Flood Estimation via Iterative Input Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hydrological models for flood estimation are prone to inaccuracies due to errors in input data and model sensitivity, leading to unreliable peak flood level predictions, and are computationally expensive and resource-intensive, making them impractical for continuous updates during a flood event.
Innovation Solution
A flood estimation method that iteratively validates and updates input data using external flood data by generating candidate input datasets through stochastic sampling, comparing model outputs against external data, and selecting satisfactory datasets to improve accuracy, allowing for more frequent and accurate flood level estimates using simpler models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex hydrological models are used to improve accuracy, then prediction accuracy is improved, but computational time and resources increase significantly
Solution Approach 1:
The system performs preliminary validation of input data against external flood data before running the hydrological model. By pre-screening and validating input data quality, the system avoids running complex models with erroneous inputs, thereby reducing unnecessary computational time while maintaining prediction accuracy.
Solution Approach 2:
The system implements a feedback mechanism where model outputs are compared against external flood data, and this comparison feeds back into validating and updating the input data for subsequent model runs. This iterative feedback loop improves prediction accuracy over time without requiring increasingly complex models, as the same model becomes more reliable through refined input validation.
2Measurement precision
If complex hydrological models are used to improve accuracy, then prediction accuracy is improved, but computational resources and expense increase
Solution Approach 1:
The system extracts and separates the data validation function from the hydrological modeling function. By taking out the input validation step and performing it independently using external flood data, the system eliminates the need to run complex, resource-intensive models multiple times, thereby reducing computational resource usage while maintaining accuracy through validated inputs.
Solution Approach 2:
The system enables self-service by automatically validating input data against external flood data without requiring manual intervention or complex model configurations. This automated self-validation process reduces the need for resource-intensive manual data processing and model tuning, thereby lowering computational expenses while improving reliability.
3Measurement precision
If hydrological models are continuously run to account for updated input data, then accuracy is improved, but it is impractical due to complexity
Solution Approach 1:
The system performs preliminary validation of updated input data against external flood data before incorporating it into the hydrological model. This pre-validation step ensures that only accurate, verified data updates are processed, allowing the system to maintain high accuracy without requiring continuous, complex model re-runs. The preliminary data screening simplifies the overall process by filtering out unnecessary computational steps.
4Productivity
If simpler models are used to reduce computational cost, then computational efficiency is improved, but prediction accuracy deteriorates
Solution Approach 1:
The system implements a feedback mechanism where simpler model outputs are validated against external flood data, and this validation feedback is used to refine and update the input data for subsequent model runs. This iterative feedback process allows simpler, computationally efficient models to achieve improved prediction accuracy over time through continuous refinement of their inputs, without requiring complex model architectures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, systems, and techniques for estimating flooding are disclosed. A flood estimation method comprises receiving input data associated with a flood event for executing a hydrological model; receiving external flood data of the flood event; validating the input data over one or more iterations based on the external flood data; and executing the hydrological model using validated input data to estimate water levels of the flood event, the validated input data based on a candidate input dataset providing a candidate model output with a satisfactory fit to the external flood data.