Understanding the Effect of Baseline Modeling Implementation Choices on Analysis of Demand Response Performance

Addy, Nathan; Mathieu, Johanna L.; Kiliccote, Sila; Callaway, Duncan S.

doi:10.1115/imece2012-86973

Cited by 11 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using 15-min interval whole-building electric demand data from a critical peak pricing (CPP) program in CA, we investigate the effect of five baseline model implementation choices: (1) source of outdoor air temperature data, (2) data resolution, (3) method for determining when buildings are occupied, (4) method for demand/temperature data alignment, and (5) method for filtering out bad data. This paper is an extension of our preliminary work on this topic [15].…”

Section: Introductionmentioning

confidence: 89%

“…For example, if the minimum observed temperature is 5 C and the maximum is 35 C, then each bin is 5 C wide with B ¼ (10,15,20,25,30). Then, we compute the temperature components associated with each temperature, for example, for T ¼ 18 C, we find T c,1 ¼ 10 C, T c,2 ¼ 5 C, T c,3 ¼ 3 C, and the remaining temperature components are 0 C. Note that P j T c;j ðt i Þ ¼ Tðt i Þ.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

How Baseline Model Implementation Choices Affect Demand Response Assessments

Addy¹,

Kiliccote

Callaway

et al. 2014

Journal of Solar Energy Engineering

Self Cite

View full text Add to dashboard Cite

The performance of buildings participating in demand response (DR) programs is usually evaluated with baseline models, which predict what electric demand would have been if a DR event had not been called. Different baseline models produce different results. Moreover, modelers implementing the same baseline model often make different model implementation choices producing different results. Using real data from a DR program in CA and a regression-based baseline model, which relates building demand to time of week, outdoor air temperature, and building operational mode, we analyze the effect of model implementation choices on DR shed estimates. Results indicate strong sensitivities to the outdoor air temperature data source and bad data filtration methods, with standard deviations of differences in shed estimates of %20-30 kW, and weaker sensitivities to demand/temperature data resolution, data alignment, and methods for determining when buildings are occupied, with standard deviations of differences in shed estimates of %2-5 kW.

show abstract

Section: Introductionmentioning

confidence: 89%

Section: Methodsmentioning

confidence: 99%

How Baseline Model Implementation Choices Affect Demand Response Assessments

Addy¹,

Kiliccote

Callaway

et al. 2014

Journal of Solar Energy Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…In selecting the best algorithm for the study, it is critical to develop a baseline of performance to compare a no-effect hypothesis model to alternatives that are more complex. Other researchers from different fields have used other terms, such as "naïve model" and "null model" (Addy et al, 2012;Schwab and Starbuck, 2013). In this work, four baseline models have been developed to represent one for each region of analysis and used as reference or control for this study with features or input parameters that are not selected through EDA.…”

Section: The Baseline Modelmentioning

confidence: 99%

Exploratory Data Analysis and Artificial Neural Network for Prediction of Leptospirosis Occurrence in Seremban, Malaysia Based on Meteorological Data

et al. 2020

View full text Add to dashboard Cite

Leptospirosis outbreaks in various parts of the world have been linked to changes in the weather. Furthermore, the effects have been shown to occur at different lags of up to 10 months, affecting the performance of simulation models that predict leptospirosis occurrence. In Malaysia, the link between different weather parameters, at different time lags, has yet to be established despite an increasing number of cases in recent years. In this study, a combination of data mining and machine learning is used to analyze, capture, and predict the relation between leptospirosis occurrence and temperature, rainfall, and relative humidity using the Seremban district in Malaysia as a case study. First, the optimal time lags for rainfall were determined using graphical exploratory data analysis (EDA) while non-graphical EDA was used for temperature. Then, an artificial neural network (ANN) model is developed to classify the combination of selected features into disease occurrence and non-occurrence using back-propagation training, optimizing the number of hidden layers and hidden nodes. The success is measured using accuracy, sensitivity, and specificity of each model. EDA has shown that leptospirosis occurrence in Seremban is highly correlated with weekly average temperature at lag 16 weeks and weekly rainfall amount at lag 12-20 weeks. Using these selected features, the ANN model achieved the highest accuracy, sensitivity, and specificity at 84.00, 86.44, and 79.33%, respectively. Overall, the EDA approach has increased the accuracy of the predictive model by 13.30-31.26% from the baseline models.

show abstract

“…For more details on the datasets, the readers are referred to [34]. Weather data: We obtained curated weather data from NOAA [35], [36] We used hourly temperature observations, which were interpolated to 15-min values. Schedule Data: It was obtained for the campus dataset comprised of information on working days, holidays, and semester durations (for campus dataset).…”

Section: A Dataset Descriptionmentioning

confidence: 99%

Prediction models for dynamic demand response: Requirements, challenges, and insights

Aman

Frîncu

Chelmis

et al. 2015

2015 IEEE International Conference on Smart Grid Communications (SmartGridComm)

View full text Add to dashboard Cite

As Smart Grids move closer to dynamic curtailment programs, Demand Response (DR) events will become necessary not only on fixed time intervals and weekdays predetermined by static policies, but also during changing decision periods and weekends to react to real-time demand signals. Unique challenges arise in this context vis-a-vis demand prediction and curtailment estimation and the transformation of such tasks into an automated, efficient dynamic demand response (D 2 R) process. While existing work has concentrated on increasing the accuracy of prediction models for DR, there is a lack of studies for prediction models for D 2 R, which we address in this paper. Our first contribution is the formal definition of D 2 R, and the description of its challenges and requirements. Our second contribution is a feasibility analysis of very-short-term prediction of electricity consumption for D 2 R over a diverse, large-scale dataset that includes both small residential customers and large buildings. Our third, and major contribution is a set of insights into the predictability of electricity consumption in the context of D 2 R. Specifically, we focus on prediction models that can operate at a very small data granularity (here 15-min intervals), for both weekdays and weekends -all conditions that characterize scenarios for D 2 R. We find that short-term time series and simple averaging models used by Independent Service Operators and utilities achieve superior prediction accuracy. We also observe that workdays are more predictable than weekends and holiday. Also, smaller customers have large variation in consumption and are less predictable than larger buildings. Key implications of our findings are that better models are required for small customers and for non-workdays, both of which are critical for D 2 R. Also, prediction models require just few days' worth of data indicating that small amounts of historical training data can be used to make reliable predictions, simplifying the complexity of big data challenge associated with D 2 R.

show abstract

Understanding the Effect of Baseline Modeling Implementation Choices on Analysis of Demand Response Performance

Cited by 11 publications

References 11 publications

How Baseline Model Implementation Choices Affect Demand Response Assessments

How Baseline Model Implementation Choices Affect Demand Response Assessments

Exploratory Data Analysis and Artificial Neural Network for Prediction of Leptospirosis Occurrence in Seremban, Malaysia Based on Meteorological Data

Prediction models for dynamic demand response: Requirements, challenges, and insights

Contact Info

Product

Resources

About