Building-level power demand forecasting framework using building specific inputs: Development and applications

Touretzky, Cara R.; Patil, Rahul

doi:10.1016/j.apenergy.2015.03.025

Cited by 25 publications

(7 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The STLF process heavily relies on the weather information and ambient environment.When the parameters are estimated, the weather information is extrapolated to forecast the load. Much research [4], [20]has looked at the most suitable features for load forecast problems. They try to explain the causality of the electric load consumption.…”

Section: Training Data Selectionmentioning

confidence: 99%

“…Therefore, we use an ANNas the meta-learner, considering correlation between the meta-features and nonlinear patterns brought by the complexity and heterogeneities of different building scenarios (noises within meta-features)might impair the modeling power of the learner. The parameter settings of the meta-learner ANN areas follows: the hidden layer size is tuned within the range of [10,20], and the transfer functions are selected between radial basis and log sigmoid. Note that the proposed meta-featuresare tentatively selected in hoping that they could effectively represent the dataset.…”

Section: Meta-learnermentioning

confidence: 99%

“…Variousdata-driven methods have been studied and implemented for building load forecasting including 1) statistical methods such as autoregressive, moving average, exponential smoothing [15], state space [16], [17], polynomial regression [18], and 2) machine learning methods such as neural networks [19] and support vector regression [8], [20].Statistical regression models simply build the correlation between the energy consumption and the simplified influential features such as weather parameters. These empirical models are developed from historical performance data to train the models.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Short-term building energy model recommendation system: A meta-learning approach

Cui

et al. 2016

Applied Energy

View full text Add to dashboard Cite

High-fidelity and computationally efficient energy forecasting models for building systems are needed to ensure optimal automatic operation, reduce energy consumption, and improve the building's resilience capability to power disturbances. Various models have been developed to forecast building energy consumption. However, given buildings have different characteristics and operating conditions, model performance varies. Existing research has mainly taken a trial-and-error approach by developing multiple models and identifying the best performer for a specific building, or presumedone universal model form which is applied on different building cases. To the best of our knowledge, there does not exist a generalized system framework which can recommend appropriate models to forecast the building energy profiles based on building characteristics. To bridge this research gap, we propose a meta-learning based framework, termed Building Energy Model Recommendation System (BEMR). Based on the building's physical features as well as statistical and time series meta-features extracted from the operational data and energy consumption data, BEMR is able to identify the most appropriate load forecasting model for each unique building.Three sets of experiments on 48 test buildings and one real building are conducted. The first experiment is to test the accuracy of BEMR when the training data and testing data cover the same condition. BEMR correctly identified the best modelon 90% of the buildings. The second experiment is to test the robustness of the BEMR when the testing data is only partially covered by the training data.BEMR correctly identified the best model on 83% of the buildings. The third experiment uses a real building case to validate the proposed framework and the result shows promising applicability and extensibility. The experimental results show that BEMRis capable of adapting to a wide variety of

show abstract

Section: Training Data Selectionmentioning

confidence: 99%

Section: Meta-learnermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Short-term building energy model recommendation system: A meta-learning approach

Cui

et al. 2016

Applied Energy

View full text Add to dashboard Cite

show abstract

“…The black-box algorithms are further categorized, such as linear autoregressive algorithms (ARAs), as shown in [8]. Touretzky and Patil proposed an ARA to predict the energy required for energy management in the building sector, specifically demand response and supervisory control [9]. Ferracuti et al assessed various complex algorithms to precisely estimate hourly energy usage requirements for district energy management [10].…”

Section: Introductionmentioning

confidence: 99%

A Comparative Analysis of Machine Learning Approaches for Short-/Long-Term Electricity Load Forecasting in Cyprus

