QR Logo
Loading...

AI Load & Demand Forecasting Solutions & Services

  • Consumers, retail and utility total & net load.

  • Intraday, 7 DA & long-term (20+ years) forecasts.
  • New forecast publication every 15 mins, 24/7.
  • Probabilistic forecasts, upper & lower uncertainty bands.
Contact us to explore a free POC trial.

QR AI Forecaster offers robust and flexible intraday, 7 days ahead, and long-term load forecasting services tailored to meet the requirements of utilities, munis, electric co-ops, energy retail providers and marketers, and grid operators.

Challenges

Clients’ profile. Load serving entities such as utilities, electric co-ops and munis, energy retail providers and marketers, grid and transmission line operators and generators, operating in real-time intraday and day-ahead 5-60 mins resolution markets.

Targets:
1. Accurate and flexible load forecasts, aggregating or drilling load by any desired criteria, i.e., geographic regions, cities, distribution nodes, or by consumer type (residential, commercial, industrial).
2. Net load forecast, i.e., demand net of behind the meter solar generation, needed to optimize the integration of microgrid fleets with utilities’ macro grid.
3. These AI demand forecast models can be short-term intraday, up to 7 days ahead, and medium to long term. In the latter case the model should include what-if-scenarios on exogenous factors, e.g., a shift in temperature up or down by 10 degrees, or some global economic or localized indicators such as consumer load growth by 9%, etc.
4. Implement real-time 24/7 automated, stand-alone and accurate load forecast solutions, as a web cloud solution that can be easily integrated with in-house trading systems, ISOs, weather data services, and internal data sources, e.g., meters, SCADA, etc.

AI Load Forecasting Advanced Ecosystem

  • Microservice architecture enables accelerated custom configuration of AI load forecast solutions.
  • API-friendly ecosystem allows flexible integration and data exchange with third party & in-house load, meter and weather data systems. Functionalities are constantly enhanced and extended to meet clients requirements.
Contact us to explore a free POC trial.

Cutting-Edge Load Forecasting Solutions

QR Data Hub

  • A key challenge for Energy Retail Providers operating through multiple utility jurisdictions is to integrate and use data through various third-party data feeds, each using different data formats. Similarly, utilities and Load Serving Entities have large amounts of meter data in multiple and often legacy systems and Excel files.
  • QR Data Hub is a robust data management platform offering great flexibility to collect, integrate, format and organize meter, SCADA and consumer consumption load into standard formats within an optimized data platform, suitable for AI load forecasting, analytics, reporting, data warehousing and export.
  • QR Data Hub is compatible with Electronic Data Interchange (EDI) billing data feeds, individual utilities legacy and third-party automated Utility Data Platform or Metering Services, and weather sites to collect the fullest range of data used in load forecasting.
  • Connectivity can be via Restful API, sql, direct file (csv, Excel, XML, Json) fetching.

Retail Energy Provider Load Forecasting

  • Energy Retail Providers operate through multiple utility jurisdictions. A first challenge is to integrate their data through different data feeds and file formats. QR Meter Data Hub is used to integrate and manage various Electronic Data Interchange (EDI) billing data feeds, individual utility and third-party automated Utility Data Platform or Metering Services.
  • Another challenge is that cumulative electricity meter consumption is read once a month or less frequently. QR AI Load Forecaster models can use approved load profiles to estimate 5-60 min load consumption for forecasting. On the other hand, our AI models can also use advanced or smart grid meter data at 5-60 minute granularity.
  • Load forecasting for individual consumers.
  • Bottom-up & top-down load forecasting models per arbitrary aggregation rules such as geolocation, per type, e.g., single or multifamily w/ air conditioning or space heater, offices, malls, commercial or industrial.
  • Short-term intraday & 7 DA load forecasts at 5-60 min temporal resolution.
  • Long-term load forecasts from months to years at daily and monthly resolutions.
  • Extreme scenario load forecasting.

Utility & LSE Load Forecasting

  • Load forecasting for individual load nodes, substations or consumers.
  • Bottom-up load forecast modeling per arbitrary aggregation rules such as geolocation, per consumer types, e.g., single or multifamily w/ air conditioning or space heater, offices, malls, commercial or industrial.
  • Intraday & 7 DA load forecasts with 5-60 min temporal resolution.
  • Medium-term load forecasts, 3+ years, are performed at daily and monthly resolution. 
  • Long-term load forecasts, 30+ years, are performed at monthly resolution.
  • Medium & Long-term forecasts are brought down to 5-60 minutes resolutions using seasonal patterns and delivery profiles.
  • Long & Medium Term load forecasts include what-if-scenarios based on exogenous factors, e.g., temperature +/- 100, demand growth + 9%, or some econometrics factors growth of 3%. Load and exogenous input data can be at any timescale, the system automatically scales them to monthly for forecasting.
  • Net load forecasting, that is, load net of behind-the-meter solar generation, scattered across fleets of individual PV stations.
  • Examples of econometrics data we provide and forecast are: Recession Probabilities, Durable Goods, Inflation Rate, 3M Yield, Total Non-farm Payroll, Consumer Sentiment, Real GDP Per Capita, Unemployment Rate, GDP per capita.
  • QR Data Hub is used to connect to your utility and third-party data feed and weathers sites to collect the fullest range of data used in load forecasting.

