Demand-side Energy Management Research Articles

A behaviour digital twin for residential demand response: Modelling intention and motivation to improve the prediction of the likelihood of reaction to behavioural triggers

Background: Residential demand response is a resource in the evolving energy infrastructure which thus far has not achieved its full potential. Amongst the reasons for this underutilisation is a lack of understanding, and therefore predictability, in relation to the uncertainty of the behaviour of human actors and its potential impact on energy demand side management. Optimal model predictive control of energy assets requires a digital twin to operate, however, most approaches so far are focused predominantly on technical indicators only and neglect the individuality of people and their behaviour in the operation. To fully integrate human led actions into such a system, the digital twin must therefore also provide social and psychological indicators to facilitate better predictability of reactions to demand response triggers. Methods: In the following a behaviour digital twin model will be presented based on the theory of planned behaviour and the self-determination theory, which provide well-established and validated tools to capture indicators of intention and motivation. The key contribution of this work is to operationalise and combine these models into a software tool, which continuously adapts its parameters to the evolving behaviour of users and provides up-to-date predictions. Results: The resulting model predicts the likelihood of each individual to react to appropriate demand response triggers, which can be used in model predictive control involving human actors to optimally select whom to target and when. Conclusions: The presented behaviour digital twin aims at bridging the gap between research in psychology to evaluate and assess drivers of behaviour and innovations in the space of model predictive control to optimally facilitate asset operation in residential settings.

A comparative study of demand-side energy management strategies for building integrated photovoltaics-battery and electric vehicles (EVs) in diversified building communities

This study compares four developed energy management strategies for a grid-connected photovoltaic-battery (PVB) system in a district energy system comprising four diverse building communities: campus, residential, office, and commercial. The proposed demand-side energy management scenarios include maximizing photovoltaic self-consumption, cost minimum strategy under time-of-use and two peer-to-peer (P2P) strategies (a Mid-Market Rate (MMR) and a novel demand response (DR) based P2P strategies). Besides, Monte Carlo integrated Markov Chain simulation method was applied to calculate the load distribution of integrated electric vehicles in diversified building communities, and P2P price was generated considering dynamic supply-demand ratio. In term of dynamic energy flow, energy cost and grid interaction, dispatch results of proposed management strategies were simulated and compared in detailed. The results demonstrate that compared to the mainstream MSC and TOU operation strategies, P2P strategies, especially the DR based P2P strategy, effectively alleviate pressure on the utility grid by actively responding to peak electricity demand and utilizing excess renewable energy generation from other communities. It is worth noting that the choice of strategy depends on the role of the building community, whether as a consumer or a prosumer. This study offers insights for decision-makers and stakeholders in managing grid-connected PVB district energy systems in diverse building communities.

Campus sustainable energy management initiatives at Goa Institute of Management

Learning outcomes The learning outcomes of this study are to identify and discuss ways in which energy consumption in a residential educational institute can be reduced by improving demand-side energy management for sustainable development; summarise the challenges that an institute faces in transitioning to a more environmentally friendly mode of operations concerning energy management; illustrate the difference between operating expense and capital expenditure methods used for solar rooftop projects from the perspective of Goa Institute of Management (GIM); and analyse different project proposals for solar rooftop power generation energy using capital budgeting techniques. Case overview/synopsis Dr Ajit Parulekar, director at GIM, was evaluating the steps taken over the past few years for sustainable energy management to understand their impact and consider ways in which to take the environmental sustainability agenda forward. One of the projects that he was considering was the rooftop solar power plant. GIM had received proposals from several different vendors and evaluated three proposals out of these. He needed to decide on the capacity of the rooftop solar power generation and the type of contract that he should get into for the implementation of the project. This case study describes the differences and highlights the advantages and disadvantages of all the mentioned models with respect to GIM. Complexity academic level This case study is suitable for post-graduate level management students, as well as for undergraduate-level finance and management students. Supplementary material Teaching notes are available for educators only. Subject code CSS4: Environmental management.

