Load Forecast Uncertainty Research Articles

A generation expansion planning method considering conditional value-at-risk

Abstract The uncertainty of load forecasting can lead to certain risks in generation expansion planning. Different from the traditional large-capacity unit planning in the past, renewable energy has the characteristics of small single-machine capacity and fast construction speed, etc. The uncertainty of load forecasting will have a more significant impact on the results of generation expansion planning including renewable energy. For this reason, this paper proposes a generation expansion planning idea considering load forecast uncertainty based on conditional value-at-risk (CVaR). Firstly, scenario technology is used to simulate the uncertainty, and CVaR is introduced to quantify the risk of generation expansion planning. Then, considering the coordinated development of traditional and renewable energy sources, a generation expansion planning model that considers risk is established to minimize the total costs. Finally, the effectiveness of the model was verified through a case study, and this model has reference significance for balancing the economy and reliability in generation expansion planning.

Cross-level steam load smoothing and optimization in industrial parks using data-driven approaches

This study focuses on the integrated energy production system in industrial parks, addressing the problem of stable load dispatch of equipment under demand fluctuations. A cross-level method for steam load smoothing and optimization is proposed, aiming to achieve stable production and optimal economic performance through three levels of integration: load forecasting, load dispatch, and load regulation. Unlike traditional methods that directly use load forecasting values, heat network elasticity is presented as a buffer between demand and supply. Constraints for minimal changes in equipment load and operational parameters are established for smooth regulation. Industrial cases demonstrate that the load forecasting model has mean absolute percentage errors of 2.44% and 1.68% for medium-pressure and low-pressure steam, respectively, meeting accuracy requirements. The modified supply-side load smoothness is effectively improved by considering heat network elasticity. The method increases boiler efficiency by 1.92%, reducing average coal consumption by 0.92 t/h. Compared to manual operation, the proposed model leads to an average increase of 5.69 MW in power generation and an average reduction of 10.81% in coal-to-electricity ratio. This study verifies the importance of smooth integration across different levels and analyzes the effective response of the proposed method to the uncertainty in load forecasting. The method demonstrates the enormous potential of data-driven methods in achieving safe, economical, and sustainable production in industrial parks.

Load forecast uncertainties can affect transmission expansion decisions

Load forecast uncertainties can affect transmission expansion decisions

A new intelligent hybrid forecasting method for power load considering uncertainty

Accurate load forecasting plays an important role in promoting low-carbon energy and high-quality utilization of electricity, as well as carbon reduction and safety in energy and power systems. The complexity of power system trends leads to widespread uncertainty in load forecasting. This article analyzes the internal mechanisms of load and meteorological factors at the level of multidimensional uncertainty and proposes a bidirectional memory feature hybrid model based on a new intelligent optimization method. The VMD optimized by gray wolf decomposes the electricity load data into different information modal components. The decomposed components are statistically analyzed with multidimensional uncertainty analysis to extract statistical patterns. The GTO part is optimized according to the fitness value to update the population and global optimal solution. Furthermore, convolutional neural networks (CNN) are used to extract potential features from load and meteorological data and enhance the correlation between input and output data. Then, short-term load forecasting is realized by using bi-directional short-term memory neural network (BiLSTM). The results show that the proposed model has smaller errors than several commonly used models, and has higher prediction accuracy. It can provide some new ideas for relevant departments to carry out power economic dispatch work in low-carbon modes.

Integrated Energy Microgrid Economic Dispatch Optimization Model Based on Information-Gap Decision Theory

To address the problems of “difficult to consume” renewable energy and the randomness of power output, we propose the CHP unit joint-operation model with power to gas (P2G) and carbon capture system (CCS) technologies and analyze the operation cost, carbon emission, and “electric-heat coupling” characteristics of this model. A dispatch optimization model is constructed based on the information-gap decision theory under the strategy to further consider the interval uncertainty of renewable energy unit output and load forecast. The optimized-dispatching model effectively solves the fate of renewable unit output and electric-thermal load and provides dispatching strategies for decision-makers to balance risk and capital management.

