Customer Baseline Load Research Articles

Customer baseline load estimation for virtual power plants in demand response: An attention mechanism-based generative adversarial networks approach

The virtual power plant (VPP) that aggregates demand-side resources, is a new type of entity to participate in the electricity market and demand response (DR) program. Accurate customer baseline load (CBL) estimation is critical for DR implementation, especially the financial settlement in incentive-based DR. However, this is a challenging task as CBLs cannot be measured and are not equal to actual loads when DR events occur. Moreover, VPPs with different aggregation scales form heterogeneous electricity customers, which increases the difficulty of CBL estimation. In order to address this challenge, this paper proposes a novel deep learning-based CBL estimation method for varied types of electricity customers with different load levels. Specifically, we first transform the CBL estimation problem into a time-series missing data imputation issue, by regarding actual load sequences as CBL sequences with missing data, during DR periods. Then, we propose an attention mechanism-based neural network model to learn load patterns and characteristics of various CBLs, and also create the DR mask to avoid the disturbance of actual loads of DR periods on CBL’s normal pattern. Further, we develop the generative adversarial networks (GAN)-based data imputation framework to produce the corresponding complete CBL sequence according to the actual load sequence, and then recover the missing values accordingly. Finally, comprehensive case studies are conducted based on public datasets, and our proposed method outperforms all benchmarks, where the mean and standard deviation of its estimation percentage error are 5.85% and 1.74%, respectively. This validates the effectiveness and superiority of the proposed method.

Strategies for electric vehicle bidding in the German frequency containment and restoration reserves market

Germany's transition to renewable energy sources for a sustainable future has introduced variability and instability in the national power grid, thus necessitating innovative solutions to ensure grid stability and reliable energy supply. The integration of electric vehicles (EVs) in the German balancing market, notably within the frequency containment reserve and automatic frequency restoration reserve sectors, is a potential stabilizing force. However, the complexity of this integration, governed by specific rules in the German energy sector, has not been fully explored. To bridge this gap, a robust schedule that maximizes the economic viability and reliability of EV integration by considering critical elements such as uniform product size, alignment with local market product duration, and customer baseline load is proposed in this study. Mixed-integer linear programming is used for the effective integration of EVs in the German balancing market. This framework emphasizes computational efficiency, providing market participants with precise insights into EV behavior and strategies for energy storage optimization. The effectiveness and reliability of the proposed method are substantiated through case studies. The proposed method consistently achieved maximum profit and a significant reduction in the number of failed cases, which underscores its applicability in real-world scenarios.

An investigation of the customer baseline load (CBL) calculation for industrial demand response participants –A regional case study from Iran

In today's power systems, performing demand response (DR) programs is considered a solution for sustainable operation, carbon reduction, and facilitating renewable energy integration. Currently in Iran, several DR programs are available to industrial customers. Moreover, the design of a market-based DR has recently been considered by the power system operator. In DR programs, customer baseline load (CBL) is used to determine the level of customers’ demand reduction and is key to the billing process; thus, the CBL calculation methodology can directly affect the efficiency of these programs. This paper investigates the current CBL calculation method in Iran's power system and, as its main contribution, proposes some effective methods to reduce the probable incentives for CBL manipulation. In order to show the effectiveness of the methods, a real study case is considered based on the historical load data of the industries in the Khorasan region.

Distributed Baseline Load Estimation for Load Aggregators Based on Joint FCM Clustering

Residential customer baseline load (CBL) estimation is critical to Demand Response (DR) implementation. The accurate estimation of baseline load is challenging because of the complexity and uncertainty of customer behavior. In addition, conventional methods are based on centralized datasets, which may be unrealistic considering the customer privacy issues and the geographical distribution of data storage under multiple load aggregators (LA). This paper proposes a fully distributed framework for CBL estimation based on joint Fuzzy C-Means (FCM). The fuzzy membership matrix is utilized to capture the load features. LAs can acquire the joint clustering result through limited information exchange in a privacy-preserving way. At last, the CBL is calculated through an inverse process of clustering. Case studies have been conducted utilizing the dynamic Time of Use trail data from Low Carbon London project. Compared with widely used averaging methods, the proposed method significantly improves the estimation accuracy.

