Prediction method for explosion characteristics of organic mixture based on support vector machine

Abstract

The invention relates to a prediction method for explosion characteristics of organic mixture based on a support vector machine, comprising the following steps: for the problem that the explosion characteristics of the organic mixture under the condition of different components and different proportion is predicted and industrial process design is optimized in industrial production, taking known component content and routine property experiment data of the organic mixture as input variable, taking corresponding explosion characteristics experiment data as output variable, effectively training and predicting inner quantitative relation of nonlinearity, non-determinacy and complexity between the input variable and the output variable with the strong machine learning algorithm support vector machine method so as to establish stable, high-efficiency support vector machine prediction model. The explosion characteristics of the other unknown mixture are predicted according to the established support vector machine model so that the method has advantages of high prediction precision, speed and convenience. According to the invention, the method can predict the explosion characteristics of the organic mixture under the condition of different components and different proportion, effectively solve the problem that the actual industry production is short of the experiment data of each mixture, and has good application prospect in industrial process design, fire protection and explosion protection.

Description

technical field [0001] The invention relates to the technical field of organic chemical industry, in particular to a method for predicting the explosion characteristics of organic mixtures, in particular to a method for predicting the explosion characteristics of organic mixtures based on support vector machines. Background technique [0002] With the continuous development of the chemical industry and the diversification of chemical products, various chemical products have been widely used in various sectors of the national economy. At present, more than 30 million chemical substances have been discovered and synthesized, among which there are more than 80,000 chemical products used by human beings daily, and this number is increasing at a rate of nearly 1,000 each year. Among the many chemical substances, there are many substances with dangerous characteristics such as flammability and explosion, and there are possibilities of fire and explosion accidents in the process of...

AI Technical Summary

Problems solved by technology

[0013] The present invention aims at the problems that the explosion characteristics of organic mixtures with different components and different proportions need to be determined through experiments in the actual industrial production, which has a long cycle, high risk and high cost, and proposes a method based on existing known organic compounds. Based on the combustion and explosion characteristics of the mixture, with a small amount of conventional physical property experiments, the prediction method of the combustion and explosion characteristics of organic mixtures based on support vector machines can be obtained without large-scale combustion and explosion experiments

Images