Accurate load shape analysis and load forecasting are essential for profitable participation in deregulated retail electricity markets. End users require precise understanding of their load profiles, the sources of variation in their load, and accurate load forecasts to evaluate alternative direct access supply options, DSM engineering projects, and self generation. Energy providers require the same information for supply planning, service cost estimation, pricing, risk management, delivery scheduling, and power trading. Ultimately, as the electricity markets mature and margins shrink, the costs of load imbalances due to load forecast inaccuracy will have significant impacts on profitability.

LOADCASTER is a comprehensive load analysis, modeling, forecasting, and settlements software system for Energy Service Providers (ESP) that incorporates high precision modeling techniques to minimize load imbalance costs. LOADCASTER incorporates a time-series, relational database structure that is optimized for efficient storage and superfast retrieval and manipulation of large volumes of hourly and subhourly data. Integrated with the database is ESC’s state-of-the-art Pooled Daily Load Shape Regression (PDLS) program (and other estimation methods), powerful model simulation tools, and a flexible report writer.

PDLS estimates end user (and/or end user group) load profile equations that accurately decompose (hourly and subhourly) load variations by source, including: normal operations and equipment cycling, business activity levels, weather, electricity price, day of week, month, holiday, etc. The technique produces statistical forecasting equations for each day, the coefficients of which can be clearly identified and interpreted to reveal the impacts of explanatory variables on load. The PDLS procedure can be extended to create equations that can algebraically approximate just about any conceivable load shape to almost any degree of accuracy required.

PDLS combines the daily load profile equations into complete load models that can be easily simulated to: produce "normal conditions" long term load forecasts for power supply planning; perform "what if" and "Monte Carlo" simulations for assessing load risk; produce short-term forecasts for scheduling power deliveries; and evaluate ex post model forecasting performance.

Availability:

LOADCASTER is a TRM^TM application product and is available for installation on client IBM and S/390® servers operating under VM/ESA®.

e-LOADCASTER is in development and will soon be available to licensed users through the internet using standard browsers.