Stochastic Demand Response Management Using Mixed-Strategy Stackelberg Game

Stochastic solutions to the demand response (DR) problem enable utility companies to address the uncertainties in customer demand. In another words, stochastic DR solutions are recently developed considering storage systems, renewable resources, and the electricity market. However, a DR approach for uncertain response of consumers to an incentive-based DR problem has not been studied before. This article presents a new prospective stochastic DR approach by using a mixed-strategy Stackelberg game (MSG) between an aggregator and residential customers during peak load periods. We use real dataset collected by a local utility. We cluster customers based on customer baseline load (CBL), select one reference customer for each cluster, and model the MSG between that customer and the aggregator. We map the MSG to subgames and prove the equilibrium using mixed-strategy Nash Equilibrium. Our results show that the peak load can be shifted to off-peak hours for almost 40 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> of load, and players gain monetary rewards in return.

Demand Response Impact Evaluation: A Review of Methods for Estimating the Customer Baseline Load

Climate neutrality is one of the greatest challenges of our century, and a decarbonised energy system is a key step towards this goal. To this end, the electricity system is expected to become more interconnected, digitalised, and flexible by engaging consumers both through microgeneration and through demand side flexibility. A successful use of these flexibility tools depends widely on the evaluation of their effects, hence the definition of methods to assess and evaluate them is essential for their implementation. In order to enable a reliable assessment of the benefits from participating in demand response, it is necessary to define a reference value (“baseline”) to allow for a fair comparison. Different methodologies have been investigated, developed, and adopted for estimating the customer baseline load. The article presents a structured overview of methods for the estimating the customer baseline load, based on a review of academic literature, existing standardisation efforts, and lessons from use cases. In particular, the article describes and focuses on the different baseline methods applied in some European H2020 projects, showing the results achieved in terms of measurement accuracy and costs in real test cases. The most suitable methodology choice among the several available depends on many factors. Some of them can be the function of the Demand Response (DR) service in the system, the broader regulatory framework for DR participation in wholesale markets, or the DR providers characteristics, and this list is not exclusive. The evaluation shows that the baseline methodology choice presents a trade-off among complexity, accuracy, and cost.

Spatio-Temporal Two-Dimensions Data Based Customer Baseline Load Estimation Approach Using LASSO Regression

Customer baseline load (CBL) estimation plays a crucial role in financial settlement for incentive-based demand response (DR). Most current CBL estimation methods utilize temporal auto-correlation or spatial cross-correlation to estimate the CBL, which would produce large errors when the load pattern suddenly changes on the DR event day or the number of CONTROL customers is insufficient. To this end, a spatio-temporal CBL estimation approach is proposed to improve the estimation accuracy in this paper. First, all CONTROL customers are grouped into several non-overlapping clusters by the K-means algorithm. Each DR customer is then matched to the most similar cluster according to the similarity between the typical load pattern and the cluster center. Second, spatio-temporal features are extracted from the historical load data of the DR customer itself and the load data of the matched CONTROL customers on the DR event day. Third, the least absolute shrinkage and selection operator (LASSO) regression model is constructed to estimate the CBL of each DR customer. The effectiveness and superiority of the proposed approach have been verified on a real-world load dataset.

Privacy-Preserving Baseline Load Reconstruction for Residential Demand Response Considering Distributed Energy Resources

Customer baseline load (CBL) reconstruction is a critical problem in residential demand response. The difficulty of residential CBL lies in the variability of both irregular consumption and on-site distributed energy resources. Targeting the CBL reconstruction of residential prosumers, a regression-based estimation scheme is proposed using stacked autoencoders (SAEs) under the federated learning (FL) framework. In the FL framework, each residential unit (RU) stores and trains data locally without sharing them with neighboring RUs or the independent third party (ITP) responsible for CBL reconstruction. Local updates containing no load information are exchanged with the ITP (server) for the training improvement. The FL framework can, thus, protect the privacy of customers. Experimental results show that the proposed FL-based cascaded SAE outperforms the baseline on all tests and achieves up to 62.5% improvement in reducing reconstruction error. Moreover, it has enhanced privacy-preserving knowledge-sharing ability, higher efficiency, and better stability.