Multi-objective optimization based demand response program with network aware peer-to-peer energy sharing

Demand-side energy management programs, such as Demand Response (DR) and Peer-to-Peer (P2P) energy sharing, have recently begun to be implemented to optimize the integration of distributed clean energy resources and to leverage generation and demand-side flexibility in the distribution system. In general, if not optimally planned, P2P energy sharing may have a negative impact on power distribution utilities’ primary objectives, such as minimizing energy losses, flattening the load profile, and improving voltage regulation. The DR program is considered as one of the potential mechanisms to alleviate the possible adverse impact of P2P energy sharing while maintaining the interest of peers participating in decentralized energy sharing. The analysis of the impact of P2P energy sharing with varying peers’ physical locations involved utilizing the Voltage Sensitivity Factor (VSF). The influence on the distribution network is contingent upon the difference in VSF values among the individual peers engaged in energy sharing. This study introduced the k-Means algorithm for VSF-based peer clustering, thereby generating the merit order and realizing the network-aware P2P energy transactions for the selected scenarios. A multi-objective optimization model-based real-time pricing-based DR scheme has been formulated to address the adverse impact of P2P energy sharing. Furthermore, this study quantifies the impact of DR potentials, which vary from 5% to 45% in increments of 10%, on the Peak-to-Average Ratio (PAR), overall energy loss, and total energy cost reduction. The modified IEEE 33 bus system incorporating industrial, commercial, and residential consumers’ load profile of 24 h at selected buses has been considered to test and validate the proposition.

Demand-side energy management reimagined: A comprehensive literature analysis leveraging large language models

The landscape of Demand-Side Energy Management (DSM) research is rapidly evolving, shaped by technological innovations and policy developments. This paper presents an exhaustive bibliometric analysis and methodological framework to explore the research trends within the DSM domain. By synthesizing data from Scopus and OpenAlex, we compile a comprehensive dataset of DSM publications that serve as the basis for our analysis. Through rigorous data acquisition and cleaning, we ensure the reliability and relevance of our dataset. We employ state-of-the-art Large Language Models (LLMs) and topic modeling techniques, including GPT and BERTopic, to perform semantic analysis and uncover thematic structures within the literature. Statistical analysis of the literature dataset reveals a steady increase in DSM publications, with significant contributions from prestigious journals and institutions worldwide. We observe that articles are the predominant publication type, while reviews often cite more references and receive higher citation counts. The distribution of publications over time indicates a growing interest in DSM, particularly since 2014. Geographical mapping of institutions highlights key regions contributing to DSM research, with notable outputs from Europe, North America, and East Asia. Coupled with citation network analysis, our approach reveals the influential works and emerging trends that define the scientific progression of DSM research. Our unsupervised topic modeling, powered by BERTopic, clusters the publications into distinct themes, while our advanced visualization techniques using UMAP and t-SNE provide insights into the semantic space of DSM literature. The resulting thematic classification is presented in a hierarchical structure, offering a comprehensive understanding of the field’s focus areas. Our citation network analysis, utilizing force-directed graph computation and edge-bundling algorithms, maps the interconnectivity and impact of research contributions, providing a dynamic view of the field’s evolution. This study not only charts the landscape of DSM research but also offers a methodological blueprint for future bibliometric analyses. The insights gained from this multi-faceted exploration serve as a valuable resource for researchers, policymakers, and industry practitioners looking to navigate the complexities of DSM and contribute to its scientific advancement.

IoT-enabled Smart Energy Management Device for Optimal Scheduling of Distributed Energy Resources

