Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Model Method for Optimizing Cement Raw Meal Quality

A technology for cement raw meal and optimization model, applied in the field of industrial Internet, can solve the problems of incomplete parameters, poor raw meal cycle and quality stability, difficulty, etc., and achieve the effect of high parameter adjustment efficiency, strong universality and stable quality.

Active Publication Date: 2022-07-12
杭州玖欣物联科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] 1. According to the detection data of raw material composition, the method of manually estimating the ratio of raw materials is used to control and adjust the quality of raw materials. This method relies on experienced operators to estimate the final ratio of raw materials based on experience, so the control of raw material quality Lack of theoretical guidance and unified adjustment standards. Due to the great difference in labor experience and level, which is seriously affected by human subjectivity, production experience cannot be accumulated, resulting in poor cycle and quality stability of raw materials.
[0005] 2. According to the raw material composition data and the raw material quality rate value standard, the raw material ratio parameters are calculated by the formula, and the operator produces according to the calculated ratio. This method will become very difficult after the raw material reaches 4 or more. Difficult or even impossible to calculate, and the method considers incomplete parameters, only uses raw material composition data, does not consider the influence of other parameters in the production process, and the optimization control of quality is not precise enough

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Model Method for Optimizing Cement Raw Meal Quality

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention is a model method for optimizing the quality of cement raw meal. Based on the historical data of various parameters in the raw meal grinding process, an optimization model for the quality of cement raw meal is constructed, and the optimal parameter value is given through the model, so as to finally achieve qualified raw meal quality, Stablize. It consists of two stages, building a raw meal quality prediction model and building a raw meal ratio optimization model.

[0038] The construction of a raw meal quality prediction model includes the following steps:

[0039] S11. Collect the raw data of each parameter in the cement raw meal production process in the past year, including the proportion of CaO in the incoming limestone, the proportion of MgO in the incoming limestone, the moisture content of the incoming limestone, the particle size of the incoming limestone out of the crusher, and each auxiliary material ( SiO in fine sandstone, bauxite, iro...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a model method for optimizing the quality of cement raw meal, which includes the following steps: collecting data of various parameters involved in the production process of cement raw meal; integrating the collected data, sorting out target variables according to target problems, and inputting variables; Perform exploratory analysis on input variables and target variables to obtain available input variables; carry out corresponding feature engineering around each target variable to obtain different model-in features; select different regression algorithm models for each target variable to construct multiple models The rate value quality prediction model is continuously iteratively trained according to the evaluation effect; after the optimal evaluation model is deployed online, the model is updated with new data; Adjustment variables; find the optimal parameters through the grid search algorithm for the adjustable variables; the obtained optimal parameter values ​​are used as the output of the model, which can achieve the purpose of optimizing the quality of cement raw meal, and has strong universality.

Description

【Technical field】 [0001] The invention relates to the technical field of the industrial Internet, in particular to a model method for optimizing the quality of cement raw meal. 【Background technique】 [0002] In the cement production process, there are multi-variable, non-linear, pure lag, large inertia, and strong time-varying situations. There are large differences in human operations, lag in operation, it is difficult to achieve refined operation, and the system operation fluctuates greatly (quality, output). To ensure the continuous, stable and economical operation of the system, the quality of raw meal fluctuates greatly. [0003] At present, there are two main ways to control the quality of domestic cement raw meal: [0004] 1. According to the test data of raw material components, the method of manually estimating the ratio of each raw material is used to control and adjust the quality of raw meal. This method relies on experienced operators to estimate the final rat...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F30/27G06N20/00G06N20/10G06N20/20
CPCG06F30/27G06N20/10G06N20/00G06N20/20Y02P90/30
Inventor 韩倩倩
Owner 杭州玖欣物联科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products