An Octane Loss Prediction Method Based on Particle Swarm Optimization Random Forest Parameters

A technology of particle swarm optimization and random forest algorithm, applied in neural learning methods, computer components, instruments, etc., can solve the problems of high complexity of the refining process, untimely response to process optimization, and high requirements for raw material analysis, etc., to achieve reduction Economic cost and time cost, avoiding the effect of overfitting, and solving the effect of comparability

Active Publication Date: 2022-07-01
KUNMING UNIV OF SCI & TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since the traditional chemical process modeling is mostly based on data association and mechanism modeling methods, but the actual refining process is highly complex, and the control variables have highly nonlinear and strong coupling relationships. The analysis requirements are high, and the response to process optimization is not timely, so the effect is not ideal

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
  • An Octane Loss Prediction Method Based on Particle Swarm Optimization Random Forest Parameters
  • An Octane Loss Prediction Method Based on Particle Swarm Optimization Random Forest Parameters
  • An Octane Loss Prediction Method Based on Particle Swarm Optimization Random Forest Parameters

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0028] Example 1: as figure 1 As shown, a method for predicting octane loss value based on particle swarm optimization random forest parameters, the method steps are as follows:

[0029] Step 1. Calculate the information gain value of the relevant features of the octane loss value, and delete the features with less impact on the octane value loss;

[0030] Step 2. Preprocess the remaining data of the features that have less impact on the octane number loss, and divide the preprocessed data into a training data set and a test data set;

[0031] Step 3. Use the training data set to train the random forest algorithm, obtain a training model, and use the test data set to verify the training model;

[0032] Step 4. Initialize the parameters of the particle swarm algorithm;

[0033] Step 5. The root mean square error of the verified random forest algorithm training model is used as the fitness function of the particle swarm algorithm, and the number of parameter decision trees n_e...

Embodiment 2

[0052] Embodiment 2: For a prediction method of octane loss value based on particle swarm optimization random forest parameters, the present invention provides the following experimental data process:

[0053] Step 1. Extract the characteristic data that affects the octane loss in the gasoline catalytic cracking process as the original data set, calculate the information gain value of each feature in the original data set, and delete the features with less impact on the octane number loss to avoid excessive generation. fitting problem. Specific steps are as follows:

[0054] 1.1. In the gasoline catalytic cracking process, the reason for the loss of octane is that in the desulfurization process, the hydrodesulfurization leads to the production of excessive olefins and causes the reaction consumption of octane. Therefore, the sensor of the hydrodesulfurization section in the gasoline catalytic cracking process per hour is selected. The collected data is used as the original da...

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 discloses a method for predicting an octane loss value based on a random forest parameter optimized by particle swarm optimization. 2. Preprocess the remaining data; step 3, use the training data set to train the random forest algorithm to obtain a training model; step 4, initialize the parameters of the particle swarm algorithm; step 5, take the root mean square error as the adaptation of the particle swarm algorithm degree function, the number of parameter decision trees in the training model and the depth of the tree are continuously solved by the particle swarm algorithm, and the optimal parameters are imported into the training model to obtain the optimal prediction model; Step 6, re-input the new test set to import The optimal prediction model is tested and the prediction result is obtained. The present invention can be effectively used for octane loss prediction.

Description

technical field [0001] The invention relates to an octane loss value prediction method based on particle swarm optimization random forest parameters, and belongs to the technical field of octane loss value prediction in a gasoline catalytic cracking process flow. Background technique [0002] In the gasoline catalytic cracking process, in order to meet the requirements of gasoline sulfur content in the new national standard environment, the requirements for desulfurization treatment in gasoline are further improved. However, excessive process operation during the desulfurization process will affect the content of octane number in gasoline. Octane number is the most important indicator to reflect the combustion performance of gasoline. Controlling the octane loss value in the process can effectively improve the economic benefits in production. Because traditional chemical process modeling is mostly based on data association and mechanism modeling, the actual refining process ...

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): G06K9/62G06N3/00G06N3/08
Inventor 杨春曦陈瑞韩世昌范升序李一鸣陈锐
Owner KUNMING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap