What is meter and portfolio demand forecasting?
Meter demand forecasting predicts individual and aggregated meter-level load for a utility’s specific service territory. These custom forecasts provide far more accuracy for load-serving entities than a grid-level forecast, as they are built from the bottom up to account for each utility’s specific geography and meter types. Read about how Amperon is helping Marin Clean Energy forecast half a million meters and manage unique microclimates.
Portfolio demand forecasting predicts load for a Retail Energy Provider’s book of business. Read about how Amperon is helping Shipley Energy grow its retail electricity business through more accurate load forecasting. Accurate load forecasting is essential for REPs and CCAs to optimize bids, reduce procurement costs and load risk, and maintain profitability on thin margins.
Meter demand short-term forecast
(Meter STF)
(Meter STF)
Individual and aggregated meter-level demand forecasting from bal-day through 14 days out. Customized for each utility’s service territory and meter types. Market-leading weather forecasts from 4+ weather vendors dynamically ensembled and translated into individual and aggregate meter-level load forecasts.
50M+
Utility meters forecasted
Meter demand mid-term forecast
(Meter MTF)
(Meter MTF)
Weather-informed sub-seasonal and seasonal meter demand forecasting to give utilities advance notice of extreme weather and load events heading for their service territory. Informs term trading, forward hedging, and the timing of planned outages. Supports risk and regulatory reporting with the industry’s first weather-informed mid-term meter demand modeling.
51
ECMWF ensemble members
Meter demand long-term forecast
(Meter LTF)
(Meter LTF)
Long-term meter-level forecasting with up to three variations for scenario analysis (zero meter growth, current obligation, and meter growth adjusted). Forecast scenarios are combined with climatological data to produce thousands of simulations, grouped into four defaults (sim-mild, sim-neutral, sim-extreme, and 15-year normal).
20GW+
Utility load forecasted
Portfolio demand short-term forecast
(Portfolio STF)
(Portfolio STF)
Individual and aggregated meter-level demand forecasting from bal-day through 14 days out. Customized for each retailer’s book of business or CCA’s service territory. Market-leading weather forecasts from 4+ weather vendors dynamically ensembled and translated into individual and aggregate meter-level load forecasts.
2.5M
Retail meters forecasted
Portfolio demand long-term forecast
(Portfolio LTF)
(Portfolio LTF)
Long-term meter-level forecasting with up to three variations for scenario analysis (zero meter growth, current obligation, and meter growth adjusted). Forecast scenarios are combined with climatological data to produce thousands of simulations, grouped into four defaults (sim-mild, sim-neutral, sim-extreme, and 15-year normal).
1000s
Of long-term load simulations
Portfolio scheduling
Scheduling saves customers 25+ hours a week by offloading bid submission and daily and monthly trades. Reduces manual work and the risk of human error. Threshold alerts flag irregular meter data such as outages, spikes, and data gaps before they impact bid accuracy.
40+ GWh
Scheduled daily on average
9 of 10
Top US gentailers served
$3/MW
Saved on wholesale purchasing
$1B+
Cumulative P&L impact
1
Data inputs
We identify and collect relevant data streams like meter data, customer hedges, and load profile data.
Hybrid modeling
A blend of physics-based and machine learning models produces our forecasts and updates them hourly as new data arrives.
2
3
Flexible outputs for any workflow
The forecasts integrate seamlessly with Amperon customers’ preferred analytical tools.
Actionable insights
Utilities, retailers and CCAs make better-informed decisions that improve operations and achieve strategic goals.
4
Meter & portfolio demand forecasting FAQs
Here are some answers about our platform, implementation process and pricing.
What is the difference between meter and portfolio demand forecasting?
Meter demand forecasting analyzes the meters owned by a utility (including investor-owned utilities, munis, and co-ops), whereas portfolio demand forecasting analyzes the meters served by a Retail Energy Provider (REP) or Community Choice Aggregator (CCA).
Amperon’s Flexible Data Interface makes it easy for all types of load-serving entities to stand up a forecast, no matter where their meter data comes from.
Amperon’s Flexible Data Interface makes it easy for all types of load-serving entities to stand up a forecast, no matter where their meter data comes from.
How is meter and portfolio load forecasting different from grid-level load forecasting?
Meter- and portfolio-level load forecasts are custom deployments that create a bottom-up view of a load-serving entity’s service territory or book of business, reflecting their specific weather patterns and meter load profiles. This approach creates an entity-specific view of forecasted load, as opposed to a grid-level forecast which aggregates demand across the entire market.
What data do you need to stand up a meter or portfolio demand forecast, and how long does it take?
For meter forecasts, we need individual meter data, such as AMI data, along with load profile data, customer hedges, and meter forward counts. For portfolio forecasts, we also ingest retail customer contract information and substation SCADA net load data. Amperon then selects the weather points best suited to each customer’s specific load profile.
Amperon offers enhanced data cleaning capabilities to resolve erroneous and inaccurate input data before modeling. After this cleaning process, we apply physics-based and machine learning techniques to deploy multiple models, weight them based on accuracy, and ultimately build a single, dynamically weighted forecast. The whole process takes roughly 2 weeks on average, depending on data readiness and cleanliness.
Amperon offers enhanced data cleaning capabilities to resolve erroneous and inaccurate input data before modeling. After this cleaning process, we apply physics-based and machine learning techniques to deploy multiple models, weight them based on accuracy, and ultimately build a single, dynamically weighted forecast. The whole process takes roughly 2 weeks on average, depending on data readiness and cleanliness.
What markets do you serve with meter demand forecasts?
Amperon can stand up meter demand forecasts anywhere in the world. We currently serve clients in all major North American ISOs/RTOs.
Can I see both individual and aggregate views?
Yes, Amperon’s meter data analytics allow users to see both individual and aggregate demand forecasts and to define segments such as residential, commercial, and industrial meters.
Does portfolio demand scheduling replace our current bid submission process?
Yes, Amperon acts as a retail client’s scheduling coordinator of record, fully offloading their day-ahead bid submission work.
Does Amperon schedule on weekends and holidays?
Yes, Amperon schedules 365 days a year. Even on FERC holidays, you can be assured that the Amperon team is here to support your scheduling needs.
Does Amperon schedule physical trades as well?
Yes, Amperon schedules physical trades including daily and monthly trades.
What markets do you serve with portfolio scheduling?
Portfolio scheduling is currently available throughout the United States.
