Dynamic Load Forecasting for Back-Feed Capacity Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power distribution network systems face challenges in accurately determining load capacities for back-feed pathways after a fault, as static load ratings do not account for varying load demands over time, which can lead to overloading and undesirable consequences like cascade tripping events.
Innovation Solution
A method and system for forecasting power use in a power distribution network that involves selecting a load profile, executing a short-term load estimation, and generating modified load profiles with safety margins to determine the maximum load values for back-feed capacity checks, ensuring alternate power sources can safely restore power without overloading.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If static rated load values are used for back-feed capacity determination, then the system is simple and can be programmed offline, but the actual load variations over time are not accounted for, leading to potential overloading
Solution Approach 1:
The patent transforms the static load rating approach into a dynamic load forecasting approach. Instead of using fixed rated values, the system implements short-term load forecasting that adapts to time-varying load demands. The load forecast is updated based on the time of day and day of week, allowing the back-feed capacity determination to reflect actual loading conditions rather than conservative rated values.
Solution Approach 2:
The patent changes the parameter used for load estimation from static rated values to dynamic forecasted values. The load forecast parameter varies based on temporal factors (time of day, day of week), enabling the system to account for load variations without requiring complex real-time measurements. This parameter transformation resolves the contradiction by providing both accuracy and computational feasibility.
2Reliability
If load forecasting is implemented to account for time-varying demands, then back-feed operation safety is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing load forecasts for different time periods (time of day and day of week combinations) before back-feed operations are needed. When a back-feed operation is required, the system simply retrieves the pre-computed forecast for the current time period rather than performing complex real-time calculations. This approach maintains reliability through accurate forecasting while minimizing the computational complexity during critical restoration operations.
3Reliability
If conservative load estimates are used, then overloading is prevented, but the available back-feed capacity is underestimated, limiting restoration options
Solution Approach 1:
The patent uses dynamic load forecasting that adapts to actual loading patterns rather than relying on conservative static ratings. By forecasting the actual load at the time of back-feed operation, the system determines the true available capacity of alternate sources. This enables more restoration path options to be identified as feasible, since the capacity assessment reflects real-world conditions rather than conservative estimates, while still maintaining overloading prevention through the forecasting accuracy.
Data Source
AI summary
A method of forecasting loads on a distribution network is disclosed. The method employs load profiles in combination with short-term load estimations to forecast expected loads on a distribution network. The predicted loads may then be used by power restoration controllers when performing back-feed capacity checks.