Impact of Demand Response on Reliability Enhancement in Distribution Networks

This paper presents an innovative instantaneous pricing scheme for optimal operation and improved reliability for distribution systems (DS). The purpose of the proposed program is to maximize the operator’s expected profit under various risk-taking conditions, such that the customers pay the minimum cost to supply energy. Using the previous information of the energy consumption for each customer, a customer baseline load (CBL) is defined; the energy price for consumption costs higher and lower than this level would be different. The proposed scheme calculates the difference between the baseline load and the consumption curve with the electricity market price instead of calculating the total consumption of the customers with the unstable price of the electricity market, which is uncertain. In the proposed tariff, the developed cost and load models are included in the distribution system operation problem, and the objective function is modeled as a mixed integer linear programming (MILP) problem. Also, the effect of demand response (DR) and elasticity on the load curve, the final profit of the distribution system operator, and payment risk and operation costs are examined. Since there are various uncertainties in the smart distribution grid, the calculations being time-consuming and volumetric is important in the evaluation of reliability indices. Thus, when computation volume can be decreased and computation speed can be increased, analytical reliability analysis methods can be used, as they were in the present work. Finally, the changes in the reliability indices were calculated for the ratio of the customers’ sensitivity to the price and the customers’ participation in the proposed tariff using an analytical method based on Monte Carlo simulation (MCS). The results showed the efficiency of the proposed method in increasing the operator profit, reducing the operation costs, and enhancing the reliability indices.

국민DR 100만kW를 위한 스마트가전 수요반응 시범사업 실적분석 및 경제성 전망

With a gradual increase of electricity demand and high penetration of renewable energy resources to electrical power grid, the demand response (DR) program is emerging as an alternative to reduce peak power and increase the power system stability. This study provides an analysis of data from a DR pilot project in which 140 appliances were participated. We proposed a customer baseline load (CBL) estimation method without historic power consumption data, and calculated peak reduction effects of smart appliances. Moreover, we analyzed the average respond success rate, and investigated the effectiveness respond signal. In addition, an economic assessment of smart appliances (S/A) DR for 1GW peak reduction is performed based on the data from the pilot project. We evaluate the benefits from the application of S/A DR in the utility and participant perspective. While participant’s benefit is an incentive from the reduction of power consumption, the utility’s benefits can include deferral effects of the generation capacity and delay of transmission and distribution infra construction. Also sensitivity analysis according to various factors that affect the benefit cost ratio of the project is performed. Finally, we proposed a moderate incentive program which can attract the customer to participate the DR project.

PV-Load Decoupling Based Demand Response Baseline Load Estimation Approach for Residential Customer With Distributed PV System

Customer baseline load (CBL) estimation is very important in demand response (DR) program. Due to the increasing installation of distributed photovoltaic system (DPVS), the load patterns of residential customers become more complex and random. The actual load power of the customer is coupled with the DPVS output power, which makes it more difficult to estimate CBL. Since the electricity meter can only measure the net load data, this article proposes a PV-load decoupling approach to improve the CBL estimation accuracy in the presence of DPVS. CBL is the difference between actual load power and DPVS output power, so the CBL estimate is converted into two sub-problems: the estimation of actual load power and the estimation of DPVS output power. First, the actual load power of DR customers is estimated based on the load power of the control group customers. Then, the DPVS output during DR period is obtained based on the DPVS output estimation model. Finally, CBL is estimated based on the actual load power and DPVS output power. In order to verify the effectiveness and feasibility of the approach, two real datasets from Sydney and Austin are used to simulate the CBL estimation. Compared with the net load directly estimating the CBL, the comparison results indicate that the proposed method shows better accuracy performance.

Building a Better Baseline for Residential Demand Response Programs: Mitigating the Effects of Customer Heterogeneity and Random Variations

Peak-time rebates offer an opportunity to introduce demand response in electricity markets. To implement peak-time rebates, utilities must accurately determine the consumption level if the program were not in effect. Reliable calculations of customer baseline load elude utilities and independent system operators, due to factors that include heterogeneous demands and random variations. Prevailing research is limited for residential markets, which are growing rapidly with the presence of load aggregators and the availability of smart grid systems. Our research pioneers a novel method that clusters customers according to the size and predictability of their demands, substantially improving existing customer baseline calculations and other clustering methods.