Solyalı

2020

Sustainability

View full text Add to dashboard Cite

Estimating the electricity load is a crucial task in the planning of power generation systems and the efficient operation and sustainable growth of modern electricity supply networks. Especially with the advent of smart grids, the need for fairly precise and highly reliable estimation of electricity load is greater than ever. It is a challenging task to estimate the electricity load with high precision. Many energy demand management methods are used to estimate future energy demands correctly. Machine learning methods are well adapted to the nature of the electrical load, as they can model complicated nonlinear connections through a learning process containing historical data patterns. Many scientists have used machine learning (ML) to anticipate failure before it occurs as well as predict the outcome. ML is an artificial intelligence (AI) subdomain that involves studying and developing mathematical algorithms to understand data or obtain data directly without relying on a prearranged model algorithm. ML is applied in all industries. In this paper, machine learning strategies including artificial neural network (ANN), multiple linear regression (MLR), adaptive neuro-fuzzy inference system (ANFIS), and support vector machine (SVM) were used to estimate electricity demand and propose criteria for power generation in Cyprus. The simulations were adapted to real historical data explaining the electricity usage in 2016 and 2107 with long-term and short-term analysis. It was observed that electricity load is a result of temperature, humidity, solar irradiation, population, gross national income (GNI) per capita, and the electricity price per kilowatt-hour, which provide input parameters for the ML algorithms. Using electricity load data from Cyprus, the performance of the ML algorithms was thoroughly evaluated. The results of long-term and short-term studies show that SVM and ANN are comparatively superior to other ML methods, providing more reliable and precise outcomes in terms of fewer estimation errors for Cyprus’s time series forecasting criteria for power generation.

show abstract

“…Such methods are often 3 38 embedded in a more complicated load model, as in [16], because the methods 39 do not sufficiently capture complex load features if used alone. Regarding 40 smaller scale power systems including residential and commercial areas, as 41 well as the building level, hybrids of multiple regression, time series, and ML 42 based methods are often used [28,29,30]. 43 Among statistical methods, time series and regression methods are widely 44 used to build short-term load models.…”

mentioning

confidence: 99%

Day-ahead hourly electricity load modeling by functional regression

Feng

Ryan

2016

Applied Energy

View full text Add to dashboard Cite

Short-term load forecasting is important for power system generation planning and operation. For unit commitment and dispatch processes to incorporate uncertainty, a short-term load model must not only provide accurate load predictions but also enable the generation of reasonable probabilistic scenarios or uncertainty sets. This paper proposes a temporal and weather conditional epi-splines based load model (TWE) using functional approximation. TWE models the dependence of load on time and weather separately by functional approximation using epi-splines, conditional on season and area, in each segment of similar weather days. Load data are transformed from various day types to a specified reference day type among similar weather days in the same season and area, in order to enrich the data for capturing the non-weather dependent load pattern. In an instance derived from an Independent System Operator in the U.S., TWE not only provides accurate hourly load prediction and narrow bands of prediction errors, but also yields serial correlations among forecast hourly load values within a day that are similar to those of actual hourly load. KeywordsShort-term load model, Forecasting, Day-ahead scenario, Epi-splines Disciplines Industrial Engineering | Systems EngineeringComments NOTICE: this is the author's version of a work that was accepted for publication in Applied Energy. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Applied Energy, v.170, 15 May (2016) AbstractShort-term load forecasting is important for power system generation planning and operation. For unit commitment and dispatch processes to incorporate uncertainty, a short-term load model must not only provide accurate load predictions but also enable the generation of reasonable probabilistic scenarios or uncertainty sets. This paper proposes a temporal and weather conditional epi-splines based load model (TWE) using functional approximation. TWE models the dependence of load on time and weather separately by functional approximation using epi-splines, conditional on season and area, in each segment of similar weather days. Load data are transformed from various day types to a specified reference day type among similar weather days in the same season and area, in order to enrich the data for capturing the non-weather dependent load pattern. In an instance derived from an Independent System Operator in the U.S., TWE not only provides accurate hourly load prediction and narrow bands of prediction errors, but also yields serial correlations among forecast hourly load values within a day that are similar to those of actual hourly load.

show abstract

Building-level power demand forecasting framework using building specific inputs: Development and applications

Cited by 25 publications

References 32 publications

Short-term building energy model recommendation system: A meta-learning approach

Short-term building energy model recommendation system: A meta-learning approach

A Comparative Analysis of Machine Learning Approaches for Short-/Long-Term Electricity Load Forecasting in Cyprus

Day-ahead hourly electricity load modeling by functional regression

Contact Info

Product

Resources

About