Load forecasting techniques for power systems with high levels of unmetered renewable generation: A comparative study

Foster, Judith; Liu, Xueqin; McLoone, Séamus

doi:10.1016/j.ifacol.2018.06.245

Cited by 7 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The non-periodic fluctuation of total demand is mainly affected by temperature. Compared with the wind speed acting on the wind turbines and the solar radiation acting on the solar PV panel, the temperature acting on the total demand is not as sensitive [10]. The main feature is that the total demand has a certain degree of response delay.…”

Section: Characteristic Of Delaysmentioning

confidence: 99%

“…solar capacity, weather and calendar variables) to net demand variations. Foster et al investigated a similar strategy using a forward selection regression technique [9] for linear models and correlationbased approach [10] for nonlinear models to select the most appropriate variables. Wang et al proposed a method in [11] to model the conditional forecast residual and use it to improve the results of probabilistic forecasting methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Simultaneous Estimation of Behind-the-Meter Solar and Wind Power at the Bulk Supply Point

et al. 2022

Self Cite

View full text Add to dashboard Cite

The presence of distributed energy resources in private premises has observed a large deployment worldwide driven by favorable governmental policies and technology development. In Northern Ireland, the small-scale generation capacity including solar and wind power has fast-grown accounting for over 17% of peak load in 2021, which makes net demands more volatile and diminishes the accuracy of electric demand forecasting activities. Different from traditional load forecasting methods based on the relationship between heterogeneous variables and net demand fluctuation, a novel approach is proposed to separately forecast the invisible components (e.g. wind & solar power) within the net demand. The critical challenge in the separate forecasting is how to extract behind-the-meter components from the measured net demand data, namely net demand disaggregation, for which continuous wavelet transform is applied to bring out the detailed fluctuation of net demand and an iterative search algorithm is proposed to separate the components in the net demand. The results reported show that the proposed demand disaggregation method contributes to reduce the forecasting error by more than 14%. Methods like the proposed net demand disaggregation can benefit utilities and system operators in performing their daily activities regarding load forecasting in serving areas and power systems they operate.

show abstract

Section: Characteristic Of Delaysmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Simultaneous Estimation of Behind-the-Meter Solar and Wind Power at the Bulk Supply Point

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…This consists of for strategies such as input parameter processing, exponential smoothing, peak load, and temperature factors. Judith Foster et.al [24] performed the day a head load demand forecasting for both linear and nonlinear models by considering new conditions such as auto regressive and exogenous inputs only models with regressors obtained from greedy selection method. J. Izzatillaev and Z. Yusupov [25] introduced two methods namely group method of data handling and ANN to predict the short term load demand for analyzing energy consumption, area of applicability and advantages and disadvantages of power consumption.…”

Section: Introductionmentioning

confidence: 99%

ANN based Day-Ahead Load Demand Forecasting for Energy Transactions at Urban Community Level with Interoperable Green Microgrids

Rao

Padma

2021

IJRER

View full text Add to dashboard Cite

Microgrids are formed with the rapid penetration of renewable energy sources (RES) into the distribution grid (DG) network, which increases the complexity of DG system and at the same time the conventional grid is getting overloaded due to the urbanization. So, to reduce the stress on the utility grid and for safe and reliable operation of microgrid (MG), it is necessary to develop proper Energy Management System (EMS) with adequate real time control. With this motivation in this paper Artificial Neural Network (ANN) is used along with EMS in the proposed microgrid cluster for making energy transactions based on the availability of localized green energy sources. Energy management system uses day-ahead load demand forecasted by ANN for scheduling the energy sources and managing the energy transactions between the proposed system and utility grid during surplus/shortage power conditions. The cluster is formed by interconnecting two interoperable areas and each area associated with two microgrids. The proposed system with ANN in this paper is modeled and implemented in MATLAB software with Simulink and NN tool box. The simulated test results showed the accuracy of ANN for forecasting day-ahead load demand of the proposed system when compared with Linear Regression (LR).

show abstract

“…Then, an Artificial Neural Network (ANN)-based application is proposed to forecast a day-ahead electricity price [9][10][11][12][13]. [14] forecast load with Multi-Layer Perceptron (MLP) compared with a linear model. Recently, Recurrent Neural Networks (RNN) with attention mechanism was applied to forecast short-term solar irradiance [15] and to predict short-term wind power [16].…”

Section: Introductionmentioning

confidence: 99%

Deep reinforcement learning for electricity energy trading

Sanayha

View full text Add to dashboard Cite

The deregulation and liberalization of the energy market in the 1990s prompted short-term electricity trading, allowing energy markets to produce net output over a range of time periods as a result of this decentralized system, most commonly minutes to days ahead of time. The energy industry urgently requires a system that has undergone substantial modernization in place to handle a variety of issues, including the current climate, renewable resources, and the energy framework. In this dissertation, we investigate a deep reinforcement learning framework for both wholesale and local energy trading, which probes the challenge of RL to optimize the real-world problem in the energy exchange. First, we introduce the MB-A3C algorithm for day-ahead energy bidding to reduce WPP's costs. Also, we have illustrated that our model can generate a strategy that obtains a more than 15% reduction in average cost per day in Denmark and Sweden (Nord Pool dataset). Second, the MB-A3C3 approach is carried out and conducted on a large-scale, real-world, hourly 20122013 dataset of 300 households in Sydney, Australia. When internal trade (trading among houses) increased and external trade (trading to the grid) decreased, our multiple agent RL (MB-A3C3) significantly lowered energy bills by 17%. In closing the gap between real-world and theoretical problems, the algorithms herein aid in reducing wind power production costs and customers' electricity bills.

show abstract

Load forecasting techniques for power systems with high levels of unmetered renewable generation: A comparative study

Cited by 7 publications

References 5 publications

Simultaneous Estimation of Behind-the-Meter Solar and Wind Power at the Bulk Supply Point

Simultaneous Estimation of Behind-the-Meter Solar and Wind Power at the Bulk Supply Point

ANN based Day-Ahead Load Demand Forecasting for Energy Transactions at Urban Community Level with Interoperable Green Microgrids

Deep reinforcement learning for electricity energy trading

Contact Info

Product

Resources

About