Implementation of a Demand-Side Management Solution for South Korea’s Demand Response Program

In this paper, we show the development of a demand-side management solution (DSMS) for demand response (DR) aggregator and actual demand response operation cases in South Korea. To show an experience, Korea’s demand response market outline, functions of DSMS, real contracted capacity, and payment between consumer and load aggregator and DR operation cases are revealed. The DSMS computes the customer baseline load (CBL), relative root mean squared error (RRMSE), and payments of the customers in real time. The case of 10 MW contracted customers shows 108.03% delivery rate and a benefit of 854,900,394 KRW for two years. The results illustrate that an integrated demand-side management solution contributes by participating in a DR market and gives a benefit and satisfaction to the consumer.

Smart Households’ Aggregated Capacity Forecasting for Load Aggregators Under Incentive-Based Demand Response Programs

The technological advancement in the communication and control infrastructure helps those smart households (SHs) that more actively participate in the incentive-based demand response (IBDR) programs. As the agent facilitating the SHs' participation in the IBDR program, load aggregators (LAs) need to comprehend the available SHs' demand response (DR) capacity before trading in the day-ahead market. However, there are few studies that forecast the available aggregated DR capacity from LAs' perspective. Therefore, this article proposes a forecasting model aiming to aid LAs forecast the available aggregated SHs' DR capacity in the day-ahead market. First, a home energy management system is implemented to perform optimal scheduling for SHs and to model the customers' responsive behavior in the IBDR program; second, a customer baseline load estimation method is applied to quantify the SHs' aggregated DR capacity during DR days; third, several features which may have significant impacts on the aggregated DR capacity are extracted and they are processed by principal component analysis; and finally, a support vector machine based forecasting model is proposed to forecast the aggregated SHs' DR capacity in the day-ahead market. The case study indicates that the proposed forecasting framework could provide good performance in terms of stability and accuracy.

Demand Response Strategy of Energy Prosumer Based on Robust Optimization Through Aggregator

In this article, the optimal operation strategy for the aggregator to participate in demand response (DR) market is proposed. First, on the day before the occurrence of DR, Customer Baseline Load (CBL)-based load forecasting is performed using historical load data and a day-ahead scheduling is implemented to minimize electric charges by using peak reduction and arbitrage trading. If the demand response occurs, distributed energy resources (DERs) bid power reduction capacity to the aggregator. In Korea tariff system, demand charges determined by peak load are very expensive. Therefore, DERs should not update their peak load due to demand response market participation. The uncertainty of load prediction is modeled using the average value of mean absolute percentage error (MAPE), and robust optimization (RO) is implemented to determine a bidding capacity, thereby preventing the peak form being updated due to prediction error. Then, the aggregator decides the capacity to participate in DR market by considering the bidding capacity and priority. It presents the method to determine the incentive for participation in DR using a logarithmic barrier function. To evaluate the performance of the proposed algorithm simulation was performed by constructing a scenario for prediction error and mandatory reduction capacity.

Customer directrix load-based large-scale demand response for integrating renewable energy sources

Large-scale demand response (DR) is a critical enabler to integrate significant renewable energy sources (RES) into power systems. Current customer baseline load (CBL)-based DR schemes face obstacles in large-scale deployments due to their centralized form, unfair DR performance measurement, and poor effect on decision making approach of customers. To bridge the gaps, this paper proposes the concept of customer directrix load (CDL), which is the desired load profile for customers from the view of the entire DR program, and a novel CDL-based DR scheme. Additionally, an optimization problem considering time-coupling constraints is formulated to help customers respond to the CDL. The computationally intensive problem is then translated into a quadratic programming problem in each time slot using Lyapunov optimization approach. A closed-form solution exists and ensures that the optimal decision is reached in real-time efficiently. Test systems are generated using data from PJM and Open Energy Information. The online algorithm and fairness performance of the proposed scheme are validated in a small system through benchmark comparisons. Further tests on a large-scale system show that the CDL-based DR scheme can help the power system integrate considerably more RES.