Stochastic multi-objective optimal sizing of battery energy storage system for a residential home

The deployment of renewable local generation sources at home can help reduce emission contributions within residential homes. Furthermore, the adoption of the residential PV system is increasing as the economics of installation continues to decrease. An optimally sized battery energy storage system can help maximise the benefits of the power generated from the PV systems while being economical. In this paper, a stochastic optimisation problem is formulated to determine the optimal size of the energy storage system and investigate the benefits of uncertainty consideration in the formulation of the sizing optimisation problem. A multi-objective problem is formulated consisting of two objectives: minimise the cost of purchasing the battery energy storage system, and minimise the amount of energy imported from the grid within the period. The stochastic problem is formulated as a two-stage scenario stochastic problem. The optimal sizing and operation problem was formulated as a mixed-integer linear programming problem and solved using the CPLEX solver. This work also utilises a Monte Carlo approach to deal with the uncertainty in load forecasting. Simulation results show that the proposed approach can estimate an optimal battery energy storage system at the current cost of BESS and clearly indicate the benefit of a stochastic approach.

Day-ahead and intraday multi-time scale microgrid scheduling based on light robustness and MPC

An optimized microgrid scheduling model is established considering demand responses, forecast errors, and the effects of uncertainties in different scheduling stages. A day-ahead, intraday, multi-time scale economic scheduling method based on light robust optimization and model predictive control (MPC) is also proposed. In the day-ahead, long-time-scale scheduling stage, light robustness optimization is used to cope with low-frequency components in prediction errors and uncertainties, and mitigates the deviations between the day-ahead scheduling plan and the actual economic scheduling outcomes under source and load forecast errors and uncertainties. At the intraday, short-time-scale stages, the MPC tracks the day-ahead light robustness economic scheduling plan, considering the high frequency components of prediction errors and uncertainties, so as to achieve robust open-loop control, better tracking results, and better economy. Analytical results demonstrate the effectiveness of the method.

An ensemble multi-step M-RMLSSVR model based on VMD and two-group strategy for day-ahead short-term load forecasting

Accurate prediction of the power load is one of the keys to guarantee stable operation of power construction. However, with the surge of power load, the uncertainty of multi-step prediction brings more challenges. To reduce the uncertainty of multi-step short-term power load forecasting, a short-term ensemble day-ahead forecasting model based on variational modal decomposition (VMD) and two-group multi-step strategy is proposed in this paper. Firstly, VMD is used to decompose the original sequence into different sub-sequences. Then, according to the curve characteristics of different sub-sequences, different multi-step strategies are selected to achieve the best prediction effect. Finally, the multi-step results of all sub-sequences are integrated to get the final day-ahead prediction results. Among them, the two groups of multi-step strategies are divided into single-output approaches (Direct, Recursive, DirRec) and multi-output approaches (MIMO, RecMO, DirMO). For the prediction tool of the single-output approaches, a hybrid prediction model with integrating decomposition, feature selection, optimization algorithm and LSSVR is proposed. For the prediction tool of the multi-output approaches, a hybrid algorithm based on multi-kernel RMLSSVR is proposed to enhance the diversity of MLSSVR, and the parameters and weights of multi-kernel MLSSVR are optimized by the PSO with extreme value disturbance without velocity update. Taking the hourly market load data of California as an example, three groups of comparative experiments are carried out on the proposed method. Experiments show that the accuracy of our model is improved by at least 19% compared with the best RecMO strategy.

Using a Bi-Level Optimization Model for Assessing the Impact of Demand Forecast Uncertainty on the Maximum Profit of Power Distribution Companies Owning Energy Storage Systems

This paper proposes a bi-level optimization model to maximize the net revenue of a power distribution company owning energy storage systems. Furthermore, the proposed model also takes into account the uncertainty of load forecast based on a set of multiple scenarios in order to assess its impact on the maximum net revenue of the power distribution company. This bi-level optimization model is transformed into the single-level mixed-integer linear programming model by using the Karush-Kuhn-Tucker optimality conditions and strong duality theorem. This single-level optimization formulation can be effectively solved by using standard commercial solvers such as CPLEX. The proposed model and the effects of the load forecast uncertainty on the net revenue of the power distribution company are validated and analyzed on an IEEE 24-bus meshed transmission grid.

