Batch process product design method based on PPLS model

Abstract

The invention discloses a batch process product design method based on a PPLS model. In historical running data, input condition data is X, a product quality parameter is Y, an expected product quality parameter is ydes, and an input condition is xpred; the batch process product design method based on the PPLS model comprises the steps of selecting an input condition matrix X and an output matrixY which are used for establishing the model according to the historical running data and running experience, and establishing the PPLS model; based on conditional probability distribution of product quality with respect to a principal element, taking the expected product quality parameter as the point estimation of the product quality, and obtaining the point estimation of the principal element; based on conditional probability distribution of the principal element with respect to an operation trajectory, equalizing the estimation of the principal element to the point estimation obtained fromthe product quality, and estimating the operation trajectory and the operation space. The method has the advantages that a product can be designed simply by utilizing the existing historical data in the process.

Description

technical field [0001] The invention relates to the field of batch process process condition design, in particular to a batch process product design method based on PPLS model. Background technique [0002] Product quality in a batch process depends on initial conditions, operating trajectories, etc., which are collectively referred to as input conditions. Different input conditions correspond to different product qualities. Therefore, given the product quality, how to obtain the input conditions has always been the focus of attention in the business community and academia. The process of how to design input conditions to obtain ideal product quality is called product design. [0003] At present, in the actual industrial process, the commonly used product design methods mainly include three-level orthogonal experiment method and response surface method. Among them, the three-level orthogonal experiment method designs different changes of a single variable while keeping ot...

AI Technical Summary

Problems solved by technology

[0006] However, the PLS model is a projection model, and there is a lot of randomness in the actual industrial process data. Obviously, the probability model is more suitable for describing the correlation between input conditions and product quality

Images