Residential Customer Baseline Load Estimation Using Stacked Autoencoder With Pseudo-Load Selection

Accurate estimation of customer baseline load (CBL) is a key factor in the successful implementation of demand response (DR). CBL technologies implemented at utilities currently are primarily designed for large industrial and commercial customers. The U.S. Federal Energy Regulatory Commission (FERC) order 745 states that DR owners, including residential customers, can sell their load reduction in the wholesale market. However, since residential load is random and un-schedulable, this tends to inherently degrade the effectiveness of existing CBL technologies. In this paper, a novel SAE based CBL method for residential customers that uses the data reconstruction capability of a stacked autoencoder (SAE) is described. In the model, two SAEs are synchronously trained—one SAE generates a pseudo-load pool and the second one is used to select a pseudo-load to reconstruct a residential CBL. A support vector machine (SVM) classifier is self-trained to conduct the pseudo-load selection. The proposed strategy is validated using a real data set consisting of 328 residential customers’ smart meter readings. Benchmarks from other machine learning techniques and existing CBL methods are compared with the proposed method. Test results show that the accuracy of the residential CBL reconstruction significantly improves when compared with existing methods, such as HighXofY and exponential moving average.

Capacity and output power estimation approach of individual behind-the-meter distributed photovoltaic system for demand response baseline estimation

Accurate customer baseline load (CBL) estimation is critical for implementing incentive-based demand response (DR) programs. The increasing penetration of grid-tied distributed photovoltaic systems (DPVS) complicates customers’ load patterns, making the CBL estimation more difficult because the volatile actual load and the intermittent PV output power are coupled together. A PV-load decoupling framework is proposed in this paper to address the above issue. The basic idea is to decouple the actual load power and the PV output power, then estimate them separately. To this end, historical PV output power data of each individual DPVS is required. However, pure historical PV output power data is usually unavailable for small-scale DPVSs, since they are normally located behind the meter, thus only the net load (i.e. actual load power minus PV output power) data is metered. Therefore, this paper proposes a machine learning approach to disaggregate the output power of each individual DPVS from net load data. The proposed approach includes two stages: DPVS capacity estimation and PV output power estimation. The first stage consists of two steps. First, a net load curve optimal pairing-based feature extraction method is proposed to extract features from the discrepancy between two different net load curves of the customers under heterogeneous weather conditions. Second, a multiple support vector regression-based ensemble model with the input features extracted in the first step is established to estimate the DPVS capacity. In the second stage, the output power of each DPVS is estimated by its capacity multiplied by the output power of a standard DPVS. Case studies using a real dataset from Sydney indicate that the proposed approach shows a promising performance on PV output power estimation and can significantly improve the CBL estimation accuracy for customers with DPVSs.

Defining virtual control group to improve customer baseline load calculation of residential demand response

One of the critical challenges in demand response is to calculate the customer baseline load, and it can be particularly challenging for residential demand response where each household’s daily electricity load can vary randomly and significantly. A general and widely accepted enhancement method for customer baseline load is to set up an independent control group, but it requires a careful selection process and exclusion of the selected customers. In this paper, we propose the concept of virtual control group that can provide the benefits of control group without requiring the main burdens. A virtual control group is adaptively formed for each demand response event using the pre-collected participation information (through a mobile app in our pilot program), and it can perform well when used with difference-in-differences that can handle the selection bias. The customer baseline load calculation method that combines virtual control group and difference-in-differences is named as V-CBL in this study. Using a real-world dataset collected from a pilot residential demand response program, we evaluate V-CBL’s robustness against selection bias and assess V-CBL’s mean error performance and mean absolute error performance against the traditional models. Besides the analysis based on the non-event days, we provide an analysis on the actual DR event days as well.

Demand Response Baseline Load Forecasting Based on the Combination of Time Series and Kalman Filter

The customer baseline load is an important reference for the industrial and commercial users to participate in the demand response project, and is affected by various factors such as the environment and user electricity usage. In order to improve the accuracy of the baseline load forecasting of industrial and commercial users, a demand response baseline load forecasting model based on time series and Kalman filter combination is proposed. The marginal contribution rate of the single forecasting model to the combined model is obtained by the Shapley value method, then gets optimal prediction results. The case results show that the Kalman filter model has higher prediction accuracy in the period of stable load fluctuation, and the ARMA model has higher prediction accuracy in the period of large load fluctuation, and the combined prediction model combines the advantages of both models and reduces the single model is affected by the time factor in the prediction process, which improves the overall prediction accuracy and expands the scope of application.