An IGDT-Based Day-Ahead Co-Optimization of Energy and Reserve in a VPP Considering Multiple Uncertainties

The increasing of grid-connected variable renewable energy (VRE) on the demand side causes balance problems in the power system. Thus, dealing with the uncertainty, variability, and consequently, flexibility requirement is becoming an urgent challenge to the power system operators. Virtual power plant (VPP), which bundles different types of small-scale distributed energy resources (DERs) into a single unit for optimization will effectively mitigate those uncertainties. An optimal VPP energy management method is proposed in this article for optimal energy and operating reserve (OR) scheduling. The studied VPP is a cluster of dispersed generating units (including dispatchable and stochastic power sources), flexible loads, as well as storage units. VPP operator has to make decisions based on the uncertainty coming from the stochastic VRE, load demand, as well as market electricity price. Thus, a dynamic risk reserve quantification method is proposed to cover both VRE power and load forecast uncertainties, while information gap decision theory is applied in the unit commitment procedures to study the impact of price uncertainty on the decision-making of VPP operators. Finally, the proposed method is implemented and verified with a case study, and optimal decisions are discussed.

Energy storage capacity planning method for urban integrated energy system considering uncertainty

The conventional energy storage capacity planning method of urban integrated energy system has the problem of fuzzy coupling characteristics, which leads to the small energy storage capacity. A planning method of energy storage capacity of urban integrated energy system considering uncertainty is designed. Calculate the expected output value, to determine the optimal control targets, considering the uncertainty of power load forecasting, load combination forecast load density and space distribution, identification of urban comprehensive energy system coupling characteristics, optimize the distribution ratio of different energies, capture of power and steam energy, energy storage capacity planning design method. Experimental results: The average energy storage capacity planning method of the urban integrated energy system in this paper is 103.844MWh, 91.657MWh and 91.152MWh compared with the other two methods, indicating that the designed urban integrated energy system energy storage capacity planning method is more perfect after fully considering the uncertainty factors.

Research on short-term load forecasting based on accumulated temperature effect and improved temporal convolutional network

Aiming at the uncertainty in short-term load forecasting, and considering the multi-dimensionality of load data and the continuity of time series, this paper constructs a short-term power load forecasting method based on accumulated temperature effect and improved Temporal Convolutional Network. First, the common load influencing factors in the short-term forecasting process were explored, the data were cleaned and correlated, and a quantitative model of temperature variables that considered the effect of accumulated temperature was established, such as the weighted average method and the discrete correction model. Secondly, fully consider and divide the multi-dimensional influencing factors, the time series characteristic quantity is formed by taking advantage of the Temporal Convolutional Network (TCN), extracting potential relationships between time-series data and non-time-series data, and the optimal load estimation value is obtained. Finally, the output of the TCN is fused with the non-time-series data to form a new input feature matrix, and the load prediction is completed through the Back Propagation (BP) Neural Network. The calculation example shows that the TCN-BP model combined with temperature correction can reduce the average absolute error by 1.77%. Under the same forecasting method, the load forecasting accuracy obtained by the weighted average method is higher than that of the discrete correction model. Compared with the single TCN and BP models, the method proposed in this paper increases the training time, but reduces the time loss of prediction and improves the prediction accuracy.