In this paper, a robust Internet-of-Thing (IoT) enabled Smart Energy Management Device (SEMD) is developed for Distributed Energy Resources (DERs) equipped energy consumers. The SEMD communicates with the existing metering infrastructure through serial communication using the Modbus RTU protocol and the RS 485 communication standard. The SEMD’s energy management framework comprises three sequentially operated software modules, viz., data preprocessing, forecasting, and an optimization module, which aims to minimize the energy costs of end consumers. The software and hardware architecture of the developed SEMD is robust enough to fit real-time implementation and makes it user-friendly. The effectiveness of the SEMD has been tested on an experimental test rig at the Indian Institute of Technology Gandhinagar. The demonstration test rig comprises two DERs equipped energy consumers (viz. Energy consumer-A & B). Energy consumer-A comprises a roof-top solar PV system, Battery Energy Storage System (BESS), and conventional flexible/non-flexible household appliances. The Energy consumer-B consists of Vehicle-to-Grid (V2G) enabled Electric Vehicle (EV), lighting, and pump loads. The experimental results obtained in this work demonstrate that using SEMD satisfies all technical constraints and benefits consumers in reducing bill payments. Thus, SEMD is a critical enabler for DER-equipped energy consumers to participate actively in various demand-side energy management programs.

Energy management for demand response in networked greenhouses with multi-agent deep reinforcement learning

Greenhouses are key to ensuring food security and realizing a sustainable future for agriculture. However, to ensure crop growth efficiency, greenhouses consume a significant amount of energy, primarily through climate control and artificial lighting systems. Owing to this high energy consumption, a network of greenhouses exhibits immense potential to participate in demand response programs for power grid stability. In this work, a multi-agent deep reinforcement learning (MADRL) control framework utilizing an actor-critic algorithm with a shared attention mechanism is proposed for energy management in networked greenhouses. A network of renewable energy integrated greenhouses is constructed to interact with the power grid, when necessary, to address the fluctuations associated with renewable energy generation and dynamic electricity prices. The viability and scalability of this multi-agent approach is demonstrated by evaluating its capabilities for a network of five greenhouses of varying capacities. The proposed MADRL-based control approach for demand-side energy management in networked greenhouses demonstrates efficiency in maintaining indoor climate in all greenhouses while ensuring a 28% reduction in net load demand as compared to well-known algorithms.

Development of prediction models for thermostat override behavior in direct load control events

Direct load control (DLC) is a prominent solution for demand-side energy management, where utility providers adjust consumers' temperature setpoints through smart thermostats. However, occupant overrides can have a detrimental impact on the overall efficacy of DLC. This study focuses on understanding override mechanisms and occupant responses during DLC. Real-world data from the Ecobee Donate Your Data (DYD) program is used to analyze user interactions with smart thermostats. Through a comprehensive analysis of four types of variables, a decision tree algorithm was found capable of accurately classifying smart thermostat users in terms of their level of compliance with the DLC program. This classification was based on general information accessible through smart thermostats enabling the evaluation of users' potential engagement in DLC without prior DLC experience. Additionally, a clustering algorithm was used to identify three distinct types of DLC participants characterized by their individual thermal comfort preferences and general habits. A predictive model for thermostat override behavior is developed based on the clustering results. These findings contribute to designing targeted and less disruptive DLC strategies that better align with participants' needs and enhance overall DLC effectiveness.

Exploring thermostat override behavior during direct load control events

Direct load control (DLC) is considered a viable solution to promote demand-side energy management, in which the utility provider adjusts consumers’ temperature setpoints via smart thermostats. Users commonly have the option to interrupt DLC and override them by adjusting their thermostat setpoints. However, the occurrence of overrides can have a detrimental impact on the overall efficacy of DLC. The user discomfort and the fact that an override may increase the load unexpectedly on the grid highlight the importance of understanding override mechanisms during DLC and the uncertainty related to occupants’ responses. This study examined user interactions with smart thermostats during DLC events using real-world data from the Ecobee Donate Your Data (DYD) program. According to the results, 35% of DLC was overridden by users, resulting in higher energy consumption during peak periods. A comprehensive analysis of four types of variables was conducted. A decision tree algorithm was used to classify smart thermostat users into two categories: “compliant users,” who rarely override DLC, and “non-compliant users,” who frequently override DLC, based on general information about their behavior and preferences and without any prior DLC experience. Moreover, three distinct types of DLC participants, characterized by their preferences and behaviors, were identified using a clustering algorithm. Classification results provide utilities with insight into where resources and efforts should be allocated to users who are more likely to comply with DLC. Clustering users into different typologies will enable utilities to design targeted and less disruptive DLC better aligned with the needs of DLC participants.

