Prediction of Sample Curing Index Data Based on Cubic Hermite Interpolation

Abstract

Cubic Hermite interpolation is a fundamental and widely-used interpolation algorithm in numerical computation. This method can generate functional expressions based on discrete sampling data to realize data inference and trend prediction. Distinct from other interpolation methods, Hermite interpolation matches not only the function values at interpolation nodes but also the first derivatives at these nodes. This characteristic enables the algorithm to fit the original function more accurately, yielding smoother and more precise curves especially in the vicinity of discrete nodes. In this paper, cubic Hermite interpolation is adopted to predict the performance evolution of civil engineering materials during curing. The test sample contains eleven performance indicators measured at weekly intervals, and the material properties of intermediate weeks are calculated via interpolation. Experimental results verify that cubic Hermite interpolation is applicable to the research problem in this paper and can effectively predict unknown sample indicators from limited measured data.