An Online Adaptive Modeling Method for Hydrocyanic Acid Production Process

Abstract

The invention discloses an online adaptive modeling method for a hydrocyanic acid production process. The online adaptive modeling method is characterized by comprising steps of determining input variable subspace of the hydrocyanic acid production process; acquiring data of the hydrocyanic acid production process; preprocessing the data of the hydrocyanic acid production process to obtain valid data which are affected by noise to the minimum extent and can effectively reflect actual characteristics of the production process; and performing modeling for the preprocessed data (X''MT, Y'') by an unscented Kalman neural network to obtain a precise hydrocyanic acid conversion rate model. The online adaptive modeling method has the advantages that the valid data which are affected by the noise to the minimum extent and can effectively reflect the actual characteristics of the production process can be obtained, the hydrocyanic acid production process can be precisely approximated by the aid of the unscented Kalman neural network in the input variable subspace, and the difficulty in modeling for complex nonlinear dynamic hydrocyanic acid production processes is solved effectively.

Description

technical field [0001] The invention belongs to the intelligent information processing technology in the production process of hydrocyanic acid, in particular to an online self-adaptive modeling method of the production process of hydrocyanic acid based on the subspace approximation of the unscented Kalman neural network. Background technique [0002] The raw material gases for the production of hydrocyanic acid (HCN) are ammonia, natural gas and air. The three raw materials are purified, mixed, oxidized and pickled in four stages to obtain pure HCN gas. The HCN industrial process is complex, with many process parameters. HCN production equipment is in contact with the air, and is affected by uncertain factors such as temperature, humidity, equipment aging, and raw material batches. It is a typical complex nonlinear dynamic characteristic chemical system. How to establish an accurate and reliable HCN industrial process model is the basis and premise of improving the conversi...

AI Technical Summary

Problems solved by technology

[0005] For the dynamic and nonlinear complex characteristics of hydrocyanic acid production process, traditional methods cannot effectively establish its high-precision model

In order to solve the above problems, the present invention proposes a kind of online adaptive modeling method based on the unscented Kalman neural network subspace approximation of the hydrocyanic acid production process, by adopting the unscented Kalman neural network in the input variable subspace to accurately approximate Hydrocyanic acid production process, effectively solving the modeling problem of complex nonlinear dynamic production process of hydrocyanic acid

Images