Review on Causes of Power Outages and Their Occurrence: Mitigation Strategies

For power plant networks in developing countries like Iraq, balancing electricity demand and generation continues to be a major challenge. Energy management (EM) in either demand-side (DS) or generation-side (GS) strategies, which is frequently utilized in Iraq due to a lack of adequate power generation, has a small impact on the power balancing mechanism. Most previous studies in similar countries discussed only the application of DS strategies. The purpose of this paper is to contrast and review various energy management methodologies being used in developing nations facing power outages, to be able to recommend suitable ones according to the country’s situation. To assess potential EM-based solutions to improve the total energy efficiency of the Iraqi electrical community, a thorough and methodical analysis was carried out. The main objective of this review paper is to discuss the causes of power outages and the energy management strategies addressed here as methods to mitigate or avoid power outages. Unlike existing reviews that concentrated on demand-side energy management, this study specifically focuses on power outage causes in developing countries like Iraq rather than all management strategies. It also introduces the consequences of power outages including analysis of distribution power losses, financial loss from power blackouts, and power blackouts in firms in a typical month. Therefore, it presents readers with state-of-the-art strategies and recommends a generation-based EM strategy to mitigate such issues.

Interpretable transformer-based model for probabilistic short-term forecasting of residential net load

Short-term residential load forecasting is of great significance for the demand-side energy management of the grid and the home energy management of resident customers. The massive penetration of distributed renewable energy, especially small-scale solar photovoltaic (PV), in the residential sector urgently requires us to move from traditional load forecasting to net load forecasting. In recent years, deep learning techniques that can improve model forecasting performance have developed rapidly in the field of residential load forecasting. However, the opaque nature of deep learning makes its practical application very difficult. This paper proposes a Transformer-based probabilistic residential net load forecasting method that utilizes quantile regression to quantify uncertainty in future load demand. Meanwhile, to improve the interpretability of deep learning model, local variable selection network is developed to automatically select relevant features and provide feature-level explanations. Additionally, interpretable sparse self-attention mechanism is proposed to extract long-term temporal dependencies. Numerical experiments are carried out with data from real household smart meters. The results show that the proposed model outperforms other state-of-the-art forecasting models. In terms of point forecasting, compared with the most common deep time series forecasting model LSTM, the proposed model decreases by 21.3%, 27.3% and 20.9% On three point forecasting performance metrics. Compared with Vanilla Transformer, the proposed InterFormer decreases by 9.9%, 10.5% and 9.0% On three point forecasting performance metrics. In terms of probabilistic forecasting, compared with LSTM and Vanilla Transformer, the average pinball loss of the proposed model decreases by 26.2% and 17.0%, respectively. Additionally, and most importantly, our model provides users with explanations in terms of feature importance and temporal patterns.

A comprehensive review of machine learning and IoT solutions for demand side energy management, conservation, and resilient operation

The energy consumption of major equipment in residential and industrial facilities can be minimized through a variety of cost-effective energy-saving measures. Most saving strategies are economically viable where several algorithms can be employed to reduce energy consumption to reduce costs to a considerable extent. Machine learning (ML) is one of these techniques. A review of recent research efforts concerning the application of ML strategies to energy conservation and management problems is presented in this study. In addition, ML approaches and strategies for energy-saving problems, management, technologies, and control methods have been discussed. A comprehensive review of all available publications is also used to make observations about past considerations. As a result, it has been concluded that ML is capable of solving a wide range of decision and management problems within a short period of time with minimal energy consumption. In addition, ML perspectives have been viewed from the perspective of emerging communication technologies, instruments, and cyber-physical systems (CPSs), along with the advancement of ultra-durable and energy-efficient Internet-of-Things (IoT) based communication sensors technology. Moreover, a comprehensive review of recent developments in ML algorithms is also included, including safe reinforcement learning (RL), Deep RL, path integral control for RL, and others not previously. Lastly, critical ML considerations such as emergency and remedial measures, integrity protection, fusion with existing robust controls, and combining preventive and emergent measures have been discussed. The implementation of recently applied ML, RL, and IoT strategies for energy management, conservation, and resilient operation is clarified in this paper. The proposed review highlights the advantages and drawbacks of the recent energy conservation strategies. Finally, the perspective solutions have been clarified to cope with the world direction for zero energy buildings.

