Applications of Data Science to Mini-Grid Smart Meter and Survey Data 3 rd Africa Smart Grid Forum - Kigali, Rwanda October 4, 2018 Nathan Williams nwilliams@cmu.edu Department of Engineering and Public Policy
Status of Electrification in East Africa Total Population 166 million Rural Population 124 million Regional Electrification 37% Urban Electrification 62% Rural Electrification 28% Pop. without Access 105 million Source: World Energy Outlook 2017 Source: NASA/AfDB
Status of Electrification in East Africa Total Population 166 million Rural Population 124 million Regional Electrification 37% Urban Electrification 62% Rural Electrification 28% Pop. without Access 105 million Source: World Energy Outlook 2017 Source: NASA/AfDB
Mini-grids in East Africa Diverse approaches AC vs. DC Revenue models System size Generation technology
Smart Meters in Mini-Grids Lots of experimentation and innovation in the sector Rich datasets are being collected with smart meters Data available in near real-time through mobile network
Payment Size/Freq. Meter Data Collected Hourly electricity consumption, voltage, current Prepaid energy transactions Hourly Load Profiles Load Growth Weekly Seasonality
Customer Survey Data Customer application surveys completed prior to construction Data collected on: Type of customer (home, business, public premises…) Household characteristics (number of members, rooms, employment status, income…) Business characteristics (business areas…) Building characteristics (type of material, owned/rented…) Energy use (sources, uses, appliances owned and planned…) Modes of transportation
Demographic Summary
Current Data Projects • Predictive Modeling for Unelectrified Communities • Customer Load Profile Segmentation • Mini-grid Load Forecasting
Predictive Modeling of Electricity Consumption for Unelectrified Communities - Overview Mini-grid revenue depends on electricity consumption Can’t be measure, never had access! Traditional approach, take inventory of appliances to be used and use patterns DOESN’T WORK! How would they know? With growing set of data, can we make data driven predictions? What data are useful for predicting electricity consumption?
Predictive Modeling for Unelectrified Communities - Method Apply regression models to relate customer survey data to subsequent consumption Models fit: Ordinary Linear Regression Ridge Regression LASSO Regression Elastic Net Regression Principal Components Regression Random Forest Regression
Predictive Modeling for Unelectrified Communities - Results Ridge & LASSO make best predictions Individual predictions are not great but better than benchmark Low correlation between customers means site level predictions are much better Ridge Regression Median baseline site level error: 50% Median site level error with Ridge: 22% Median site level error with LASSO: 23% Median site level error with Random Forest: 30%
Predictive Modeling for Unelectrified Communities - Results Ridge & LASSO make best predictions Individual predictions are not great but better than benchmark Low correlation between customers means site level predictions are much better Random Forest Median baseline site level error: 50% Median site level error with Ridge: 22% Median site level error with LASSO: 23% Median site level error with Random Forest: 30%
Predictive Modeling for Unelectrified Communities - Results LASSO Selected Variables Coeff. Sign Diesel Use: Heat + What kind of variables are useful in Transport: Other Low Freq. + prediction? Transport: Boat + Gasoline Use: Electricity Gen. + Customer class (Home, Business, • Business Type: Bar + Home/Business, Public Premises) Building Type: Wood + Nature of business • Pre-MG Appliances: Other Low Freq. + Pre-MG Appliances: High Watt TV + Employment status • Pre-MG Appliances: Light Bulb + Current source of energy and uses • Pre-MG Appliances: Low Watt TV + Current mode of transport Business Type: Other + • Business Connection + Appliances owned prior to mini-grid • Pre-MG Appliances: Other + connection Firewood Use: Cooking - Building construction materials • Resp. Employment: Self-Employed Ag. - Transport: Bicycle - Energy Source: Firewood - Business Type: Restaurant - Home Connection -
Predictive Modeling for Unelectrified Communities - Results What kind of variables are useful in prediction? Customer class (Home, Business, • Home/Business, Public Premises) Nature of business • Employment status • Current source of energy and uses • Random Forest Importance Current mode of transport • Appliances owned prior to mini-grid • connection Building construction materials •
Customer Load Profile Segmentation - Overview Beyond aggregate consumption, what do daily consumption patterns looks like? Are there groups of typical load profile class among customers? If so, what kind of customers fall into these classes?
Customer Load Profile Segmentation - Method Compute mean daily load profiles for each customer Run k- means cluster analysis to find ‘typical’ load profiles Group customers by aggregate consumption level What customers fall into consumption/profile classes? clusters
Customer Load Profile Segmentation - Results Five load profile classes Cluster Description 0 Large evening peak with a morning peak 1 Large evening peak with small morning bump 2 Large daytime use with no evening peak 3 Growing use during day with evening peak 4 Evening peak with continued night time use
Customer Load Profile Segmentation - Results
Mini-grid Load Forecasting - Overview Electricity consumed by mini-grid customers is both seasonal and expected to grow over time Once a customer is connected, how well can we forecast consumption in the future? Work by Fred Otieno, now at IBM Research Africa, Nairobi
Mini-grid Load Forecasting - Method Focus on time scales of weeks due to data limitations Models attempted: Persistence & Unconditional benchmark First order autoregressive Autoregressive Model with trend and seasonality Exponential Smoothing Test performance using out-of-sample forecasts measured by NRMSE Assess how far into the future one might reliably achieve accurate forecasts
Mini-grid Load Forecasting - Results
Mini-grid Load Forecasting - Results Intraweek seasonality is significant at certain sites, not intraannual Low correlation between customers Trend is not significant Exponential smoothing model forecasts up to 4 months with NRMSE of 55 – 99% relative to unconditional benchmark
On-Going Work Classification model for load profiles based on customer surveys Further investigation into prepaid transaction data (trends, seasonality, segmentation) Digging deeper on forecasting Impact of interventions (tariff structure changes, price reductions, short term promotional pricing, appliance finance programs)
Carnegie Mellon University Africa Two masters programs in Information Technology and Electrical & Computer • Engineering Energy concentration for both programs • First class graduated in 2014 • Currently 129 students from 15 countries • What do graduates do? • Information Technology • Financial Services • Public Sector • PhD studies • Education • Energy •
Recommend
More recommend
