Electric Load Forecasting: Fundamentals and Best Practices
Education Category Image


  
Electric Load Forecasting: Fundamentals and Best Practices
BELF : BELF
This course introduces electric load forecasting from both statistical and practical aspects using language and examples from the power industry. Through conceptual and hands-on exercises, participants experience load forecasting for a variety of horizons from a few hours ahead to 30 years ahead. The overall aims are to prepare and sharpen the statistical and analytical skills of participants in dealing with real-world load forecasting problems and improve their ability to design, develop, document, and report sound and defensible load forecasts. According to statistics gathered on the first five offerings, this course was highly rated by students who ranged from new graduates with no industry or SAS experience to forecasting experts with over 30 years of industry experience and over 20 years of SAS programming background. The students represented all sectors of the industry: G&T, ISO, distribution companies, REPs, IOU, co-op, municipal, regulatory commission, and consulting firm. Titles of the participants ranged from analyst, engineer, manager, to director and vice president. For advanced topics, pair this course with Electric Load Forecasting: Advanced Topics and Case Studies. The two courses are offered on contiguous days.
Learn How To
  • Classify load forecasts.
  • Use basic graphic methods to discover the salient features of load profiles.
  • Build a benchmark model for a wide range of utilities.
  • Capture special effects for a local utility.
  • Forecast loads for both small and large utilities.
  • Improve very short-term forecasting accuracy.
  • Perform weather normalization
  • Use macroeconomic indicators for long-term load forecasts.
  • Detect outliers
  • Continue improving forecasting practice.
  • Avoid making frequently made mistakes.
    • Who Should Attend
    Load/price forecasters, energy traders, quantitative/business analysts in the utility industry, power system planners, power system operators, load research analysts, and rate design analysts
    Prerequisites
    Before attending this course, you should:;
  • Have a basic knowledge of the utility industry.
  • Have a basic understanding of forecasting.
    • SAS Products Covered
    SAS/ETS;SAS/STAT
    Course Outline
    Introduction to Electric Load Forecasting
  • Overview of the electric power industry.
  • Business needs of load forecasts.
  • Driving factors of electricity consumption.
  • Classification of load forecasts.
  • Software applications.
    • Salient Features of Electric Load Series
  • General approach to electric load forecasting.
  • Overview of the data pool.
  • Trend and seasonality.
  • More salient features.
    • Multiple Linear Regression
  • Naive models.
  • Trend.
  • Class variables.
  • Polynomial regression.
  • Interaction regression.
  • Rolling regression.
    • A Naive Benchmark for Short-Term Load Forecasting
  • Motivation.
  • Criterion.
  • A naive MLR benchmark.
  • Applications.
  • Two more salient features.
    • Customizing the Benchmarking Model
  • Recency effect.
  • Weekend effect.
  • Holiday effect.
  • Case studies.
  • Two more salient features.
    • Very Short-Term Load Forecasting
  • Hour-ahead load forecasting.
  • Weighted least squares regression.
  • Dynamic regression.
  • Two-stage method.
  • Extensions.
    • Medium/Long-Term Load Forecasting
  • Macroeconomic indicator.
  • Weather normalization.
  • Forecasting with weather variation.
  • Forecasting with cross scenarios.
    • Variables, Methods, Techniques, and Further Readings
  • Load, weather, calendar, macroeconomic indicator, and so on.
  • Similar day and hierarchy.
  • Regression.
  • ARIMA.
  • Exponential smoothing.
  • Support vector machine.
  • Artificial neural networks.
  • Fuzzy systems and fuzzy regression.
  • Relevant and readable books.
  • Load forecasting papers.
  • Training courses.
    • Frequently Made Mistakes
  • Counterexamples.
  • Expectation.
  • Data.
  • Models.
  • Decisions.
    • Software Applications
  • Software Applications
    		•
    Skip Earn this badge

    Earn this badge

    Badge Image