AI N-Value Prediction Using Augmented Borehole Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate identification of N-values in undrilled regions is challenging due to limited drilling data, leading to errors in pile design and quantity calculation, which causes delays and increased costs in construction projects.
Innovation Solution
A method and apparatus using artificial intelligence (AI) for N-value prediction that employs data augmentation by generating hypothetical N-values around actual locations, utilizing circular augmentation to increase data sets and improve model accuracy, incorporating techniques like automatic machine learning and neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If limited drilling data is used for N-value estimation, then the bidding period and investigation time are reduced, but the accuracy of N-value prediction at undrilled points deteriorates
Solution Approach 1:
The patent creates virtual copies of actual drilling data by generating hypothetical N-values at hypothetical points around actual locations. This copying approach allows the system to expand limited actual data into a larger training dataset, improving prediction accuracy without requiring additional physical drilling investigations during the bidding period.
Solution Approach 2:
The patent introduces an artificial intelligence model as an intermediary between limited drilling data and N-value predictions at undrilled points. This AI intermediary learns patterns from actual data and hypothetical augmented data, then applies this learned knowledge to predict N-values at locations where no drilling was performed, bridging the gap between sparse measurements and comprehensive ground characterization.
2Reliability
If more drilling investigations are conducted to improve N-value accuracy, then the reliability of ground characteristics improves, but the cost and time required for investigation increase
Solution Approach 1:
Instead of performing additional physical drilling investigations, the patent creates virtual drilling data through hypothetical points and generated N-values. This copying strategy allows the training dataset to expand from limited actual drilling results to a comprehensive dataset covering the entire project area, achieving high reliability without proportional increases in drilling quantity.
Solution Approach 2:
The patent performs data augmentation and AI model training in advance before the actual bidding and design processes. By preparing the prediction model beforehand with augmented data, the system establishes reliable ground characteristic predictions without requiring extensive drilling investigations at the time of bidding, thus improving reliability while minimizing investigation quantity during critical time periods.
3Ease of manufacture
If engineer's empirical judgement is used for N-value estimation at undrilled points, then no additional drilling is required, but errors in pile design and quantity calculation increase
Solution Approach 1:
The patent replaces the mechanical system of manual empirical judgement with an artificial intelligence-based prediction system. Instead of relying on engineers' subjective experience and manual estimation, the system uses trained AI models that objectively analyze drilling data and hypothetical data to generate N-value predictions, thereby improving design accuracy while maintaining ease of use through automated calculations.
Solution Approach 2:
The patent enables the system to automatically generate N-value predictions at undrilled points without requiring manual intervention for each prediction point. The AI model, trained on augmented data, serves itself by automatically processing input coordinates and generating reliable N-value estimates, eliminating the need for engineers to manually estimate values and reducing human error in the design process.
Data Source
AI summary
An N-value prediction apparatus according to an embodiment of the present invention includes a hypothetical learning data augmentation unit, based on an actual N-value measured at an actual location according to the Standard Penetration Test through drilling investigation, generating at least one of hypothetical N-values corresponding to a preset hypothetical point based on the actual location, an N-value prediction model learning unit learning ground characteristic data corresponding to each of the actual location and the hypothetical point by artificial intelligence, the ground characteristic data including the actual N-value and the hypothetical N-values, and an N-value prediction result calculation unit predicting an N-value at an arbitrary prediction target point by using an N-value prediction model generated by artificial intelligence learning executed in the N-value prediction model learning unit.