IOT Based Smart Energy Meter

Abstract: The labour involved in gathering utility metre readings for energy. The main goal of this project is to use the Internet of Things (IoT) to measure electricity consumption in home appliances and produce a bill automatically. IoT is an efficient and cost-effective way to transfer information about energy consumers wirelessly. A distributed architecture must be used to create the energy grid so that it can dynamically include various energy sources. IoT may be used for a variety of smart grid applications, including energy demand side management, distributed energy plant metering, smart metres for energy generation and consumption, and many areas of energy production.

Can social impacts promote residents’ pro-environmental intentions and behaviour: Evidence from large-scale demand response experiment in China

Energy demand-side management is crucial for deep decarbonization. Consumers’ behavioral intentions and real pro-environmental behaviour have been widely discussed separately. However, few studies assessed the intention-behaviour gap and limited empirical evidence has been provided on how to promote the intention-behaviour on residential electricity demand response (DR) project. Here, we identified four intention-behaviour relationship by k-means cluster and analyzed the impacts of normative social influence and peer influence on the intention and behaviour based on large scale incentive-based DR experiment and matching questionnaires data in China. We found that the intention-behaviour gap exists in 77.94% respondents, only 16.09% respondents belong to inclined actors who has both high intention and high electricity savings behaviour. Peer influence and normative social influence can significantly increase intention to participate the DR project and energy-saving behaviour during peak hours, and the impact is larger for respondents with high family endowments and educational level. Specifically, 30.6% of the electricity saving behaviour has been explained directly by the peer influence, while normative social influence electricity saving behaviour through behavioural intention. Besides these, effort expectancy and facilitating conditions also play an important role in promoting the intention and electricity-saving behaviour. Our study further provides feasible solutions to facilitate the residential DR for policymakers.

Efficient and Private Scheduling of Wireless Electric Vehicles Charging Using Reinforcement Learning

Future vehicle-to-grid (V2G) systems require more flexible scheduling to adjust and flatten the peak energy demand. For efficient scheduling and energy trading, the utility provider (UP) needs to keep track of the state of charge (SoC) of vehicle batteries (VBs). However, sharing of SoC of VBs from electric vehicles (EVs) to UP may compromise owner privacy by analyzing the electricity usage in EVs. Therefore, we propose Reinforcement learning (RL)-based demand-side energy management using a rechargeable battery (RB) for enhanced cost-friendly privacy of EVs, efficient scheduling, and accurate billing. With existing Q-Learning-based RL (using <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\epsilon $ </tex-math></inline-formula> -greedy exploration and exploitation), we find that the reward maximization of efficient and private scheduling is often sluggish and incurs convergence issues. Therefore, we develop a genetic algorithm (GA)-based exploration and exploitation, which solves the convergence problems. We develop theoretical analysis and implement numerical results to demonstrate that the proposed GA-based RL framework accelerates convergence and enhances cost-friendly privacy considerably.

An energy demand-side management and net metering decision framework

Demand side management (DSM) and net metering (NM) are two important strategies that can be used in smart energy management systems to help utilities reduce peak loads and increase the penetration of renewable energy, while also enabling customers to save energy and reduce energy bills. In order to effectively implement DSM and NM programs, it is important to identify the customers who are most likely to benefit from these programs and to quantify the economic impact and benefit for both utilities and end-users. This paper proposes a decision framework that utilizes smart meter data and clustering analysis to identify representative load patterns, and a five-stage method for segmenting load patterns to create a consumption intensity matrix. A fuzzy ensemble based multi-criteria decision making framework is then applied to identify the most potential customer groups and quantify the economic benefit and impact. This paper presents a novel approach for identifying potential customers for DSM and NM programs and developing personalized services for energy demand management. The proposed framework is evaluated through a case study, which demonstrates its robustness in decision making and identifies 13 typical load patterns, seven for weekdays and six for weekends, representing diverse consumption behaviors. The weekday load patterns perform better than the weekend load patterns in terms of the five-stage method.

