2020(华为杯)研究生数学建模B题汽油辛烷值建模优秀论文我要分享

2020 (Huawei cup) Postgraduate mathematical modeling title b excellent paper on gasoline octane numb

华为杯 研究生数学建模 B题 汽油辛烷值 论文

关注次数: 981

下载次数: 101

文件大小: 1.11MB

代码分类: 数学建模

开发平台: matlab

下载需要积分: 20积分

版权声明:如果侵犯了您的权益请与我们联系,我们将在24小时内删除。

代码描述

中文说明:

目标

依据从催化裂化汽油精制装置采集的325个数据样本(每个数据样本都有354个操作变量),通过数据挖掘技术来建立汽油辛烷值(RON)损失的预测模型,并给出每个样本的优化操作条件,在保证汽油产品脱硫效果(欧六和国六标准均为不大于10μg/g,但为了给企业装置操作留有空间,本次建模要求产品硫含量不大于5μg/g)的前提下,尽量降低汽油辛烷值损失在30%以上。

 

问题

数据处理:请参考近4年的工业数据(见附件一“325个数据样本数据.xlsx”)的预处理结果,依“样本确定方法”(附件二)对285号和313号数据样本进行预处理(原始数据见附件三“285号和313号样本原始数据.xlsx”)并将处理后的数据分别加入到附件一中相应的样本号中,供下面研究使用。

寻找建模主要变量: 由于催化裂化汽油精制过程是连续的,虽然操作变量每3 分钟就采样一次,但辛烷值(因变量)的测量比较麻烦,一周仅2次无法对应。但根据实际情况可以认为辛烷值的测量值是测量时刻前两小时内操作变量的综合效果,因此预处理中取操作变量两小时内的平均值与辛烷值的测量值对应。这样产生了325个样本(见附件一)。

建立降低辛烷值损失模型涉及包括7个原料性质、2个待生吸附剂性质、2个再生吸附剂性质、2个产品性质等变量以及另外354个操作变量(共计367个变量),工程技术应用中经常使用先降维后建模的方法,这有利于忽略次要因素,发现并分析影响模型的主要变量与因素。因此,请你们根据提供的325个样本数据(见附件一),通过降维的方法从367个操作变量中筛选出建模主要变量,使之尽可能具有代表性、独立性(为了工程应用方便,建议降维后的主要变量在30个以下),并请详细说明建模主要变量的筛选过程及其合理性。(提示:请考虑将原料的辛烷值作为建模变量之一)。

建立辛烷值(RON)损失预测模型:采用上述样本和建模主要变量,通过数据挖掘技术建立辛烷值(RON)损失预测模型,并进行模型验证。

主要变量操作方案的优化:要求在保证产品硫含量不大于5μg/g的前提下,利用你们的模型获得325个数据样本(见附件四“325个数据样本数据.xlsx”)中,辛烷值(RON)损失降幅大于30%的样本对应的主要变量优化后的操作条件(优化过程中原料、待生吸附剂、再生吸附剂的性质保持不变,以它们在样本中的数据为准)。

模型的可视化展示:工业装置为了平稳生产,优化后的主要操作变量(即:问题2中的主要变量)往往只能逐步调整到位,请你们对133号样本(原料性质、待生吸附剂和再生吸附剂的性质数据保持不变,以样本中的数据为准),以图形展示其主要操作变量优化调整过程中对应的汽油辛烷值和硫含量的变化轨迹。(各主要操作变量每次允许调整幅度值Δ见附件四“354个操作变量信息.xlsx”)。


English Description:

target
Based on 325 data samples collected from FCC gasoline refining unit (each data sample has 354 operating variables), the prediction model of gasoline octane number (RON) loss is established through data mining technology, and the optimal operating conditions of each sample are given to ensure the desulfurization effect of gasoline products (European and national standards are no more than 10) μ G / g, but in order to leave space for enterprise plant operation, the sulfur content of the product is required to be no more than 5 in this modeling μ G / g), try to reduce the loss of gasoline octane number by more than 30%.
Question
Data processing: please refer to the preprocessing results of industrial data in recent 4 years (see Annex I "325 data samples. Xlsx"), preprocess data samples 285 and 313 according to the "sample determination method" (Annex II) (see Annex III "285 and 313 sample original data. Xlsx") and add the processed data to the corresponding sample numbers in Annex I respectively, For the following research.
Find the main variables for modeling: because the FCC gasoline refining process is continuous, although the operating variables are sampled every 3 minutes, the measurement of octane number (dependent variable) is troublesome, and it is impossible to correspond only twice a week. However, according to the actual situation, it can be considered that the measured value of octane number is the comprehensive effect of operating variables within two hours before the measurement time. Therefore, in the pretreatment, the average value of operating variables within two hours corresponds to the measured value of octane number. This resulted in 325 samples (see Annex I).
The establishment of octane number reduction loss model involves 7 variables such as raw material properties, 2 Properties of spent adsorbents, 2 Properties of regenerated adsorbents, 2 Properties of products, and another 354 operating variables (367 variables in total). The method of dimension reduction before modeling is often used in engineering technology applications, which is conducive to ignoring secondary factors, Find and analyze the main variables and factors affecting the model. Therefore, according to the 325 sample data provided (see Annex I), please screen out the main modeling variables from 367 operating variables by dimensionality reduction method to make them as representative and independent as possible (for the convenience of engineering application, it is recommended that the main variables after dimensionality reduction be less than 30), and please explain in detail the screening process and rationality of the main modeling variables( Tip: please consider the octane number of raw materials as one of the modeling variables).
Establish an octane number (RON) loss prediction model: using the above samples and modeling main variables, establish an octane number (RON) loss prediction model through data mining technology and verify the model.
Optimization of main variable operation scheme: it is required to ensure that the sulfur content of the product is not greater than 5 μ On the premise of g / g, use your model to obtain the main variables corresponding to the 325 data samples (see Annex IV "325 data samples. Xlsx") in which the octane number (RON) loss decreases by more than 30% (the properties of raw materials, spent adsorbents and regenerated adsorbents remain unchanged during the optimization process, subject to their data in the samples).
Visual display of the model: in order to stabilize the production of industrial units, the optimized main operating variables (i.e. the main variables in question 2) can only be gradually adjusted in place. Please keep the data of sample 133 (raw material properties, properties of spent adsorbent and regenerated adsorbent unchanged, subject to the data in the sample), The corresponding changes of gasoline octane number and sulfur content during the optimization and adjustment of its main operating variables are displayed graphically( Each main operating variable is allowed to adjust the amplitude value each time Δ See Annex IV "354 operation variable information. Xlsx").


代码预览

B20105330033.pdf