Virtual sample generation method based on interpolation algorithm

Abstract

The invention discloses a virtual sample generation method based on an interpolation algorithm, which expands the sample size under the condition of unbalanced and incomplete samples and improves thesoft measurement modeling precision of a purified terephthalic acid production device. According to the invention, the projection of a high-dimensional original sample in a low-dimensional space is obtained by using a multi-dimensional scale analysis algorithm, a virtual sample is generated in a sample sparse region according to an interpolation algorithm, and finally, the value of the virtual sample in the original sample space is obtained by constructing an extreme learning machine neural network, so the virtual sample generation method is formed. According to the invention, the neural network is trained by expanding the sample set, and the precision and stability of the soft measurement model can be improved. The virtual sample generation method based on the interpolation algorithm is easy to use and obvious in effect, has excellent generalization performance and good stability, and can be widely applied to small sample modeling in the chemical production process.

Description

technical field [0001] The invention relates to the technical field of purified terephthalic acid production, in particular to a virtual sample generation method based on an interpolation algorithm. Background technique [0002] As data-driven methods are widely used in the modern process industry to build soft-sensing models, many algorithms that utilize collected data sets to learn data trends have been proposed. Sufficient effective samples and uniform sample distribution are two key requirements for building accurate data-driven models. Sufficient effective samples provide a guarantee for improving the accuracy and robustness of the soft sensor model. In the modern process industry, the main reasons for the difficulty in obtaining valid samples are: 1. Due to the stable process and low volatility of the process industry, it is difficult to collect a large amount of representative data; Due to the characteristics of linear and random noise, it is difficult to extract a ...

AI Technical Summary

Problems solved by technology

However, the diffusion function and diffusion coefficient cannot be determined efficiently and do not take into account the asymmetry of real samples

Images