Electricity Market System Demand

  • Demand forecasting service for world electricity system operators and markets, also known as Independent System Operators (ISOs), Regional Transmission Organizations (RTOs), Transmission System Operators (TSO).
  • We compute demand forecasts per regional demand or transmission zones, or Trading Hubs.
  • Forecast temporal granularity is 5 to 60 min. Forecast frequency is every 10 min for intraday forecasts and 10 am local time for 7 DA forecasts.
  • This service does not require any data from clients. QR Data Hub automatically fetches and manages demand data from market operator data services, as well as weather data vendors.

QR AI Load Forecaster Features

  • Medium-term load forecasts, 3+ years, are at daily resolution. 
  • Long-term load forecasts, 30+ years, are at monthly resolution.
  • Short-term intraday & 7 DA forecasts are at 5–60 minutes resolution.
  • Forecasts computation frequency: 9:00 for DA, and every 10 mins, 24/7 for intraday forecasts.
  • Bottom-up load forecast modeling aggregates load data on the fly, by any criteria, e.g., from individual meters, to geolocations, consumer types, or subsidiaries.
  • Medium and long-term load forecasts include what-if-scenarios on exogenous factors, e.g., temperature +/- 10o, demand growth + 9%, etc.
  • Forecast models can be probabilistic with an average, high, and low forecasts at set percentiles, e.g., 5 & 95, 10 & 90, etc.
  • Extreme case scenarios can be simulated on top of AI load forecasts. We can furthermore combine these with probabilistic forecasts to account for the volatility of external drivers such as weather.
  • Numerous deep and machine learning AI forecast models.
  • Multiple weather data indicators can be added as features to any AI load forecast models. We accommodate any weather data feed.
  • Peak load forecasting and reporting.
  • Data visualization, dashboards, reporting, alerting.

No-coding on-screen contextual feature engineering: similar points, and temporal features.

On screen load data preprocessing (gaps, outliers, quantization), contextual feature engineering

No-coding on-screen drag & drop  feature engineering: weather, radiance and other time series.

On screen drag & drop machine learning AI load forecast modeling. Build, fine-tune & deploy models.

No coding drag & drop interface to build & deploy complex deep learning load forecast models. 

Easily build and deploy probabilistic AI load forecast modeling On screen drag & drop.

Bottom up load forecasting model: aggregating load pre-forecasting and making a single forecast.

Bottom up load forecasting model: forecasting individual load drivers and aggregating load post-forecast.

Custom data visualization (forecasts and actual), interactive dashboards and reports. Data exporting.

Benefits

  • You can always start with a free POC Trial.
  • Two flexible delivery options:
    • Forecast as a Service (FaaS). Our AI platform and expert team do everything, and you receive, via API and in web dashboards, intraday and DA forecasts every 5–10 minutes, 24/7. No software installation, hardware and IT resources are needed.
    • Software as a Service (SaaS) where we implement QR AI Forecaster on a private cloud of your choosing, or on-site, and you control the data & of the AI models.
  • Cost-effective pricing. Whether you use QR AI Forecaster as FaaS or SaaS, you pay one single annual subscription fee. The fee comprises the license, maintenance and upgrade releases.
  • Ready-to use platform. We offer a large number of preconfigured Deep & Machine Learning models especially designed for load forecasting. These can be instantly trained on any data set of interest.
  • Expert support team. Creating and maintaining forecast solutions in-house, using generic AI tools, require extensive use of data scientists. When working with QuantRisk our data scientists configure, fine-tune and deploy your forecast models and maintain them in time as data behavior changes.
  • Forecast results are published via API and in web dashboards with data visualization and manual download options.

How to Proceed

Whether we install QR Load Forecaster on your private cloud or on-site, or you use it as a service, the road map to follow is the same. You can always start with a POC Trial. Our team is here to assist and work with you every step of the way.
How to Proceed-Load Forecast-Mobile

Case Studies & Video Demos

Download Brochures

Let's talk about your needs

We look forward to exploring the range of options for your projects. Please write to us and one of our project managers will get back to you at once.