Life Prediction of Support Structures Based on Multi-Layer Perceptron (MLP) Algorithm

With the development of infrastructure construction and intelligent design, the life prediction of engineering structures has gained increasing attention. This paper proposes a life prediction method for support structures based on a Multi-Layer Perceptron (MLP) neural network, taking the pile foundation system of a sewage treatment plant as the research object. A high-dimensional dataset is constructed by collecting ten types of feature variables including geological characteristics, load parameters, pile type ID, concrete grade, and construction errors. The MLP model is then built and trained to perform life regression prediction. Experimental results show that the proposed method significantly outperforms traditional Support Vector Machine (SVM) and Random Forest (RF) methods in terms of regression accuracy, generalization ability, and stability, achieving a coefficient of determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) of 0.92. This study provides an effective algorithmic support for intelligent life assessment of support structures in civil engineering, showing promising engineering applications.