Building Electricity Demand Forecasting
Download as PPTX, PDF•0 likes•1,430 views
This document describes a proposed electricity demand forecasting pipeline for predicting power consumption at the IIIT-Delhi campus. It discusses collecting high-resolution smart meter data, preprocessing the data, and using various forecasting models like ARIMA, ANN, and hybrid models. Parameters like granularity, forecast horizon, and similar days selection are explored. The pipeline is implemented in R and a Shiny app is developed for data visualization and parameter selection. Time-series clustering is also discussed to group loads and assign the best forecasting model.
Building Electricity Demand Forecasting
- 1. Building Electricity Demand Forecasting SHUBHAM SAINI, PANDARASAMY ARJUNAN, AMARJEET SINGH As part of the work done at Mobile and Ubiquitous Computing Group
- 2. OVERVIEW The IIIT – Delhi campus has more than 200 smart meters installed, collecting around 10 electrical parameters every 30 seconds. Important to calculate an accurate baseline, and monitor any deviations from it. A forecasting pipeline is proposed for predicting the power consumption of an electric load at any given point of time.
- 3. Motivation Energy Consumption Increasing Worldwide India – Energy Forecasting has important role in formulation of effective energy policies Electricity consumption analysis useful for monitoring environmental issues
- 4. FORECASTING MODELS Auto-Regressive Integrated Moving Average (ARIMA) Artificial Neural Networks (ANN) Hybrid ARIMA+ANN EnerNOC
- 5. ARIMA (p,d,q)(P,D,Q) (p,P) - number of lagged variables (d,D) - difference necessary to make the time series stationary (q,Q) moving average over the number of last observations. Where yt and Et are actual value and random error at time t
- 6. Artificial Neural Networks Popular for flexible non-linear modeling Single hidden layer feed-forward network Where wj and wi,j are model model parameters called connection weights, p is the number of input nodes and q is the number of hidden nodes.
- 7. Hybrid ARIMA+ANN Power consumption composed of linear and non- linear structure Yt = Lt + Nt ARIMA able to model linear component Lt Residuals modeled by ANN et = Yt - YFt Final fitted value: YFt = LFt + NFt
- 8. EnerNOC Based on averaging the load on X days for each interval D-3 12-1am 1-2am 2-3am 3-4am 4-5am D-2 12-1am 1-2am 2-3am 3-4am 4-5am D-1 12-1am 1-2am 2-3am 3-4am 4-5am Event Day 12-1am 1-2am 2-3am 3-4am 4-5am
- 9. Prediction Pipeline Multiple models can be learned by using different sub-models at each of these stages.
- 10. Initial Parameters - Granularity Very high resolution data available, sampled every 30 seconds Too small and too large time intervals detrimental to a model's performance Experimented with 1Hour, 30Minutes, 15 Minutes
- 12. Initial Parameters – Forecast Horizon Forecasting Horizon implies the number of data points a model forecasts into the future. Days maybe be divided into working/non-working hours, day/night hours, peak/off-peak hours.
- 14. SELECTION OF SIMILAR (Y) DAYS CRITERIA: Previous Business Days Previous Same Days Lookback Window 4,7,10
- 15. 7 similar days 14 similar days
- 16. Sub-sampling (X) Days Criteria High X Days Makes sense for demand-response Excluding Highest and Lowest Days anomalies could be either due to load failure, holiday, unpredicted occupancy etc
- 17. X:Y = 8:10 X:Y = 6:10
- 18. Adjustments – ARIMA+ANN Training data used to forecast future values includes an additional 2-4 hours of data from the event day. For example, in order to forecast consumption on the event day for 12PM - 5PM, we use 10AM - 5PM data on the X similar days, as well as 10AM - 12PM data on the event day. This additional data more accurately reflect load conditions on the event day.
- 20. Adjustments - EnerNOC To adjust the forecasted value of a time interval, for example 12PM - 1 PM, adjustments are done at 11AM Mean of difference between actual values and the forecasted values between 8AM - 11AM is added(subtracted) to(from) the 12PM - 1PM forecasted value. Event Day data not always available !!
- 22. Results Brute-force approach to find optimal parameters Over 700 different combinations of parameters tested Varying Parameters: 1. No. of similar days - 4, 7, 10 2. Similarity Criteria - Previous Business Days, Previous Same Days 3. Sub-sampling: High X of Y 4. X:Y Ratio - 6:10, 8:10 5. Models - Hybrid ARIMA+ANN, EnerNOC, Adjusted EnerNOC 6. Time Duration - 12AM - 12AM, 12AM - 7AM, 7AM - 12PM, 12PM - 5PM 7. Dates - 13-March-2014, 11-March-2014, 5-March-2014, 3-March-2014, 28-February-2014
- 23. Results (Contd.) Load #1: Academic Building - Floor Total - First Floor
- 24. Sample Result Load #1: Academic Building - Floor Total - First Floor Number of Similar Days (Y) – 7 X : Y Ratio – 0.8 Similarity Criteria - Previous Same Days Time Duration - 12AM – 7AM Model - Adjusted EnerNOC
- 25. Implementation Developed using the R language for statistical computing version 3.0(RStudio IDE) Reasons for choosing R over other statistical computing languages like Matlab are: 1. Free and Open-Source 2. Graphics and Data Visualization 3. Flexible statistical analysis toolkit 4. Powerful, cutting-edge analytics 5. Robust, vibrant community
- 26. UI Design and Layout GUI for simple data visualization using Shiny web framework v0.98 Tab layout with a sidebar Sidebar contains options to set the forecasting parameters Main window - training data, and output of various forecasting models
- 28. Time-Series clustering (In Progress) Global features extracted from the time series through statistical operations trend seasonality periodicity serial correlation skew, kurtosis chaos nonlinearity self-similarity
- 29. Time-Series clustering (In Progress) Clustering – K-Means or Heirarchical Using global characteristics, group all available streams into optimal number of clusters For each cluster, find optimal forecasting model (through the prediction pipeline) For any new stream – assign the stream to one of the clusters and apply the optimal forecasting model
- 30. Questions ???