Mixture moisture detection value correction method based on feedback adaptive prediction model

Abstract

The invention relates to a mixture moisture detection value correction method based on a feedback adaptive prediction model. The method comprises the following steps: 1, collecting the moisture content information of a raw material; 2, calculating the moisture content of a mixture through an ingredient calculation system; and 3, screening model input quantities through an MIV algorithm, training a BP neural network through a collected sample, enabling the trained adaptive neural network system to predict and correct the moisture value of the mixture, comparing the predicted value and the target value regularly, and if the deviation exceeds a certain value, carrying out feedback correction. According to the technical scheme, the research range of the sintering mixture detection process in the past is broken through, and the feedback adaptive prediction model is innovatively applied to predict and correct the moisture value of the mixture after training is conducted according to the known sintering mixture, so the detection precision can be greatly improved; and in addition, feedforward control can be conveniently combined to avoid the influence of 3-5 minutes of delay detection on quality control, and an advanced adjustment process is provided.

Description

technical field [0001] The invention relates to a dredging component, in particular to a method for correcting the moisture detection value of a mixture based on a feedback self-adaptive prediction model, and belongs to the technical field of moisture and humidity detection. Background technique [0002] Sintering mixture moisture is one of the parameters that need to be strictly controlled in sintering production, which directly affects the quality and quality of sintering production. The moisture content of the sinter mixture is too low, the cohesive force between the particles is small, and the ground ore powder and other additives cannot be agglomerated into pellets with a certain particle size. Efficiency decreases in . If the moisture content of the sintering mixture is too high, although the granules have good ball forming performance, the granulation will stick due to the high viscosity, and too much free water will be precipitated in the cold material layer of the ...

AI Technical Summary

Problems solved by technology

In terms of detection accuracy, it seems to have reached a bottleneck period, and the biggest problem with this detection technology is that after the sintering raw materials are mixed once, water is added, and then the moisture detector is used to detect real-time moisture during the process of transporting to the next process on the belt conveyor system. There is often a time delay of 3-5 minutes. At this time, a considerable part of the moisture content of the mixed material is not within the standard range, which may easily cause the moisture data to get out of control.

[0004] With the development of neural network and intelligent detection technology, more and more factories and enterprises have improved their production technology, especially in the sintering process site, the key sintering process procedures such as the detection and control of the moisture content of the mixture, the temperature control of the sintering end point, etc. Extensive attention and research, but due to the complex working conditions of the sintering site and the great influence of external factors, the improvement of these technologies often only stays in the theoretical research stage, and the application in the actual site cannot completely solve the existing technical problems

Images