Paper-pulp washing process and optimizing control

Abstract

A pulp-dressing technology and an optimizating method in pulp-dressing are provided in the invention. It is a multiple-stage countercurrent washing technology .Plasm becomes cleaner from the first stage to the third stage while the washer solvent becomes less concentrated .We adopt double-aim optimazing method which includes an elementary step by orthogonal and diluting factor method; found a dynamic mathematic model by NN method; produce a static model sample collection. Advantages are cleanness and optimal black liquor desity.

Description

technical field [0001] The invention relates to a pulp washing process and an optimization control method, which are used in a papermaking production line and belong to the technical field of papermaking control systems. Background technique [0002] The purpose of pulp washing is to separate the pulp from the cooking liquid as much as possible. The main technical indicators are the degree of washing (residual alkali) and the density of black liquor (degrees Baume). The degree of washing requires that the amount of residual alkali be controlled below 0.05g / L. The black liquor density requires Baume to be controlled within the range of 7-9. In the actual operation process, residual alkali and Baume are two contradictory parameters. To improve the cleaning degree, it is required to increase the washing water, and such a In the future, the Baume degree will be reduced, thereby increasing the steam volume in the evaporation section of the alkali recovery workshop; on the contrar...

AI Technical Summary

Problems solved by technology

In many paper mills, the concentration of black liquor is manually measured off-line after a certain period of time, and the control of black liquor concentration is not very accurate. Because foreign pulp washing control systems are often matched with beating equipment and technological processes, most domestic paper mills are short of funds. , there is no ability to import a full set of foreign equipment, coupled with the variety of domestic pulp types, foreign systems may not be applicable

[0004] In the past neural network identification process of steady-state models, a large amount of steady-state data is often used for direct identification. This method requires continuous perturbation of the process to obtain more data to form a sample set. The sample set generally has Hundreds of sets of data can fully express the steady-state characteristics of the actual process. Such frequent application of perturbation is generally not allowed in industrial production processes, which will seriously affect production.

Images