Load forecasting is a fundamental element in utility business operations and planning processes. During the past 120-plus years, load forecasting methodologies have evolved as the industry and related technologies advanced. Consequently, many classical methods are no longer suitable in addressing today’s challenges in the utility industry.
This course offers a comprehensive and in-depth treatment to long-term load forecasting. The content includes:
- A review of the fundamental concepts and classical methods
- A statistical approach that leverages high resolution data and modern computing power
- Probabilistic forecasting that helps better quantify the uncertainties of the unpredictable future
- Several advanced and emerging challenges triggered by big data
- Renewable energy integration and demand side management programs
This course brings the long-term load forecasting methodologies to the most widely-used business forecasting tool, spreadsheets. It starts with an overview of various options available in the market, and then zooms into spreadsheet operations. We will demonstrate four different ways of doing forecasting in spreadsheets, and help the participants understand the limitations of each. The participants will gain hands-on experience with building models and generating load forecasts in spreadsheets, while competing and collaborating with others. Real-world examples and case studies are embedded throughout the course when introducing the theories and methodologies.
This will be a working session. Each participant should bring a laptop with MS Excel 2007 (or newer versions) with add-ins of Solver and Data Analysis.
This is an intermediate course. Individuals not having experience in forecasting should consider attending the preceding program on Fundamentals.
Course Learning Outcomes
Attendees will cover materials and engage in discussions that will allow them to:
- Review the fundamental concepts of load forecasting
- Recognize the evolution of classical load forecasting methods
- Prepare and build regression models with hourly data
- Build and conduct ex post forecasting to analyze the source of errors
- Evaluate point and probabilistic load forecasts
- Apply and conduct what-if analysis
- Employ and conduct weather normalization
- Evaluate the distribution of forecast errors
- Manage how to perform data cleansing
- Discuss emerging topics and the applicable problem-solving principles
EUCI has been accredited as an Authorized Provider by the International Association for Continuing Education and Training (IACET). In obtaining this accreditation, EUCI has demonstrated that it complies with the ANSI/IACET Standard which is recognized internationally as a standard of good practice. As a result of their Authorized Provider status, EUCI is authorized to offer IACET CEUs for its programs that qualify under the ANSI/IACET Standard.
EUCI is authorized by IACET to offer 1.0 CEUs for this event.
Requirements for Successful Completion of Program
Participants must sign in/out each day and be in attendance for the entirety of the course to be eligible for any continuing education credit.
PowerPoint presentations, case studies, and workshop exercises will be used in this program.
Thursday, June 8, 2017
8:00 – 8:30 a.m. :: Registration and Continental Breakfast
8:30 – 10:15 a.m. :: Introduction to Long-Term Load Forecasting
- Utility applications
- How much is 1% error?
- Principles of forecasting
- Load forecasting terminologies
- Exercise: calculating error measures
10:15 – 10:30 a.m. :: Morning Break
10:30 a.m. – 12:00 p.m. :: Evolution of Classical Methods
- The magic ruler
- Regression models with monthly or daily data
- S-curve and spatial load forecasting
- End use studies
12:00 – 1:15 p.m. :: Group Luncheon
1:15 – 3:00 p.m. :: A Statistical Approach
- Golden insights from hourly data
- Polynomials or piecewise functions?
- A large but small model
- Is it economists’ fault?
3:00 – 3:30 p.m. :: Networking Break
3:30 – 5:00 p.m. :: Exercise: Load Forecasting with Hourly Data
5:00 p.m. :: Course Adjourns for the Day
Friday, June 9, 2017
8:00 – 8:30 a.m. :: Continental Breakfast
8:30 – 10:00 a.m. :: Probabilistic Load Forecasting
- Evaluation methods
- Exercise: calculating pinball loss
- Scenario analysis
- Exercise: scenario-based probabilistic load forecasting
10:00 – 10:15 a.m. :: Morning Break
10:15 – 11:00 a.m. :: Probabilistic Load Forecasting (Continued)
- Load normalization against weather
- Exercise: probabilistic load forecasting and normalization
11:00 – 11:45 a.m. :: Challenges and Emerging Topics
- Data, data, data
- Distributed generation
- Energy efficiency and demand response
- Electric vehicles
- Forecast override, override what, and how?
11:45 a.m. :: Course Adjourns
Tao Hong, Director of BigDEAL (Big Data Energy Analytics Laboratory), University of North Carolina at Charlotte and Chief Data Scientist, Hong Analytics
Dr. Tao Hong is the Director of BigDEAL (Big Data Energy Analytics Laboratory) at University of North Carolina at Charlotte and Chief Data Scientist of Hong Analytics. He has been providing training and consulting services to more than 100 organizations in the energy industry worldwide. He is the Founding Chair of IEEE Working Group on Energy Forecasting, General Chair of Global Energy Forecasting Competition, lead author of the online book Electric Load Forecasting: Fundamentals and Best Practices, and author of the blog Energy Forecasting. Dr. Hong received his B.Eng. in Automation from Tsinghua University in Beijing and his PhD with co-majors in Operations Research and Electrical Engineering from North Carolina State University.
The Hotel Minneapolis, Autograph Collection
215 S 4th St.
Minneapolis, MN 55401
To reserve your room, please call 1-612-340-2000
Please indicate that you are with the EUCI group to receive the group rate.
The room rate is $229.00 single or double plus applicable taxes.
Room Block Dates:
A room block has been reserved for the nights of June 4 – 8, 2017.
Rate Available Until:
Make your reservations prior to May 17, 2017. There are a limited number of rooms available at the conference rate. Please make your reservations early.