Least squares support vector machine soft measurement method based on distributed parallel local optimization parameters

Abstract

The invention discloses a least squares support vector machine soft measurement method based on distributed parallel local optimization parameters. The least squares support vector machine soft measurement method comprises the following steps: firstly, modeling a least squares support vector machine to obtain a locally optimized least squares support vector machine model, then averagely dividing samples in a training set into blocks, labeling the samples in each block with data block labels, grouping the training samples with the same label into one data block, then modeling training data in each data block in a distributed parallel manner on multiple computers by using the locally optimized LSSVM, predicting test set data, and taking a mean value of prediction results of a plurality of data blocks as a final prediction result. By the least squares support vector machine soft measurement method, the randomness of training sample division is guaranteed, the prediction accuracy is guaranteed, and parallel modeling is performed in each data block and a test set is predicted, so that computational overhead is greatly reduced, and an LSSVM soft measurement algorithm based on the local optimization parameters can also be applied to a large-scale data set.

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