Comparison of three short-term load forecast models in Southern California

Zhang, Ning; Li, Zhiying; Zou, Xun; Quiring, Steven M.

doi:10.1016/j.energy.2019.116358

Cited by 53 publications

(23 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It should be noted that most statistical approaches are based on probabilistic forecasts, and the distribution of forecast values is helpful for risk management (Cabrera and Schulz, 2017). On the other hand, machine learning techniques have attracted attention in recent years, such as support vector machine (SVM); (Chen et al, 2017;Jiang et al, 2018;Yang et al, 2019); neural networks (He, 2017;Guo et al, 2018b;Kong et al, 2018;Bedi and Toshniwal, 2019;Wang et al, 2019), gradient boosting (Zhang et al, 2019), and hybrids of multiple forecasting techniques (Miswan et al, 2016;Liu et al, 2017;de Oliveira and Cyrino Oliveira, 2018;Haq and Ni, 2019). These techniques capture complex nonlinear structures; therefore, high forecast accuracies are expected.…”

Section: Introductionmentioning

confidence: 99%

Interpretable Modeling for Short- and Medium-Term Electricity Demand Forecasting

Hirose

2021

Front. Energy Res.

View full text Add to dashboard Cite

We consider the problem of short- and medium-term electricity demand forecasting by using past demand and daily weather forecast information. Conventionally, many researchers have directly applied regression analysis. However, interpreting the effect of weather on the demand is difficult with the existing methods. In this study, we build a statistical model that resolves this interpretation issue. A varying coefficient model with basis expansion is used to capture the nonlinear structure of the weather effect. This approach results in an interpretable model when the regression coefficients are nonnegative. To estimate the nonnegative regression coefficients, we employ nonnegative least squares. Three real data analyses show the practicality of our proposed statistical modeling. Two of them demonstrate good forecast accuracy and interpretability of our proposed method. In the third example, we investigate the effect of COVID-19 on electricity demand. The interpretation would help make strategies for energy-saving interventions and demand response.

show abstract

Section: Introductionmentioning

confidence: 99%

Interpretable Modeling for Short- and Medium-Term Electricity Demand Forecasting

Hirose

2021

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…Some studies combine fuzzy models and similar day methods for training [31,32,33]. Some studies use holidays as the input by binary encoding (i.e., 0 for a non-holiday, and 1 for a holiday), but efforts in manually collecting calendar data and confirming the building type are inevitable [34,35]. Due to the tedious efforts to collect calendar data, most past research focused on buildings in a single region, and no research has yet to propose a general method that can be widely applied across building types or countries.…”

Section: Introductionmentioning

confidence: 99%

Using Google Trends as a proxy for occupant behavior to predict building energy consumption

Fu,

Miller

2021

Preprint

View full text Add to dashboard Cite

In recent years, the availability of larger amounts of energy data and advanced machine learning algorithms has created a surge in building energy prediction research. However, one of the variables in energy prediction models, occupant behavior, is crucial for prediction performance but hard-to-measure or time-consuming to collect from each building. This study proposes an approach that utilizes the search volume of topics (e.g., education or Microsoft Excel) on the Google Trends platform as a proxy of occupant behavior and use of buildings. Linear correlations were first examined to explore the relationship between energy meter data and Google Trends search terms to infer building occupancy. Prediction errors before and after the inclusion of the trends of these terms were compared and analyzed based on the ASHRAE Great Energy Predictor III (GEPIII) competition dataset. The results show that highly correlated Google Trends data can effectively reduce the overall RMSLE error for a subset of the buildings to the level of the GEPIII competition's top five winning teams' performance. In particular, the RMSLE error reduction during public holidays and days with site-specific schedules are respectively reduced by 20-30% and 2-5%. These results show the potential of using Google Trends to improve energy prediction for a portion of the building stock by automatically identifying site-specific and holiday schedules.

show abstract

“…Almost all models include temperature from one or more locations linearized either piecewise or through another function like a 3rd degree polynomial. Several examples can be found in [6,8,[22][23][24][25][26]. Long term trends are usually captured by linear or quadratic polynomials of time [10], but also can be modeled by a moving function of previous loads [6] and it can even ignored for shorter period of time.…”

Section: Introductionmentioning

confidence: 99%

Use of Available Daylight to Improve Short-Term Load Forecasting Accuracy

et al. 2020

View full text Add to dashboard Cite

This paper introduces a new methodology to include daylight information in short-term load forecasting (STLF) models. The relation between daylight and power consumption is obvious due to the use of electricity in lighting in general. Nevertheless, very few STLF systems include this variable as an input. In addition, an analysis of one of the current STLF models at the Spanish Transmission System Operator (TSO), shows two humps in its error profile, occurring at sunrise and sunset times. The new methodology includes properly treated daylight information in STLF models in order to reduce the forecasting error during sunrise and sunset, especially when daylight savings time (DST) one-hour time shifts occur. This paper describes the raw information and the linearization method needed. The forecasting model used as the benchmark is currently used at the TSO’s headquarters and it uses both autoregressive (AR) and neural network (NN) components. The method has been designed with data from the Spanish electric system from 2011 to 2017 and tested over 2018 data. The results include a justification to use the proposed linearization over other techniques as well as a thorough analysis of the forecast results yielding an error reduction in sunset hours from 1.56% to 1.38% for the AR model and from 1.37% to 1.30% for the combined forecast. In addition, during the weeks in which DST shifts are implemented, sunset error drops from 2.53% to 2.09%.

show abstract

Comparison of three short-term load forecast models in Southern California

Cited by 53 publications

References 41 publications

Interpretable Modeling for Short- and Medium-Term Electricity Demand Forecasting

Interpretable Modeling for Short- and Medium-Term Electricity Demand Forecasting

Using Google Trends as a proxy for occupant behavior to predict building energy consumption

Use of Available Daylight to Improve Short-Term Load Forecasting Accuracy

Contact Info

Product

Resources

About