Demand Response Optimization Research Articles

Integrated demand response optimization strategy considering risk appetite under multi‐dimensional uncertain information

AbstractIntegrated demand response (IDR) is deemed as an effective tool to balance energy supply and demand. User’s uncertain information containing prior uncertain information and posterior uncertain information is a key factor affecting the implementation effectiveness of IDR, but existing studies fail to consider the two types of uncertain information, response risk caused by the uncertain information, and risk appetite comprehensively. Based on the principal‐agent theory of optimal incentive contract under uncertain information and Markowitz's mean‐variance portfolio theory, a new IDR model is established in this paper, and an IDR optimization strategy considering risk appetite under uncertain information is proposed. By proposing the user model considering multi‐dimensional uncertain information and the risk appetite‐based integrated energy service providers (IESP) model based on the principal‐agent theory and Markowitz's mean‐variance portfolio theory, we have achieved effective modelling of the user’s uncertain information and the risk borne by IESP. The arithmetic examples have verified advantages of the model in enhancing the accuracy of user’s actual response prediction and the superiority of incentive strategies, which is beneficial to reduce the cost of IESPs and enhance the benefit of users participating in IDR.

Joint capacity configuration and demand response optimization of integrated energy system considering economic and dynamic control performance

This paper proposes a joint optimization framework for the demand response-capacity configuration design of the integrated energy system (IES), in which both the economic cost and closed-loop dynamic regulation performance under source-load fluctuations are considered. The joint optimization consists of two layers. The upper layer optimizes the time of use electricity price to adjust the demand response of electric load; while the lower layer optimizes the capacity configuration, scheduling instructions and incentive-based thermal load demand response under typical scenarios. To fairly evaluate the impact of equipment capacity and demand response on the dynamic control performance of the IES, multi-parameter programming-based predictive control approach is applied to develop an offline design and scheduling-perceptive control system. A closed-loop dynamic scheduler is then developed based on the control system, which can optimize the operating instruction of each equipment considering their dynamic operating practice. Simulation studies on a typical off-grid combined heat and power IES show that the proposed approach can effectively match the demand response to the capacity configuration of the IES, helping the system achieve a dynamically flexible and economic energy supply.

Integrating few-shot personalized thermal comfort model and reinforcement learning for HVAC demand response optimization

Heating, ventilation, and air conditioning (HVAC) is the most common equipment to maintain personalized thermal comfort (PTC), and a valuable demand response (DR) resource. The challenges to be addressed in HVAC DR optimization include uncertainties of environment, PTC varying by users and rapid decision-making. This paper proposed a fast HVAC DR optimization method that integrates deep reinforcement learning (DRL) and few-shot PTC model. Firstly, we established an HVAC DR optimization framework integrating few-shot PTC model and DRL, in which the few-shot PTC model was utilized to generate user's PTC temperature range served as the constraint and state variable of DRL. Secondly, we proposed a few-shot PTC model, which only required several field samples and had excellent engineering usability. Thirdly, we designed the HVAC DR optimization algorithm considering user's PTC based on proximal policy optimization, which automatically dealt with the uncertainties and made rapid decisions for setting HVAC setpoints of temperature dynamically. Simulation results show that the few-shot PTC model only requires 9 field samples, which is less than 1 % of existing advanced methods. Furthermore, the proposed method converges within 12.6 min under uncertainties on an i7-12700 processor, which is only about 1/2 of the standard deep deterministic policy gradient.

Incentive-based demand response optimization method based on federated learning with a focus on user privacy protection

Considering the flexible capacity and privacy needs of numerous flexible energy users in the context of demand response (DR), this study establishes an influence model (IM) to describe the DR participation capabilities of users considering privacy budget. Using the designed Stackelberg game mechanism that can achieve optimal selection for DR responders from users with different characteristics described by IM, a federated learning (FL)-based optimization method that uses differential privacy (DP) as the data transmission mechanism for DR economic optimal dispatch is proposed. Ideas of controlling the number of participating users and financially compensating for the privacy leakage risk by the FL-based optimization method are the guarantees for users with private data to participate in DR. The performance of the proposed optimization method is also compared with that of the Moth-flame optimization algorithm in a case study, and the guiding value of the former in selecting among user groups with different characteristics is then discussed. Results show that the proposed method exhibits good economic benefits and universal applicability.

