Multi-period forecasting and scenario generation with limited data

Ríos, Ignacio; Wets, Roger J.-B.; Woodruff, David L.

doi:10.1007/s10287-015-0230-5

Cited by 35 publications

(30 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Algorithmic details of regression using epi-splines are provided in [17]. That paper provides an example illustrating how to fit a regression function from the weather for day d to R 24 , using data from some set of daysD as…”

Section: Estimating Regression Functionsmentioning

confidence: 99%

“…Scenarios are then constructed by constructing paths whose trajectories are determined by selecting a specific error category for each day part. This process is a specific instantiation of the general scenario generation methodology detailed in [17]. Let H be the set of hours that define a partition of the hours in a day, specified as follows:…”

Section: Scenario Generationmentioning

confidence: 99%

“…To mitigate this difficulty, a general method of generating scenarios based on approximating rather than sampling from error distributions was proposed and motivated by load scenario generation in [17]. Our contributions in this paper are (1) the development of methods to incorporate critical data such as seasons, weather patterns, day-of-the-week, and load zones; (2) description of the details necessary to actually generate scenarios; (3) in combination with the companion paper, optimization models that use them and tests on the efficacy of those scenarios; and (4) the scenarios themselves as a resource for the research community.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward scalable stochastic unit commitment. Part 1: load scenario generation

et al. 2015

Self Cite

View full text Add to dashboard Cite

Unit commitment decisions made in the day-ahead market and during subsequent reliability assessments are critically based on forecasts of load. Tra-ditional, deterministic unit commitment is based on point or expectation-based load forecasts. In contrast, stochastic unit commitment relies on multiple load sce-narios, with associated probabilities, that in aggregate capture the range of likely load time-series. The shift from point-based to scenario-based forecasting necessi-tates a shift in forecasting technologies, to provide accurate inputs to stochastic unit commitment. In this paper, we discuss a novel scenario generation method-ology for load forecasting in stochastic unit commitment, with application to real data associated with the Independent System Operator for New England (ISO-NE). The accuracy of the expected scenario generated using our methodology is consistent with that of point forecasting methods. The resulting sets of realistic scenarios serve as input to rigorously test the scalability of stochastic unit com-mitment solvers, as described in the companion paper. The scenarios generated by our method are available as an online supplement to this paper, as part of a novel, publicly available large-scale stochastic unit commitment benchmark. Abstract Unit commitment decisions made in the day-ahead market and during subsequent reliability assessments are critically based on forecasts of load. Traditional, deterministic unit commitment is based on point or expectation-based load forecasts. In contrast, stochastic unit commitment relies on multiple load scenarios, with associated probabilities, that in aggregate capture the range of likely load time-series. The shift from point-based to scenario-based forecasting necessitates a shift in forecasting technologies, to provide accurate inputs to stochastic unit commitment. In this paper, we discuss a novel scenario generation methodology for load forecasting in stochastic unit commitment, with application to real data associated with the Independent System Operator for New England (ISO-NE). The accuracy of the expected scenario generated using our methodology is consistent with that of point forecasting methods. The resulting sets of realistic scenarios serve as input to rigorously test the scalability of stochastic unit commitment solvers, as described in the companion paper. The scenarios generated Disciplines Industrial Engineering | Systems Engineering

show abstract

Section: Estimating Regression Functionsmentioning

confidence: 99%

Section: Scenario Generationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Toward scalable stochastic unit commitment. Part 1: load scenario generation

et al. 2015

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the end, only about 15 to 25 days per year may remain on which the forecast must rely, making traditional techniques based on time-series or stochastic differential equations inaccurate. Here, we briefly describe the epi-spline-based construction of a stochastic process of the next day's electricity load as laid out in Feng et al [13,14]; see also Rios et al [16]. Epi-splines enter at three points.…”

Section: Examplesmentioning

confidence: 99%

“…We can repeat this process for each hour h independently, but that would not take into account conditioning. Clearly, if the load at hour h is substantially higher than the load tentatively projected by the regression curve, one should take this into account when looking at the distribution of the errors at a later time h + h. This can be done systematically by restricting the samples of the error at time h + h to those that come from (observed) load trajectories that at time h had similar deviations from the overall drift of the process, i.e., from the regression curve (for details, see Feng et al [13,14], Rios et al [16]). Since this limits the number of data points on which a single density estimate must rely, the significance of being able to include external information in the epi-spline framework is further highlighted.…”

Section: Examplesmentioning

confidence: 99%

From Data to Assessments and Decisions: Epi-Spline Technology

Røyset¹,

Wets²

2014

Bridging Data and Decisions

Self Cite

View full text Add to dashboard Cite

Please scroll down for article-it is on subsequent pagesINFORMS is the largest professional society in the world for professionals in the fields of operations research, management science, and analytics. For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org Abstract Analysts in every field face the challenge of how to best use available data to estimate performance, quantify uncertainty, and predict the future. The data are almost never "just right," but rather scarce, excessive, corrupted, uncertain, and incomplete. External information derived from experiences, established "laws," and physical restrictions offer opportunities to remedy the situation and should be utilized. Applications in sustainable energy, natural resources, image reconstruction, financial planning, uncertainty quantification, and reliability engineering are rich with problems where decisions rely on data analysis under such circumstances. We address these problems within a framework that identifies a function that according to some criterion best represents the given data set and satisfies constraints derived from the data as well as external information. Epi-splines provide the linchpin that allows us to handle shape restrictions, information growth, and approximations.

show abstract

Statistical metrics for assessing the quality of wind power scenarios for stochastic unit commitment

et al. 2015

View full text Add to dashboard Cite

In power systems with high penetration of wind generation, probabilistic scenarios are generated for use in stochastic formulations of day-ahead unit commitment problems. To minimize the expected cost, the wind power scenarios should accurately represent the stochastic process for available wind power. We employ some statistical evaluation metrics to assess whether the scenario set possesses desirable properties that are expected to lead to a lower cost in stochastic unit commitment. A new mass transportation distance rank histogram is developed for assessing the reliability of unequally likely scenarios. Energy scores, rank histograms and Brier scores are applied to alternative sets of scenarios that are generated by two very different methods. The mass transportation distance rank histogram is best able to distinguish between sets of scenarios that are more or less calibrated according to their bias, variability and autocorrelation. ABSTRACTIn power systems with high penetration of wind generation, probabilistic scenarios are generated for use in stochastic formulations of day-ahead unit commitment problems. To minimize the expected cost, the wind power scenarios should accurately represent the stochastic process for available wind power. We employ some statistical evaluation metrics to assess whether the scenario set possesses desirable properties that are expected to lead to a lower cost in stochastic unit commitment. A new mass transportation distance (MTD) rank histogram is developed for assessing the reliability of unequally likely scenarios. Energy scores, rank histograms, and Brier scores are applied to alternative sets of scenarios that are generated by two very different methods. The MTD rank histogram is best able to distinguish between sets of scenarios that are more or less calibrated according to their bias, variability and autocorrelation.

show abstract

Multi-period forecasting and scenario generation with limited data

Cited by 35 publications

References 15 publications

Toward scalable stochastic unit commitment. Part 1: load scenario generation

Toward scalable stochastic unit commitment. Part 1: load scenario generation

From Data to Assessments and Decisions: Epi-Spline Technology

Statistical metrics for assessing the quality of wind power scenarios for stochastic unit commitment

Contact Info

Product

Resources

About