing methods.

Energy Management and Capacity Optimization of Photovoltaic, Energy Storage System, Flexible Building Power System Considering Combined Benefit

Building structures themselves are one of the key areas of urban energy consumption,therefore,are a major source of greenhouse gas emissions.With this understood,the carbon trading market is gradually expanding to the building sector to control greenhouse gas emissions.Hence,to balance the interests of the environment and the building users,this paper proposes an optimal operation scheme for the photovoltaic,energy storage system,and flexible building power system(PEFB),considering the combined benefit of building.Based on the model of conventional photovoltaic(PV)and energy storage system(ESS),the mathematical optimization model of the system is proposed by taking the combined benefit of the building to the economy,society,and environment as the optimization objective,taking the near-zero energy consumption and carbon emission limitation of the building as the main constraints.The optimized operation strategy in this paper can give optimal results by making a trade-off between the users’costs and the combined benefits of the building.The efficiency and effectiveness of the proposed methods are verified by simulated experiments.

Energy Management System with Power Offering Strategy for a Microgrid Integrated VPP

In the context of both the Virtual Power Plant (VPP) and microgrid(MG), the Energy Management System (EMS) is a key decision-maker forintegrating Distributed renewable Energy Resources (DERs) efficiently. TheEMS is regarded as a strong enabler of providing the optimized schedulingcontrol in operation and management of usage of disperse DERs and RenewableEnergy reSources (RES) such as a small-size wind-turbine (WT) andphotovoltaic (PV) energies. The main objective to be pursued by the EMSis the minimization of the overall operating cost of the MG integrated VPPnetwork. However, the minimization of the power peaks is a new objective andopen issue to a well-functional EMS, along with the maximization of profitin the energy market. Thus, both objectives have to be taken into accountat the same time. Thus, this paper proposes the EMS application incorporatingpower offering strategy applying a nature-inspired algorithm such asParticle Swarm Optimization (PSO) algorithm, in order to find the optimalsolution of the objective function in the context of the overall operating cost,the coordination of DERs, and the energy losses in a MG integrated VPPnetwork. For a fair DERs coordination with minimized power fluctuationsin the power flow, the power offering strategies with an active power controland re-distribution are proposed. Simulation results show that the proposedMG integrated VPP model with PSO-based EMS employing EgalitarianreDistribution (ED) power offering strategy is most feasible option for theoverall operating cost of VPP revenue. The total operating cost of the proposedEMS with ED strategy is 40.98$ compared to 432.8$ of MGs only withoutEMS. It is concluded that each MGs in the proposed VPP model intelligentlyparticipates in energy trading market compliant with the objective function,to minimize the overall cost and the power fluctuation.

Base Station Energy Management in 5G Networks Using Wide Range Control Optimization

The traffic activity offifth generation(5G)networks demand for new energy management techniques that is dynamic deep and longer duration of sleep as compared to the fourth generation(4G)network technologies that demand always for varied control and data signalling based on control base station(CBS)and data base station(DBS).Hence,this paper discusses the energy management in wireless cellular networks using wide range of control for twice the reduction in energy conservation in non-standalone deployment of 5G network.As the new radio(NR)based 5G network is configured to transmit signal blocks for every 20 ms,the proposed algorithm implements withstanding capacity of on or off based energy switching,which in-turn operates in wide range control by carrying out reduced computational complexity.The proposed Wide range of control for base station in green cellular network using sleep mode for switch(WGCNS)algorithm toon and off the base station will work in heavy load with neighbouring base station.For reducing the overhead duration in air,heuristic versions of the algorithm are proposed at the base station.The algorithm operates based on the specification with suggested protocol-level to give best amount of energy savings.The proposed algorithm reduces 40%to 83%of residual energy based on the traffic pattern of the urban scenario.

Development of Energy Management System for Micro Grid Operation

The introduction of several small and large-scale industries,malls,shopping complexes,and domestic applications has significantly increased energy consumption.The aim of the work is to simulate a technically viable and economically optimum hybrid power system for residential buildings.The proposed micro-grid model includes four power generators:solar power,wind power,Electricity Board(EB)source,and a Diesel Generator(DG)set,with solar and wind power performing as major sources and the EB supply and DG set serving as backup sources.The core issue in direct current to alternate current conversion is harmonics distortion,a five-stage multilevel inverter is employed with the assistance of an intelligent control system is simulated and the optimum system configuration is estimated to reduce harmonics and improve the power quality.The monthly demand for residential buildings is 13-15 Megawatts.So,almost 433 Kilo-Watts(KW)of electricity is required every day,and if it is used for 8 h per day,50-60 KW of electricity is needed per hour.The overall micro-grid model’s operation and performance are established using MATLAB/SIMULINK software,and simulation results are provided.The simulation results show that the developed system is both cost-effective and environment friendly resulting in yearly cost reductions.