Forecasting Energy Consumption with AI: A Review for Sustainable Energy Management

This paper provides a comprehensive review of the integration of artificial intelligence (AI) within the context of Industry 4.0, emphasizing its transformative impact on various industries and its specific applications in energy consumption forecasting for sustainable energy management. Beginning with a historical perspective on industrial evolution, from automation to the current cyber-physical systems era, the review highlights the pivotal role of AI in reshaping manufacturing processes. The article explores the diverse applications of AI in the energy sector, particularly its effectiveness in short-term load forecasting, demand response optimization, and accurate predictions for renewable energy sources like solar and wind. The growing complexity of power systems due to decentralization and the proliferation of grid-connected devices is discussed, underscoring the importance of effective information exchange facilitated by AI. Additionally, the review delves into various models used for energy forecasting, including supervised learning models, artificial neural networks, and deep learning models. The practical applications of AI in power system control, management, energy market pricing, and policy recommendations are outlined, showcasing its potential in optimizing energy efficiency and balancing electricity production and consumption. The practical examples of AI's role in improving predictions of supply and demand, such as Google's subsidiary DeepMind enhancing wind power output forecasts, highlight the real-world impact of these technologies. However, the abstract also acknowledges existing challenges, including insufficient theoretical background, practical expertise, and financial constraints hindering widespread AI adoption in the energy industry. In conclusion, the article offers valuable insights into the current state, challenges, and potential of AI in forecasting energy consumption, providing a roadmap for sustainable energy management across diverse industries.

Artificial Intelligence Integration Driven Smart Grid Alternative Transforming the Future With Renewable, and Non-renewable Energy Sources

The evolving energy landscape necessitates a paradigm shift towards more sustainable, efficient, and resilient systems. This article explores integrating renewable and non-renewable energy sources within Smart Grid alternatives, highlighting the transformative role of Artificial Intelligence (AI). Renewable sources, like solar and wind, are essential for reducing carbon emissions but present challenges due to their intermittent nature. Non-renewable sources provide stability, ensuring a consistent power supply. AI's capabilities in predictive maintenance, demand response, energy forecasting, and grid optimization enhance grid efficiency and adaptability, facilitating the seamless integration of diverse energy inputs. Essential components of standalone systems, including batteries, charge controllers, and power conditioning equipment, are also examined for their contributions to system robustness and reliability. The Smart Grid, supported by AI and advanced technologies, represents a reimagined energy infrastructure poised to meet modern demands while promoting sustainability and resilience. This holistic approach promises a secure energy future, balancing efficiency with environmental stewardship.

Integrated Demand Response Optimization for Consumer With Herd Mentality: A Genetic Simulated Annealing Approach

Integrated Demand Response Optimization for Consumer With Herd Mentality: A Genetic Simulated Annealing Approach

The optimization analysis of multi-type demand-side flexibility resources for renewable energy accommodation in electrical power systems

The significant increase in the proportion of renewable energy access has led to a burgeoning and urgent demand for flexible resources within the power system, and the prominence of flexible resources on the demand side has been underscored. The existing research on demand-side response primarily focuses on addressing peak load challenges, while the optimization of demand response for accommodating renewable energy remains inadequate. Considering this issue, the present study initially examines the flexibility mechanism offered by demand-side resources and proposes a calculation approach for determining the adjustable capacity of demand-side flexible resources. Furthermore, considering the limitations of the load that can be shifted, transferred, and curtailed, we develop a collaborative optimization operational model for accommodating renewable energy with multiple types of demand response. Finally, based on the provincial power grid example, the simulation study verifies the efficacy of a multi-type demand response. The results demonstrate that shiftable and transferable loads can effectively facilitate the accommodation of new energy, and their optimal allocation capacity ratios are 5.8% and 2.3%, respectively.

Economic optimization of microgrid with demand response under source-load uncertainty

ABSTRACT The installed capacity of renewable energy producers has increased, which has accelerated the development of the microgrid system. One of the most significant and difficult issues in the field of microgrids is economic optimization. The reliability of the microgrid is threatened by the unpredictability of renewable energy and the variety of load types. In this study, a two-layer microgrid demand response optimization model that takes into account source-load uncertainty. To address the instability of renewable energy and load demand, this study introduces a hybrid scenario reduction strategy that combines Latin Hypercube sampling and probability distance. Then, a two-layer model is developed an upper optimization model that aims to minimize the demand response cost and a lower optimization model that aims to minimize the overall system cost. Finally, the paper then suggests a teaching-learning crow search algorithm that combines the teaching and learning optimization algorithm with the crow search method to resolve the two-layer optimization model. The simulation results demonstrate the effectiveness of the two-layer optimization model and the superiority of the suggested algorithm in optimizing the economic operation of microgrids.

