Multiple Thermal Parameter Inversion for Concrete Dams Using an Integrated Surrogate Model

An efficient and accurate method for concrete thermal parameter inversion is essential to guarantee the reliable prompt analysis results of dams. Traditional methods either suffer from low efficiency or are limited in accuracy. Thus, this paper presents a multiple based on an integrated surrogate model (ISM) Jaya algorithm. This replaces finite element with ISM incorporating three machine learning algorithms, Kriging, support vector regression (SVR), radial basis function (RBF), describe mapping relationship between parameters structure temperature responses. The input datasets training testing generated by uniform design approach. Subsequently, simple global optimization algorithm, Jaya, used identify minimizing error calculated monitored temperatures. effectiveness practicality verified applying data two strength grades dam. verification indicate that proposed approach can obtain more than above individual models. Compared these models, errors using reduced 8.45%, 3.93% 20.85%, respectively C35 concrete, 6.53%, 23.82% 44.43%, C40 concrete. Additionally, maintains powerful computational surrogate-based optimization, compared directly invert swarm intelligence improved about 111.7 times.