Low Carbon Building Design Optimization Based on Intelligent Energy Management System

The construction of relevant standards for building carbon emission assessment in China has just started,and the quantitative analysis method and evaluation system are still imperfect,which hinders the development of low-carbon building design.Therefore,the use of intelligent energy management system is very necessary.The purpose of this paper is to explore the design optimization of low-carbon buildings based on intelligent energy management systems.Based on the proposed quantitative method of building carbon emission,this paper establishes the quota theoretical system of building carbon emission analysis,and develops the quota based carbon emission calculation software.Smart energy management system is a low-carbon energy-saving system based on the reference of large-scale building energy-saving system and combined with energy consumption.It provides a fast and effective calculation tool for the quantitative evaluation of carbon emission of construction projects,so as to realize the carbon emission control and optimization in the early stage of architectural design and construction.On this basis,the evaluation,analysis and calculation method of building structure based on carbon reduction target is proposed,combined with the carbon emission quota management standard proposed in this paper.Taking small high-rise residential buildings as an example,this paper compares and analyzes different building structural systems from the perspectives of structural performance,economy and carbon emission level.It provides a reference for the design and evaluation of low-carbon building structures.The smart energy management system collects user energy use parameters.It uses time period and time sequence to obtain a large amount of data for analysis and integration,which provides users with intuitive energy consumption data.Compared with the traditional architectural design method,the industrialized construction method can save 589.22 megajoules(MJ)per square meter.Based on 29270 megajoules(MJ)per ton of standard coal,the construction area of the case is about 8000 m2,and the energy saving of residential buildings is 161.04 tons of standard coal.This research is of great significance in reducing the carbon emission intensity of buildings.

Effective Energy Management Scheme by IMPC

The primary purpose of the Energy Management Scheme(EMS)is to monitor the energy fluctuations present in the load profile.In this paper,the improved model predictive controller is adopted for the EMS in the power system.Emperor Penguin Optimization(EPO)algorithm optimized Artificial Neural Network(ANN)with Model Predictive Control(MPC)scheme for accurate prediction of load and power forecasting at the time of preoptimizing EMS is presented.For the power generation,Renewable Energy Sources(RES)such as photo voltaic(PV)and wind turbine(WT)are utilized along with that the fuel cell is also presented in case of failure by the RES.Such a setup is connected with the grid and applies to the household appliances.In improved model predictive control(IMPC),the set of constraints for the powerflow in the system is optimized by the ANN,which is trained by EPO.Such a tuning based prediction model is presented in the IMPC technique.The proposed work is implemented in the MATLAB/Simulink platform.The energy management capability of the proposed system is analyzed for different atmospheric conditions.The total system cost,life cycle cost and annualized cost for IMPC are 48%,45%and 15%,respectively.From the performance analysis,the cost obtained by the proposed method is very low compared to that obtained by the existing techniques.

Autonomous Multi-Factor Energy Flows Controller (AmEFC): Enhancing Renewable Energy Management with Intelligent Control Systems Integration

The transition to sustainable energy systems is one of the defining challenges of our time, necessitating innovations in how we generate, distribute, and manage electrical power. Micro-grids, as localized energy hubs, have emerged as a promising solution to integrate renewable energy sources, ensure energy security, and improve system resilience. The Autonomous multi-factor Energy Flow Controller (AmEFC) introduced in this paper addresses this need by offering a scalable, adaptable, and resilient framework for energy management within an on-grid micro-grid context. The urgency for such a system is predicated on the increasing volatility and unpredictability in energy landscapes, including fluctuating renewable outputs and changing load demands. To tackle these challenges, the AmEFC prototype incorporates a novel hierarchical control structure that leverages Renewable Energy Sources (RES), such as photovoltaic systems, wind turbines, and hydro pumps, alongside a sophisticated Battery Management System (BMS). Its prime objective is to maintain an uninterrupted power supply to critical loads, efficiently balance energy surplus through hydraulic storage, and ensure robust interaction with the main grid. A comprehensive Simulink model is developed to validate the functionality of the AmEFC, simulating real-world conditions and dynamic interactions among the components. The model assesses the system’s reliability in consistently powering critical loads and its efficacy in managing surplus energy. The inclusion of advanced predictive algorithms enables the AmEFC to anticipate energy production and consumption trends, integrating weather forecasting and inter-controller communication to optimize energy flow within and across micro-grids. This study’s significance lies in its potential to facilitate the seamless incorporation of RES into existing power systems, thus propelling the energy sector towards a more sustainable, autonomous, and resilient future. The results underscore the potential of such a system to revolutionize energy management practices and highlight the importance of smart controller systems in the era of smart grids.