Rough knowledge enhanced dueling deep Q-network for household integrated demand response optimization

Implementing a household integrated demand response (HIDR) can be an effective solution to save energy and reduce carbon emissions in household multi-energy system (HMES). However, the HMES are subject to uncertainties, and HIDR optimization needs to automatically adapt to the uncertainty and make decisions within 15 min. Therefore, we propose an HIDR optimization method combining rough knowledge and a dueling deep Q-network (DDQN). Firstly, we propose an HIDR optimization framework of rough knowledge and DDQN fusion, in which knowledge is utilized to generate good DDQN samples and serves as action guidance value input to DDQN. Secondly, we formulate the household equipment models and knowledge rules. Subsequently, we design the HIDR optimization algorithm that incorporates rough knowledge into DDQN, focusing on dynamic probability of knowledge sample participation in DDQN learning, knowledge serving as DDQN action adviser, knowledge diversification and dynamic adjustment of DDQN’s random exploration probability. Simulation results show that our method saves energy costs by 5.6% and 0.9% compared to rule-based methods and DDQN, respectively. Additionally, based on an i5-10300H CPU, our method’s convergence time is no more than 13 min even for scheduling interval of 15 min. The average convergence time of our method is 29% of DDQN.

An approach towards demand response optimization at the edge in smart energy systems using local clouds

The fourth and fifth industrial revolutions (Industry 4.0 and Industry 5.0) have driven significant advances in digitalization and integration of advanced technologies, emphasizing the need for sustainable solutions. Smart Energy Systems (SESs) have emerged as crucial tools for addressing climate change, integrating smart grids and smart homes/buildings to improve energy infrastructure. To achieve a robust and sustainable SES, stakeholders must collaborate efficiently through an energy management framework based on the Internet of Things (IoT). Demand Response (DR) is key to balancing energy demands and costs. This research proposes an edge-based automation cloud solution, utilizing Eclipse Arrowhead local clouds, which are based on Service-Oriented Architecture that promotes the integration of stakeholders. This novel solution guarantees secure, low-latency communication among various smart home and industrial IoT technologies. The study also introduces a theoretical framework that employs AI at the edge to create environment profiles for smart buildings, optimizing DR and ensuring human comfort. By focusing on room-level optimization, the research aims to improve the overall efficiency of SESs and foster sustainable energy practices.

Low-carbon demand response strategy of buildings considering load rebound

With the carbon emission ratio of buildings increasing, including buildings in the carbon market, an inevitable trend of social development is created. Buildings are characterized by their strong seasonality, high simultaneity rate, and large load, which is not only the main factor leading to the peak load of a power grid but also an important resource for load control. Under the guidance of dynamic carbon emission prices and real-time electricity prices, it is worth considering how best to balance the peak response ability of buildings and the cause of the secondary peak leading to load rebound. Therefore, this paper proposes a low-carbon demand response strategy for buildings considering load rebound. First, a building load is classified based on how schedulable the load remains, and a building demand response model which considers how the load schedulability is constructed. Second, according to the different rebounds of different building loads, a load rebound optimization model based on linear attenuation was built, and the rebound parameters of various loads were designed. Then, the real-time carbon emission factor, load fluctuation reward, punishment factor, and real-time electricity price are introduced. Thus, the lowest running cost of the building load is taken as the objective function needed to construct a building demand response optimization model while also considering load rebound. Finally, an example analysis is used to verify the effectiveness of the proposed model in reducing the comprehensive operation cost of buildings. The proposed model not only effectively reduces the comprehensive operation cost of buildings, but also reduces carbon emissions and load fluctuations under the premise of satisfying the actual building load, which provides a new idea for the research of a low-carbon demand response strategy of buildings with load rebound.

Artificial Intelligence and Machine Learning for Real-time Energy Demand Response and Load Management

Within this compendium, an exhaustive examination is undertaken to scrutinize the intricate amalgamation of artificial intelligence (AI) and machine learning (ML) techniques within the paradigm of real-time energy demand response and load management. Placing paramount importance on the pervasive significance of AI and ML, this research expounds upon their profound capabilities to adroitly harmonize the delicate interplay between supply and demand, meticulously calibrate the multifarious dimensions of grid stability, and optimize the boundless potential inherent in renewable energy resources. An in-depth analysis ensues, encompassing the deployment of AI algorithms, poised at the vanguard of demand response optimization, and the judicious utilization of ML techniques, flawlessly calibrated to deliver unerring accuracy across varying temporal scales in the realm of load forecasting. Furthermore, the seamless integration of AI into the very fabric of intelligent appliances and Internet of Things (IoT)-enabled systems unfolds, illuminating the path towards energy consumption optimization, ascertaining an intricate tapestry of interconnected devices, and engendering an ecosystem of intelligent load management. Notably, this comprehensive exposition delves into the far-reaching implications for optimal load management and resource allocation, magnifying the transformative potential that AI-driven algorithms hold in precisely balancing energy utilization and deftly managing the intricate interdependencies that permeate load distribution. Through meticulous elucidation, this illuminating manuscript emboldens the reader with insights into the progressive advancements and myriad benefits that the tandem of AI and ML confers upon the dynamic energy sector, charting an unyielding course towards unprecedented resilience and sustainable utilization of our cherished renewable energy resources.

