PH neutralization process model identification method based on Newton iteration algorithm

Abstract

The invention provides a PH neutralization process model identification method based on a Newton iteration algorithm, and the method comprises the following steps: 1) constructing a PH neutralization process fractional order Hammerstein CAR model, and obtaining a PH neutralization process identification model according to a constructed system model; and 2) constructing an identification process of a Newton iteration algorithm. The PH value neutralization process model identification method based on the Newton iteration algorithm has the beneficial effects that a parameter identification result carried out by using the PH value neutralization process model identification method based on the Newton iteration algorithm can show that the method is relatively high in identification precision and relatively small in output parameter estimation error; meanwhile, the identification method also shows that the identification method has good applicability to the PH neutralization process model.

Description

technical field [0001] The invention relates to the technical field of sewage treatment system identification, in particular to a method for identifying a fractional order model of a PH neutralization process based on a Newton iterative algorithm. Background technique [0002] Environmental protection and sustainable development are important links in the national development strategy. In the treatment of chemical industry and three wastes, the environmental protection department will focus on monitoring the pH value of industrial wastewater discharge. In the process of neutralizing the pH value of sewage treatment, it generally has a strong nonlinearity. In order to better analyze, predict and control the pH neutralization process, it is necessary to establish a system model for the pH neutralization process control, and at the same time identify the parameters of the established model. For this reason, many scholars have established integer-order models of the PH neutral...

AI Technical Summary

Problems solved by technology

[0003] When building a model for an actual system, the commonly used integer-order model is sometimes not accurate enough, resulting in insufficient identification accuracy

In addition, most of the nonlinear parts used are polynomials of known order, which cannot fully describe all nonlinear phenomena

In terms of identification methods, the least squares method is lacking in identification accuracy, and at the same time, data saturation is prone to occur when there are many identification parameters.

The convergence speed of the gradient iteration method is slow, and it is easy to cause the identification result to fall into a local optimum. When selecting the iteration step size, it is necessary to consider the divergence of the identification result.

Images