Pile foundation integrity classification and recognition method based on convolutional neural network

Abstract

The invention discloses a pile foundation integrity classification and recognition method based on a convolutional neural network. The method comprises the steps: detecting a pile foundation through low-strain collection equipment, and attaching class labels to the pile foundation to construct a standby data set; scaling and normalization preprocessing are carried out on the images in the data set; performing data expansion on the preprocessed data set; dividing into a training set and a test set according to a preset proportion; inputting the training set into the constructed model for classification training, and optimizing related parameters until a globally optimal solution is obtained to obtain a required convolutional neural network model; and inputting the test set into the convolutional neural network model for verification, outputting an recognition result and performing evaluation. According to the invention, the feature extraction capability of the convolutional neural network is utilized to identify the low-strain wave pattern image, so that the problems of high cost and high subjectivity of manual analysis and detection are solved; meanwhile, the calculation capabilitybased on the neural network has relatively high practicability, and has huge practical significance and value for pile foundation integrity detection.

Description

technical field [0001] The invention relates to the field of computer deep learning, in particular to a pile foundation integrity classification and recognition method based on a convolutional neural network. Background technique [0002] With the development of my country's engineering construction industry, pile foundation engineering is the most widely used foundation form in construction engineering. Pile foundation is a concealed project, and it is one of the effective methods to deal with weak foundations and reduce building settlements. It is widely used in urban high-rise buildings, factories, bridges and other important projects that are closely related to people's lives. The quality of the pile foundation project is directly related to the structure and stress safety of the building, and even more related to people's life, property and social security. The pile foundation project is located underground or underwater, which makes the quality inspection and pile con...

AI Technical Summary

Problems solved by technology

At present, the low-strain dynamic pile testing method is generally used to detect its integrity. The low-strain stress wave reflection method occupies a dominant position in the integrity detection of pile foundations due to its fast, convenient and economical advantages. However, the accuracy of dynamic pile testing is largely It mainly depends on the experience of the testing personnel, high testing cost, low efficiency, low degree of automation and strong subjectivity, etc.

[0003] The defect identification of pile foundation is a nonlinear problem, it is difficult to express it with an accurate function

Images