Economic Optimization Scheduling Based on Load Demand in Microgrids Considering Source Network Load Storage

As a large number of flexible elements such as distributed power and flexible load are connected to microgrids, the economic improvement of microgrids has become an important topic through rational energy distribution. In this paper, an economic scheduling model considering the load demand for a microgrid system under the mechanism of a peak–valley tariff is proposed. A mathematical model of the microgrid components is proposed to determine the exchange power between the microgrid and main network. Meanwhile, an improved War Strategy Optimization (WSO) algorithm is proposed to investigate three scenarios: (i) without batteries, (ii) with batteries and (iii) with batteries and demand response. Additionally, demand response optimization is carried out with the Particle Swarm Optimization (PSO) algorithm and the improved WSO is compared with four other algorithms in scenario (iii). The comparison shows that the improved WSO algorithm has a better optimization performance in solving the proposed scheduling model.

Artificial Intelligence Application in Demand Response: Advantages, Issues, Status, and Challenges

In recent years, there has been a significant growth in demand response (DR) as a cost-effective technique of providing flexibility and, as a result, improving the dependability of energy systems. Although the tasks associated with demand side management (DSM) are extremely complex, the use of large-scale data and the frequent requirement for near-real-time decisions mean that Artificial Intelligence (AI) has recently emerged as a key technology for enabling DSM. Optimization algorithm methods can be used to address a variety of problems, including selecting the optimal set of consumers to respond to, learning their attributes and preferences, dynamic pricing, device scheduling, and control, as well as determining the most effective way to incentive and reward participants in DR schemes fairly and effectively. The implementation optimization algorithm needs proper selection to mitigate the cost of energy consumption. Due to that reason, this paper outlines various challenges and opportunities in developing, utilizing, controlling, and scheduling the DR scheme’s optimization algorithm. In addition, several issues in applications and advantages of optimization techniques in artificial intelligence approaches are discussed. The importance of implementing demand response mechanisms in developing countries is also presented. In addition, the status of demand response optimization in demand-side management solutions is also illustrated congruently.

A behavior-orientated prediction method for short-term energy consumption of air-conditioning systems in buildings blocks

The short-term prediction of air-conditioning (AC) energy consumption is a crucial part of building operation optimization and demand-response strategies. However, the AC energy consumption in some types of buildings can be highly stochastic due to arbitrary occupants' behavior of AC usage, which make it had to predict. Currently, very few prediction models considered the impacts of stochasticity related to occupants’ behavior of AC, which leads to less accuracy. To predict the AC energy consumption under stochastic energy use scenarios, this study proposes a behavior-orientated prediction method for short-term AC energy consumption with a case study on teaching building blocks. Firstly, cluster analysis is employed to identify typical patterns to quantify stochastic AC energy use. Then, the AC usage rate is predicted based on weighted k-Nearest-Neighbors method, which can provide a precise prediction of stochastic AC usage rate. Based on these, RandomForest method is used to develop a basic prediction model of AC energy consumption. The importance of each variable is also evaluated. Finally, an enhanced part of prediction is implemented by support vector machine to achieve higher accuracy under stochastic AC energy use scenarios. The results show that both proposed models can accurately predict the AC energy consumption, while the proposed enhanced model can reap significant accuracy improvement under the more stochastic scenarios. This study provides a new approach to predict AC energy consumption in buildings with stochastic AC energy use, which can match the real-time amount with good accuracy. The proposed model can provide an accuracy reference for the demand response strategies and optimization of AC systems operation.

