SVD-based visualisation and approximation for time series data in smart energy systems

Khoshrou, Abdolrahman; Dorsman, A.B.; Pauwels, Eric J.

doi:10.1109/isgteurope.2017.8260303

Cited by 4 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, the right panel in Fig. 1, illustrates the evolution of daily load profiles in the year 2010; this figure was obtained by recasting the time series into a × 24 365 matrix, where every column contains 24 hourly values for each daily profile [26]. As expected, in spring and fall, where the temperature is moderate, electricity demand tends to be lower than any other time of the year (winter and summer times).…”

Section: Datamentioning

confidence: 89%

“…There are some conspicuous diurnal patterns in the data, reflecting typical patterns for daily or weekly human activities. Furthermore, the overall structure of the data is affected by a combination of those fast diurnal patterns superimposed on slower seasonal variations [26]. We herein consider two Gradient Boosting based methods to predict the day-ahead load prognoses.…”

Section: Our Proposed Forecasting Modelsmentioning

confidence: 99%

“…Model II To reflect the lagging effect of temperature on load changes, in the second model, we add the daily dynamics of the temperature and load profiles (1st and 2nd derivatives) [26,35]. For a given hour slot h the corresponding first derivative of the variable z L T { , } can be obtained by…”

Section: Our Proposed Forecasting Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Short-term scenario-based probabilistic load forecasting: A data-driven approach

Khoshrou

Pauwels

2019

Applied Energy

Self Cite

View full text Add to dashboard Cite

Scenario-based probabilistic forecasting models have been explored extensively in the literature in recent years. The performance of such models evidently depends to a large extent on how different input (temperature) scenarios are being generated. This paper proposes a generic framework for probabilistic load forecasting using an ensemble of regression trees. A major distinction of the current work is in using matrices as an alternative representation for quasi-periodic time series data. The singular value decomposition (SVD) technique is then used herein to generate temperature scenarios in a robust and timely manner. The strength of our proposed method lies in its simplicity and robustness, in terms of the training window size, with no need for subsetting or thresholding to generate temperature scenarios. Furthermore, to systematically account for the non-linear interactions between different variables, a new set of features is defined: the first and second derivatives of the predictors. The empirical case studies performed on the data from the load forecasting track of the Global Energy Forecasting Competition 2014 (GEFCom2014-L) show that the proposed method outperforms the top two scenario-based models with a similar setup .

show abstract

Section: Datamentioning

confidence: 89%

Section: Our Proposed Forecasting Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Short-term scenario-based probabilistic load forecasting: A data-driven approach

Khoshrou

Pauwels

2019

Applied Energy

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the case at hand, it is clear that prices would show daily patterns that might change relatively slowly over the year. To apply SVD, we rewrite the time series as a matrix where each column records the 24 hourly values for a particular day [10]. In this way, a time series representing a typical year is recast as a matrix A of size 24 × 365.…”

Section: B Extracting Underlying Trendsmentioning

confidence: 99%

Quantifying Volatility Reduction in German Day-ahead Spot Market in the Period 2006 through 2016

Khoshrou

Pauwels

2018

2018 IEEE Power &Amp; Energy Society General Meeting (PESGM)

Self Cite

View full text Add to dashboard Cite

In Europe, Germany is taking the lead in the switch from the conventional to renewable energy. This poses new challenges as wind and solar energy are fundamentally intermittent, weather-dependent and less predictable. It is therefore of considerable interest to investigate the evolution of price volatility in this post-transition era. There are a number of reasons, however, that makes the practical studies difficult. For instance, EPEX prices can be zero or negative. Consequently, the standard approach in financial time series analysis to switch to logarithmic measures is inapplicable. Furthermore, in contrast to the stock market prices which are only available for trading days, EPEX prices cover the whole year, including weekends and holidays. Accordingly, there is a lot of underlying variability in the data which has nothing to do with volatility, but simply reflects diurnal activity patterns. An important distinction of the present work is the application of matrix decomposition techniques, namely the singular value decomposition (SVD), for defining an alternative notion of volatility. This approach is systematically more robust toward outliers and also the diurnal patterns. Our observations show that the day-ahead market is becoming less volatile in recent years.

show abstract

“…Inspired by the work in Ref. [6], we herein consider matrices as an alternative representation of the electricity market data where the time series demonstrates periodic patterns. We accordingly define a new, yet easy-to-quantify, notion of volatility using a popular and numerically stable matrix decomposition technique, namely the singular value decomposition (SVD).…”

Section: Introductionmentioning

confidence: 99%

The evolution of electricity price on the German day-ahead market before and after the energy switch

2019

Self Cite

View full text Add to dashboard Cite

a b s t r a c tGermany is a forerunner in developing renewable energies. It is therefore of considerable interest to investigate the impact of switch to renewables on the market during transition era. The aim of this study is in two parts: 1) Investigating the volatility; and 2) Conducting a descriptive study on the evolution of daily profiles and emergence of non-positive prices. In terms of volatility quantification, the following characteristics of EPEX prices should be recognized: 1) Covering the whole year (24/7); 2) Taking nonpositive values; 3) Depending on calendar information; and 4) Changing according to demand and supply availability. We, therefore, propose a robust and generic approach to account for diurnal or seasonal patterns of human activities in volatility analysis. An important distinction of our work is in introducing an alternative representation (as matrices) for quasi-periodic price data. We, herein, propose a new notion of volatility using a matrix decomposition technique, namely the singular value decomposition (SVD). Our observations indicate price volatility reduction, in the recent years. The second part of this article provides evidences of effect of renewables on daily price profiles e emergence of nonpositive prices and also shifts of peak price values to hours where solar is less available.

show abstract

SVD-based visualisation and approximation for time series data in smart energy systems

Cited by 4 publications

References 7 publications

Short-term scenario-based probabilistic load forecasting: A data-driven approach

Short-term scenario-based probabilistic load forecasting: A data-driven approach

Quantifying Volatility Reduction in German Day-ahead Spot Market in the Period 2006 through 2016

The evolution of electricity price on the German day-ahead market before and after the energy switch

Contact Info

Product

Resources

About