Enhancement, implementation, and performance of an adaptive short-term load forecasting algorithm

Grady, W.M.; Groce, L.A.; Huebner, T.M.; Lu, Qian; Crawford, Melba M.

doi:10.1109/59.116982

Cited by 24 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Forecast errors can be partly mitigated by artificially setting everyday as a weekend day, but this alone is far from sufficient in closing the accuracy gaps [5]. Previous research also investigated adaptive learning schemes where different regions or seasons are considered [8], [9], yet no literature have considered the forecast tasks under an unexpected pandemic. Figure 2 shows the published day-ahead forecast errors for Germany and CAISO on randomly selected days in April of 2019 and April of 2020.…”

Section: Introductionmentioning

confidence: 99%

Using Mobility for Electrical Load Forecasting During the COVID-19 Pandemic

Chen,

Yang,

Zhang

2020

Preprint

View full text Add to dashboard Cite

The novel coronavirus (COVID-19) pandemic has posed unprecedented challenges for the utilities and grid operators around the world. In this work, we focus on the problem of load forecasting. With strict social distancing restrictions, power consumption profiles around the world have shifted both in magnitude and daily patterns. These changes have caused significant difficulties in short-term load forecasting. Typically algorithms use weather, timing information and previous consumption levels as input variables, yet they cannot capture large and sudden changes in socioeconomic behavior during the pandemic.In this paper, we introduce mobility as a measure of economic activities to complement existing building blocks of forecasting algorithms. Mobility data acts as good proxies for the populationlevel behaviors during the implementation and subsequent easing of social distancing measures. The major challenge with such dataset is that only limited mobility records are associated with the recent pandemic. To overcome this small data problem, we design a transfer learning scheme that enables knowledge transfer between several different geographical regions. This architecture leverages the diversity across these regions and the resulting aggregated model can boost the algorithm performance in each regions day-ahead forecast. Through simulations for regions in the US and Europe, we show our proposed algorithm can outperform conventional forecasting methods by more than three-folds. In addition, we demonstrate how the proposed model can be used to project how electricity consumption would recover based on different mobility scenarios.

show abstract

Section: Introductionmentioning

confidence: 99%

Using Mobility for Electrical Load Forecasting During the COVID-19 Pandemic

Chen,

Yang,

Zhang

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…These modified methods include algorithms to produce load forecasts on the short term. Using measurements and automated systems, these algorithms can incorporate the most upto-date data, assess current errors, and adjust regression parameters dynamically (Gross and Galiana 1987;Lu et al 1989;Grady et al 1991;Paarmann and Najar 1995;Huang 1997;Barakat, Al-Qassim, and Rashed 1992;Chen, Wang, and Huang 1995).…”

Section: Advanced Mathematical Algorithmsmentioning

confidence: 99%

Load Forecasting for the Moroccan Electricity Sector

Cox

Silva

Jorgenson

et al. 2021

View full text Add to dashboard Cite

For the three timescales described a total of eight methodologies were tested (three for longterm, three for medium-term, and two for short-term), six of which were included in this report. These methodologies were linear and multivariable regression on the annual (long-term) and hourly (long-and medium-term) basis, load clustering (medium-term), and two auto regressive models for short-term analysis.Based on long-term analysis, depending on growth scenarios, Morocco can expect an annual peak demand in 2030 ranging approximately 8-11 GW with annual energy consumption ranging 55-77 TWh. This is based on an assumed annual gross domestic product (GDP) growth of 2.5%-4.0%. Analysis was conducted for PV deployment between 1-7 GW and resulting net load curves are forecast. See Section 9.2. Additional work will need to be done to evaluate system integration potential, depending on the geospatial placement of PV and other renewable energy (RE) and non-RE deployments.

show abstract

“…Dynamic filters are well suited to on-line digital processing as data are processed recursively. They had been used extensively in estimation problems for dynamic systems [6]. Dynamic filters have the advantage of being able to handle measurements that change with time.…”

Section: Introductionmentioning

confidence: 99%