Solar and wind power data from the Chinese State Grid Renewable Energy Generation Forecasting Competition

Accurate solar and wind generation forecasting along with high renewable energy penetration in power grids throughout the world are crucial to the days-ahead power scheduling of energy systems. It is difficult to precisely forecast on-site power generation due to the intermittency and fluctuation characteristics of solar and wind energy. Solar and wind generation data from on-site sources are beneficial for the development of data-driven forecasting models. In this paper, an open dataset consisting of data collected from on-site renewable energy stations, including six wind farms and eight solar stations in China, is provided. Over two years (2019–2020), power generation and weather-related data were collected at 15-minute intervals. The dataset was used in the Renewable Energy Generation Forecasting Competition hosted by the Chinese State Grid in 2021. The process of data collection, data processing, and potential applications are described. The use of this dataset is promising for the development of data-driven forecasting models for renewable energy generation and the optimization of electricity demand response (DR) programs for the power grid.

Hierarchical Low-Carbon Economic Dispatch with Source-Load Bilateral Carbon-Trading Based on Aumann–Shapley Method

Introducing carbon trading is an essential way to decarbonize the power system. Many existing studies mainly consider source-side unilateral carbon trading (UCT). However, there are still rare studies considering source-load bilateral carbon trading (BCT). The effect of source-load BCT on system-wide carbon mitigation is worth studying. To fill this research gap, a hierarchical low-carbon economic-dispatch model with source-load BCT based on the Aumann–Shapley method was proposed. In the first layer, economic-dispatch was conducted to minimize the power-generation costs and source-side carbon-trading costs. Then, based on the carbon-emission flow (CEF) theory, the actual load carbon emissions can be obtained and passed to the second layer. At the second layer, the demand-response optimization was performed to minimize the load-side carbon-trading costs. Finally, the proposed model was tested on the modified New England 39-bus and IEEE 118-bus systems using the MATLAB/YALMIP platform with the Gurobi solver. The results indicate that the proposed model can effectively facilitate peak-load shifting, wind-power consumption, and carbon mitigation. Furthermore, compared with the models only considering source-side or load-side UCT, the proposed source-load BCT model has obvious advantages in carbon mitigation.

Optimal scheduling strategy of a regional integrated energy system considering renewable energy uncertainty and heat network transmission characteristics

In order to overcome the curtailment of wind energy caused by “power determined by heat” of combined heat and power units, this study proposes an integrated demand response strategy and optimization method for heat-electric energy dispatching in a regionally integrated energy system. Firstly, a scenario analysis algorithm based on Latin Hypercube Sampling (LHS) and Backward Reduction (BR) is applied to characterize the probability characteristics of wind power output. In addition, a novel integrated demand response scheduling model with the consideration of transmission characteristics of the heating network, aiming at minimizing the operation cost, is proposed for a regional integrated energy system. To solve this model, the thermos-electric coupling relationship is evaluated and the proportion of thermoelectric demand response is also weighted. The user satisfaction and economy will finally reach a more optimal level, and thereby is solved via CPLEX solver in MATLAB. The simulation results on a actual integrated energy system reveal that the proposed method is capable of optimizing energy consumption and lowering operational costs while maintaining user satisfaction with the uncertain wind generation by leveraging thermal inertia of the integrated energy system.

A comprehensive and modular set of appliance operation MILP models for demand response optimization

Demand response programs are essential to enable accommodating larger shares of variable power generation based on renewable sources, the deployment of which is imperative for decarbonizing the economy and mitigating global warming. Consumers/prosumers are increasingly exposed to and may benefit from time-differentiated energy prices aimed to induce changes in regular consumption patterns. These changes are also beneficial for retailers and grid operators in face of the variability of wholesale market prices, renewable energy availability and grid conditions. The optimization models to be implemented in autonomous home energy management systems require a rigorous modeling of appliance operation to generate effective load scheduling solutions, respecting their physical operation principles and use patterns in everyday life. A balance should be sought between the detail level of optimization models and the computational requirements to generate usable solutions having in mind their implementation in low-cost processors. This paper presents a comprehensive and modular set of mixed-integer linear programming models aimed at enabling their seamless incorporation in home energy management systems, allowing for the integrated optimization of all energy resources (exchanges with the grid, load management, electric vehicle and stationary battery, local microgeneration). Detailed energy consumption optimization models for shiftable, interruptible and thermostatic loads are presented, also including the power cost component and ways of dealing with user's discomfort. The modular models are presented in a building block manner enhancing the flexibility of their utilization in overall models with different objective functions encompassing the economic and comfort dimensions. Computational results are presented for a case study using actual data, which considers a time-of-use tariff with six periods. In addition to comparing with a plain tariff scheme, different consumer profiles are simulated to assess the impact of comfort requirements on cost. These results show that whenever consumers have the flexibility to change their consumption patterns, they are able to lower the net electricity bill by having an energy management system endowed with the models herein proposed to make optimized decisions on their behalf.