This document provides an overview of load forecasting in power systems. It discusses short, medium, and long term load forecasting and their objectives. Short term forecasting spans 1 hour to 1 week and is used for power system operations. Medium term is 1 week to several weeks and used for fuel planning. Long term is 1 year to 20 years and used for transmission planning. The document outlines different forecasting methods and considerations for each term and discusses important factors like weather and historical load data that influence load forecasting.
This document provides an overview of load forecasting in power systems. It discusses short, medium, and long term load forecasting and their objectives. Short term forecasting spans 1 hour to 1 week and is used for power system operations. Medium term is 1 week to several weeks and used for fuel planning. Long term is 1 year to 20 years and used for transmission planning. The document outlines different forecasting methods and considerations for each term and discusses important factors like weather and historical load data that influence load forecasting.
This document provides an overview of load forecasting in power systems. It discusses short, medium, and long term load forecasting and their objectives. Short term forecasting spans 1 hour to 1 week and is used for power system operations. Medium term is 1 week to several weeks and used for fuel planning. Long term is 1 year to 20 years and used for transmission planning. The document outlines different forecasting methods and considerations for each term and discusses important factors like weather and historical load data that influence load forecasting.
This document provides an overview of load forecasting in power systems. It discusses short, medium, and long term load forecasting and their objectives. Short term forecasting spans 1 hour to 1 week and is used for power system operations. Medium term is 1 week to several weeks and used for fuel planning. Long term is 1 year to 20 years and used for transmission planning. The document outlines different forecasting methods and considerations for each term and discusses important factors like weather and historical load data that influence load forecasting.
Electrical Engineering Department Umm Al-Qura University Announcements 2
Virtual Office hours are upon your request; send an email at
yahazmi@uqu.edu.sa
Term Project topics will be posted this week
2 Load Forecast
Objectives 1. Defining the different classes of load forecasting for distribution systems • Short load forecast • Medium load forecast • Long load forecast. 2. Data preparation for short and medium load forecast 3. Presenting the classification for short and medium load forecast. 4. Identifying the load forecasting methods suitable for distribution systems. 5. Presenting the mathematical formulation for different load forecasting methods. 6. Presenting the state of the arts load forecasting methods. 10/4/2020 33 Introduction
• Load forecasting at the distribution level is a major component for the planning and proper operation of the distribution network.
• Frequently there is a time lag between awareness of an impending increase in load
demand and the occurrence of that increase.
• This time allows electrical engineers to perform the task of planning and forecasting needed to meet the expected demand increase.
• Load forecast is needed to determine when an increase on load will occur so that appropriate actions can be taken
10/4/2020 44 Introduction
The factors that affect load forecast are:
1. Historical load data
2. Load density 3. Land use 4. City planning 5. Industrial plans 6. Community development plan 7. Alternative energy source available 8. Population growth
10/4/2020 55 Introduction
10/4/2020 66 Load Forecast
10/4/2020 77 Short Term Load Forecast (STLF)
• It spans the period from one hour up to one week
• It is mainly utilized for: o Power system operation studies o Losses reduction o Voltage regulations o Unit commitment o Maximizing the utility revenues in the deregulated environment.
10/4/2020 88 Medium Term Load Forecast (MTLF)
• It spans the period from one week to several weeks
• It is mainly utilized for: o Predicting the necessary power to purchase or sell from other neighboring networks (inter-tie exchanged power) o The fuel required by the utility in the near future.
10/4/2020 99 Long Term Load Forecast (LTLF) • It spans over a period ranging from one year up to twenty years • It is necessary for planning and expansion studies. • It is mandatory to know how much power we will need in the future and where this power is needed. • Accurately estimating the required amount of capacity in future with poor estimation about the location where these capacities are required will lead to installing more capacities in areas which do not required expansion • May lead to shortage of power in some regions and unutilized capacities in some other regions. 10/4/2020 1010 Requirements for Different Forecast Models • In short and medium electric load forecast, it is required to know how much power we will need and at what time of the day is provided • The information regarding where this demand is required is not of a major concern. • Accurate STLFM required data which is associated mainly with o the time dimension o historical load data o historical weather conditions o the predicted weather condition o the nature of the day and the season 10/4/2020 1111 Requirements for Different Forecast Models
A variety of methods have been deployed for short and medium
electric load forecast such as
• naïve approaches, • simple regression analysis, • time series analysis, • methods which are based on soft-computing.
10/4/2020 1212 Requirements for Different Forecast Models
Beside time and weather conditions a variety of factors will affect the long term forecasting; these factors include but are not limited to: • economic growth, • city plans, • social factors, • political factors.
10/4/2020 1313 Requirements for Different Forecast Models
• Several approaches have been utilized for long term electric
load forecasts
• These approaches should consider the spatial information
in their formulations.
• Trending and simulation methods are the most powerful and
acceptable approaches which deal with the spatial load forecast.
10/4/2020 1414 Data Collection and Preprocessing
The following data are important for proper load forecast studies:
• Load in MW and MVAR
• Weather conditions (temperature, dew point, pressure, sky cover, visibility, weather forecasts, etc) • Economic indicators (local industrial production, housing starts, etc.) • Customer counts and types as a function of system, substation, or buss. 10/4/2020 1515 Data Collection and Preprocessing
The following data are important for proper load forecast studies:
• Load in MW and MVAR
• Weather conditions (temperature, dew point, pressure, sky cover, visibility, weather forecasts, etc) • Economic indicators (local industrial production, housing starts, etc.) • Customer counts and types as a function of system, substation, or buss. 10/4/2020 1616 Data Collection and Preprocessing
In electric load forecast studies three factors should be handled
carefully:
1- Time factor 2- Weather conditions 3- The historical load demand.
10/5/2020 1717 Data Collection and Preprocessing
The time factors take into account:
the time of the year, the day of the week, the hour of the day.
An index can be utilized in load forecasting studies which distinguish
between weekdays and weekends. The load on different weekdays also can behave differently. For example, Mondays and Fridays being adjacent to weekends, may have structurally different loads than Tuesday through Thursday. 10/5/2020 1818 Data Collection and Preprocessing
The weather conditions that influencing the load
Various weather variables are considered by different utilities and research engineers to capture the effect of weather condition on electric load forecast • Temperature, • wind, • humidity.
10/5/2020 1919 Data Collection and Preprocessing
The weather conditions that influencing the load
Utilities widely utilize two factors to capture these effects:
the first factor is related to temperature and humidity index and is usually utilized in summer to capture the effect of heat and humidity on the electric consumption. Other indices which are related to wind speed, temperature and rate of ice falling are utilized in winter.
10/5/2020 2020 Data Collection and Preprocessing
The historical load demand data
The raw data collected from different sources should be pre-processed
in order to achieve proper load forecast quality; the following chronicle steps are generally required:
• Remove all negative and zero values.
• Remove outlying data that have values that are much greater than, or less than, the average values. • Remove patterns that have time gradients that are too large. 10/5/2020 2121 Data Collection and Preprocessing
The historical load demand data
• Each individual data set should be inspected manually.
• However, when there are large data sets to analyze, manually cleaning-up each data file individually will require substantial time and effort • Automated examination may be the best alternative, using some well established statistically-based methods.