Design of type-2 fuzzy logic controller in a smart home energy management system with a combination of renewable energy and an electric vehicle

Home energy management is one of the most important parts of a smart home that manages the efficient use of energy in the smart home. This paper aims to design two type-2 fuzzy logic controllers in the demand-side energy management system. For this purpose, a combination of renewable energy sources, such as fuel cells, photovoltaic solar panels, vertical axis wind turbines with a helical savonius rotor, and an electric vehicle and energy storage system along with an external grid are utilized to supply the electricity usage of home appliances in a smart home. By using the proposed method, higher quality, more economical, and environmentally friendly energy can be accomplished. To this end, the outputs of the proposed controllers are defined to make appropriate decisions about the generated energy, how to supply energy for own consumption, and controllable loads. The number and distribution of membership functions of type-2 fuzzy logic controllers are selected according to the measured real input data over a year in Tehran, Iran. To demonstrate the effectiveness of the proposed method it is simulated by MATLAB/Simulink software, which indicates that the proposed system causes the smart home to receive 49.186 kWh less electricity energy from the grid to supply daily power for home appliances and therefore can weekly consume approximately 343.95 kWh less electricity energy from the grid. it is shown that after applying the proposed strategy, electricity costs were reduced by 71.5%, and the peak-to-average ratio was reduced by 64.6%.

Development of fuzzy logic-based demand-side energy management system for hybrid energy sources

Demand-side energy management techniques, such as load shielding, shifting, and delaying appliance operation during peak periods, are typically used to reduce electricity costs at the expense of users’ comfort. To address these challenges, especially where operational delays are unacceptable, distributed generation (DG) is frequently incorporated into the grid system to improve power balance and total energy costs. However, dynamic load demands and varying outputs from renewable DG sources such as solar photovoltaic (PV) systems make energy management in microgrids (MGs) extremely challenging. Moreover, most of the existing studies in this domain focus on objective functions that are geared on optimising the economic balance between cost and value of MG operation over a certain time period. Nevertheless, research that took into consideration the stochastic behaviour of DG’s subsystems in addition to cost and benefit of MG operation are still limited. This current study proposed a fuzzy logic control (FLC) integrated energy management system (EMS) for commercial loads with hybrid grid-solar PV/battery energy system. The proposed technique intelligently selects energy sources using the grid energy cost and the state of charge (SoC) of the solar PV/battery at any time of the day. The EMS operate the loads at a reduced cost without any operational delay or shifting. The system was implemented in the MATLAB/Simulink environment, and the techno-economic feasibility of energy cost savings was investigated by comparing the developed scheme with the Homer hybrid energy system model for a hotel building. The developed EMS reduced energy costs by an average of 11.87 % per day and 7.94 % over a 20-year lifetime.

A Bi-Level Framework for Supply and Demand Side Energy Management in an Islanded Microgrid

Time-varying electricity prices in demand response (DR) programs motivate users to change their consumption behavior with the aim of electricity bill cost minimization. Coordinated DR schemes aim to flatten the aggregate load profile and mitigate rebound peaks. However, a comprehensive model which integrates supply and demand sides has to be developed to consider system constraints and renewable energy resources. In this regard, this article proposes a bi-level framework for supply and demand side energy management in an islanded microgrid. In the first level, consumers schedule their appliances’ operation to minimize their expenses. In the second level, a dynamic economic dispatch problem is solved which finds the power generation schedule for the generators resulting in the minimum electricity production cost. A game-theoretic turn-based decentralized algorithm is suggested which manages the interaction between both levels. Simulation results show that the proposed algorithm minimizes users’ electricity bill costs while satisfying power system constraints.

Modeling and implementation of demand-side energy management system

Modeling and implementation of demand-side energy management system