Soft sensor method for least squares support vector machine based on distributed parallel local optimization parameters

Abstract

The invention discloses a least squares support vector machine soft sensor method based on distributed parallel local optimization parameters. First, the least squares support vector machine is modeled to obtain a locally optimized least squares support vector machine model, and then the The samples in the training set are evenly divided into blocks, and the samples in each block are marked with data block labels, and the training samples with the same label are classified into one data block, and then the locally optimized LSSVM distribution is used on multiple computers to parallelize each The training data in each data block is modeled and the test set data is predicted, and the average of the prediction results of multiple data blocks is taken as the final prediction result. This method not only guarantees the randomness of the training sample division, but also guarantees the prediction accuracy. At the same time, it performs modeling and prediction on the test set in parallel in each data block, which greatly reduces the computational cost and makes the local optimization based The parameterized LSSVM soft sensor algorithm can also be applied on large-scale data sets.

Description

technical field [0001] The invention belongs to the field of industrial process prediction and control, and in particular relates to a least square support vector machine soft sensor method based on distributed parallel local optimization parameters. Background technique [0002] Soft sensor modeling technology means that in the actual industrial production process, some process variables and quality variables are difficult to use sensors for direct measurement or the cost of measurement is too high. Therefore, people often use some process variables that are easier to measure and build mathematical models. The methods to estimate those difficult-to-measure process or quality variables can control product quality well and improve production efficiency. [0003] Among the commonly used soft sensor mathematical models, the least squares support vector machine model (hereinafter referred to as LSSVM) is widely used in the soft sensor of various industrial process variables due ...

AI Technical Summary

Problems solved by technology

[0005] However, the LSSVM algorithm based on parameter optimization has the disadvantage of slow calculation speed, that is, before using LSSVM to model, it is necessary to use an intelligent optimization algorithm to perform LSSVM modeling on various model parameters. Therefore, when there are too many training sample sets and the population size, When the number of iterations is too large, huge computing overhead will be generated. At present, the data used for industrial soft sensors is becoming larger and larger, gradually forming industrial big data problems. Therefore, the traditional LSSVM algorithm based on parameter optimization must be improved.

Images