Electric Power Market Price Estimation Using Multi-Resolution Transition Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately estimate electric power market prices in units of short time periods over extended future periods, such as a few months, due to limitations in predicting fuel availability and market price fluctuations.
Innovation Solution
An estimation system that stores sample data for estimation objects and factors, generates transition patterns, specifies pattern factor dependency relationships, and identifies an estimation model to produce accurate estimates of electric power market prices in short time units over long future periods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If monthly estimation methods are used for long-term predictions, then the estimation period can be extended to several months, but the time resolution deteriorates to monthly units only
Solution Approach 1:
The patent segments the estimation process into multiple time-resolution levels. It generates both monthly estimation results and intra-monthly (hourly, 15-minute, or 1-hour) estimation results by dividing the long-term period into shorter intervals. This segmentation allows the system to provide detailed short-term estimates while covering extended future periods, resolving the contradiction between long estimation duration and fine time resolution.
2Measurement precision
If short-term estimation methods are used, then time resolution can be maintained at 15 minutes or 30 minutes, but the estimation period is limited to relatively short future periods
Solution Approach 1:
The patent introduces a temporal dimension hierarchy by implementing multiple estimation models operating at different time scales. The system maintains high-time-resolution estimates (15-minute or 30-minute intervals) by applying specific estimation models designed for short-term predictions, while simultaneously generating monthly estimates for long-term planning. This multi-dimensional temporal approach allows the system to achieve both fine time resolution and extended estimation periods.
3Duration of action of moving object
If weather prediction data with large deviations is used, then long-term weather forecasts can be obtained, but the accuracy of electric power market price estimation deteriorates
Solution Approach 1:
The patent implements a dynamic model selection mechanism that adapts the estimation approach based on the forecast period and data reliability. For short-term predictions where weather data is reliable, the system uses high-resolution weather-based models. For long-term predictions where weather uncertainty increases, it transitions to using historical patterns and statistical models that are less sensitive to weather prediction deviations. This dynamic adaptation maintains estimation accuracy across different time horizons.
4Adaptability or versatility
If multiple estimation models are used for different time resolutions, then both short-term and long-term estimates can be generated, but the system complexity increases
Solution Approach 1:
The patent implements a universal estimation framework where a core estimation system handles multiple time resolutions and forecast periods through configurable parameters rather than completely separate models. The system uses a unified architecture that can switch between different estimation approaches (weather-based, historical patterns, statistical methods) based on the required time resolution and forecast horizon. This multi-functional design reduces overall system complexity compared to maintaining entirely separate estimation systems for each time scale.
Data Source
AI summary
An estimation system generates, from a value of a time series of an estimation object in a past period, a plurality of patterns of a transition of a value of the estimation object. Based on the plurality of generated patterns and a value of a time series of a factor in the past period, the estimation system specifies a dependency relationship between a transition pattern and a value of the factor and a transition pattern at a past (or future) time point and identifies a model in accordance with the specified dependency relationship. By inputting a value of a time series of the factor in a future period to the estimation model, the estimation system specifies a time series of a value in the future period of the estimation object using at least one transition pattern.