A hybrid deep learning approach by integrating extreme gradient boosting‐long short‐term memory with generalized autoregressive conditional heteroscedasticity family models for natural gas load volatility prediction

AbstractNatural gas load forecasting provides decision‐making support for natural gas dispatch and management, pipeline network construction, pricing, and sustainable energy development. To explain the uncertainty and volatility in natural gas load forecasting, this study predicts the natural gas load volatility. As the natural gas load volatility has the time‐series features, along with long‐term memory, volatility aggregation, asymmetry, and nonnormality, this study proposes a natural gas load volatility prediction model by combining generalized autoregressive conditional heteroscedasticity (GARCH) family models, XGBoost algorithm, and long short‐term memory (LSTM) network. The model first takes the GARCH family models parameters of sliding estimation and meteorological factors as the influencing factors of volatility, and then it screens these influencing factors through the extreme gradient boosting (XGBoost) algorithm. Finally, the selected important features are input into the LSTM network to predict the volatility, and the 90% confidence interval of the volatility is calculated. Compared with a variety of single and combined models, the model proposed in this study has an average reduction of 45.404% in the evaluation index of mean squared error. The experimental results show that the model proposed in this study has a good performance and accuracy in predicting the volatility of natural gas load.

Why have U.S. electricity sales flattened?

Since the 2007–2009 financial crisis, U.S. electricity sales have remained essentially flat despite economic recovery, with all but five states registering declines in electricity sales growth between the two decades on either side of the financial crisis. This paper examines the factors that contributed to the decline in annual average electricity sales growth from 1997 to 2007 to 2009–2019 across states. It finds that, at a national level, most (∼60%) of the decline in sales growth was in the commercial sector. The factors driving declines in sales growth across states were heterogeneous. The only statistically significant factor (p < 0.05) was a decline in the growth of the number of commercial customers, which occurred in all but two states. Over the last decade, customer count growth slowed across both commercial and residential classes in almost all states. These changes may reflect structural economic and demographic changes and require further analysis. The results highlight the structural economic factors — as opposed to energy efficiency and other technological factors — that are currently shaping electricity demand in the United States. They underscore the importance, for electric utilities, system operators, and their regulators, of developing more effective strategies to manage load forecast uncertainty.

조명부하를 고려한 BTM 태양광 발전설비용량 추정 알고리즘

In the era of energy transitions, solar photovoltaic(PV) resources and wind generators are rapidly increasing. Most small-scale solar PV generators in South Korea are located behind-the-meter(BTM). It is difficult for system operators to monitor the amount of PV generations in real time. Therefore, the amount of BTM solar PV generations have caused increasing the uncertainty of load forecasting. In order to improve uncertainty of load forecasting, this paper proposes an algorithm for BTM solar PV capacity estimation considering lighting load. The capacities of BTM solar PV resources are estimated using reconstituted load based temperature filtering process and lighting load estimation. The proposed algorithm was stable for performance on BTM solar PV capacity estimation and improved the accuracy of load forecasting throughout case studies.

Virtual Energy Storage-Based Energy Management Algorithm for Optimally Sized DC Nanogrid

The optimal configuration of energy sources in a nanogrid is important due to intermittency of renewable energy sources and variability in load. An oversized configuration gives the maximum power supply availability to load. However, this would result in higher cost of electricity and excess power generation. Since, dc nanogrids are seldom connected to central grid, the surplus power cannot be sold back. On the other end, economical configuration leads to undersized system, i.e., insufficient generation to meet the load demand. Hence, in general, a multiobjective Grey-Wolf optimization-based optimal sizing algorithm is proposed to get optimal configuration for the area of photovoltaic (PV) panel installation, rating of the wind energy conversion system and capacity of battery energy storage system. Furthermore, there is need for real-time energy management in dc nanogrid due to uncertainty in power generation and load forecasting. A virtual energy storage (VES)-based energy management is proposed in this article to enhance the availability of power supply. The VES concept models the high thermal inertia loads such as air-conditioner and refrigerator analogous to electrochemical battery. A priority-based strategy in VES algorithm turn- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on</small> and turn- <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">off</small> thermal loads during the power deficit and surplus renewable generation receptively. The VES-based energy management does not compromise the comfort of customers and does not change the usual number of switching in thermal loads.

