H-Infinity Filter Bias Correction for Future Prediction Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing post-processing methods for predicting physical quantities, such as weather and solar power generation, are limited by their dependence on statistical information, which restricts the precision of future bias modeling and prediction.
Innovation Solution
A prediction value correction method using an H-infinity filter to estimate future bias from past prediction and measurement values, allowing for precise prediction value correction without relying on statistical information, implemented in a prediction value correction apparatus with modules for past value provision, filtering, bias estimation, and future value correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If statistical information-based models are used to model future bias, then the model can be built using past prediction and measurement values, but the precision of future bias modeling is restricted
Solution Approach 1:
The patent changes the fundamental parameter of the modeling approach from statistical information-based to H-infinity filter-based. This parameter change enables more precise future bias modeling by using a different mathematical framework that does not rely on statistical assumptions about the bias characteristics, thereby resolving the contradiction between modeling precision and model complexity.
Solution Approach 2:
The patent substitutes the statistical modeling mechanism with an H-infinity filter mechanism. Instead of using statistical methods to model future bias, the system uses the H-infinity filter which provides a more robust framework for handling uncertainties and achieving precise bias estimation without being constrained by statistical information requirements.
2Measurement precision
If complex filtering processes are used to improve prediction accuracy, then prediction precision is improved, but program execution time increases
Solution Approach 1:
The patent changes the filtering approach parameter from traditional complex filters to H-infinity filter with specific optimizations. This parameter change maintains high prediction accuracy while reducing computational complexity and execution time by using an efficient implementation of the H-infinity filtering algorithm that is better suited for real-time prediction applications.
3Measurement precision
If traditional post-processing methods are used, then the process is simple, but the prediction accuracy is limited due to dependence on statistical information
Solution Approach 1:
The patent substitutes traditional statistical post-processing methods with an H-infinity filter-based approach. This substitution eliminates the dependence on statistical information while achieving superior prediction accuracy. The H-infinity filter provides a more sophisticated yet computationally efficient mechanism for bias correction that overcomes the limitations of traditional methods.
Data Source
AI summary
Provided are a prediction value correction method and apparatus. The prediction value correction method includes steps of: (a) determining a prediction condition to be predicted; (b) receiving past prediction values and past measurement values according to the determined prediction condition; (c) filtering the past prediction values and the past measurement values by using an H-infinity filter to obtain an output value for a final time point; (d) estimating a future bias for a date and time point to be predicted by using the output value of the H-infinity filter; and (e) correcting a future prediction value for the date and time point to be predicted by using the estimated future bias to obtain a corrected future prediction value for the date and time point to be predicted.