Ethylene cracking carbon emission accounting method, device and system and storage medium

Abstract

The invention provides an ethylene cracking carbon emission accounting method, device and system and a storage medium, and belongs to the technical field of energy conservation and carbon reduction, and the method comprises the steps: extracting characteristic variables from real-time process data of an ethylene cracking device; the characteristic variables are input into a target prediction model, a predicted value of the carbon emission in the ethylene cracking and scorching process is generated, the target prediction model is a neural network model after training and parameter optimization, and samples adopted during training are constructed by utilizing the characteristic variables which are determined from historical process data of an ethylene cracking device and reflect the carbon emission. Compared with a traditional mode based on a carbon dioxide and carbon monoxide concentration on-line monitoring hardware instrument, a target prediction model based on a neural network is constructed through a data driving mode, the carbon emission in the ethylene cracking and scorching process is predicted, and the carbon emission is combined with electric power and thermal power indirect carbon emission determined through an emission factor method, so that the carbon emission is predicted. And accurate and reliable detection of the carbon emission of the ethylene cracking device is realized.

Images