Robust capacity optimization methods for integrated energy systems considering demand response and thermal comfort

Integrated Energy System can realize the cascade utilization of energy, which improves the utilization of energy efficiently and reduce the carbon emission. Taking into account the uncertainty of multi-energy load and renewable energy forecasting, this paper presents a bi-level robust optimization model with demand response and thermal comfort, for the capacity planning and operation problem of Integrated Energy System. The outer level optimization is planning to find the optimal integrated energy system configuration to minimize economic investment, while the inner level is to robustly optimize the system scheduling to simultaneously reduce carbon emissions and dissatisfaction of residents’ participation in demand response. In the simulation, the Pareto front of the multi-objective problem is obtained via NSGA-II algorithm and Gurobi solver, and the best design plans on the Pareto front selected by Topsis method are analyzed and discussed. The results from three modes are compared in simulation, which illustrates the economic and environmental benefits from demand response and thermal comfort. In addition, the impact of different thermal time scales and forecast uncertainties on integrated energy system planning are also discussed. Finally the sensitivity of the influence of the low carbon grid constrains on the optimization is analyzed.

A Comprehensive Network Restoration Model for Active Distribution Network Considering Forecast Uncertainty

The active distribution network management (ADNM) equipped by active distribution networks (ADNs) can enhance the resilience of the network after failure. This paper proposes a novel comprehensive post-event network restoration model and its chance-constrained variant for ADN that considering the dispatch of distributed generation (DG), energy storage system (ESS), demand response (DR), static var compensator (SVC), and network reconfiguration to fully investigate the potential of the ADNM in network restoration. In this paper, the line failure, bus failure, and the uncertainty from load and DG output forecast error are considered. Specifically, the line and bus failures are modeled by an enhanced fictitious network technique, while the uncertainty of load and DG output forecast is modeled by the chance-constrained optimization. The power flow is described by a linearized DistFlow model, and thus the deterministic and chance-constrained network restoration models are programmed by the mixed-integer linear programming (MILP). The proposed deterministic and chance-constrained restoration models are tested on the IEEE 33 bus system. Results demonstrate the effectiveness of the proposed deterministic and chance-constrained network restoration model.

BTM 발전량을 고려한 평일 단기 전력수요예측 알고리즘

The short-term load forecasting(STLF) is necessary for stable and efficient power system operation. With the spread of renewable energy, solar photovoltaic(PV) capacity is increasing. Most solar PV generator in South Korea is classified Behind-the-Meter(BTM) generator, such as small capacity of solar PV and distributed energy resources(DER). BTM generation is one of the main factors that change the load, and it causes the uncertainty of load forecasting. As a high penetration of BTM solar PV generator, load forecasting error is increased. The load forecasting algorithm using reconstituted load method is proposed to accurate load forecasting considering the BTM generation. The load forecast is performed for weekday excluding special days for the case studies in 2019. The proposed algorithm is improved a lot of the accuracy of STLF compared to the former STLF algorithms that do not considering BTM generation.

Multi-objective Planning for Distributed Generation with Consideration of the Randomness of DG output

DG as an alternative clean energy has a rapid development with the advantages of economy, environment protection and so on. But the randomness of power output has an influence on the security and stability of power system. At the same time, the trade-off among multiple optimization goals is often involved in the actual planning process. Therefore, using multi-objective optimization method to plan DG has wider application prospects. In this paper, a multi-objective optimization model of DG planning is established from three aspects of economy, safety and environmental protection. The flexible parameters of power system are introduced to represent the multi-objective optimization sub-functions and constraint conditions. Considering the influence of the stochastic feature of DG output, the multi-objective programming model is transformed into a flexible integrated optimization problem. One of the goals is lower cost, the others are safety and environmental protection. At the same time the objectives of economy, security and environmental protection are optimized comprehensively after DG integration, which are greatly adapted to the actual needs of the power grid planning department. In addition, the introduction of the static voltage stability margin optimization sub objective is of great significance for solving the problems caused by the uncertainty of load forecasting during the planning process.