Prediction apparatus and method for N value using artificial intelligence and data augmentation
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.
1 . An N-value prediction apparatus, comprising;
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,
wherein the hypothetical learning data augmentation unit uses circular augmentation where multiple hypothetical points are set up radially in a certain distance centering the actual location-,
wherein the multiple hypothetical points consist of 8 points in a radius of 0.5 m, 12 points in a radius of 1 m, and 16 points in a radius of 2 m centering around the actual location.
2 . The N-value prediction apparatus of claim 1 , wherein the ground characteristic data includes a location information and a soil information corresponding to the actual location and the hypothetical point.
3 . The N-value prediction apparatus of claim 2 , wherein the location information includes at least one of latitude, longitude, altitude, depth, and an absolute position based on depth and altitude.
4 . The N-value prediction apparatus of claim 2 , wherein the N-value prediction model learning unit uses an artificial intelligence learning structure constructed by at least one of automatic machine learning technique, decision tree technique, and artificial neural network technique, which contains an input layer, an output layer, and at least one hidden layer built between the input layer and the output layer, and
the N-value prediction model learning unit is repeatedly learnt by the said artificial intelligence learning structure so that the location information and the soil information corresponding to the actual location and the hypothetical point is input in the input layer, and the actual N-value or hypothetical N-value is output in the said output layer.
5 . The N-value prediction apparatus of claim 4 , wherein a learning through the artificial intelligence learning structure selects one technique with the minimum error value with the application of the Mean Absolute Percentage Error (MAPE).
6 . The N-value prediction apparatus of claim 1 , wherein the N-value prediction result calculation unit outputs a predicted N-value derived from the input of a location information of the prediction target point and an actual N-value at a ground corresponding to the prediction target point by using the N-value prediction model.
7 . An N-value prediction method, comprising;
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;
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
predicting an N-value at an arbitrary prediction target point by using an N-value prediction model generated in learning by artificial intelligence,
wherein in generating at least one of hypothetical N-values, circular augmentation where multiple hypothetical points are set up radially in a certain distance centering the actual location is used,
wherein the multiple hypothetical points consist of 8 points in a radius of 0.5 m, 12 points in a radius of 1 m, and 16 points in a radius of 2 m centering around the actual location.
8 . The N-value prediction apparatus of claim 7 , wherein the ground characteristic data includes a location information and a soil information corresponding to the actual location and the hypothetical point.
9 . The N-value prediction apparatus of claim 8 , wherein the location information includes at least one of latitude, longitude, altitude, depth, and an absolute position based on depth and altitude.