WO2021036546A1 - 基于有偏估计的近红外定量分析模型构建方法 - Google Patents
基于有偏估计的近红外定量分析模型构建方法 Download PDFInfo
- Publication number
- WO2021036546A1 WO2021036546A1 PCT/CN2020/101622 CN2020101622W WO2021036546A1 WO 2021036546 A1 WO2021036546 A1 WO 2021036546A1 CN 2020101622 W CN2020101622 W CN 2020101622W WO 2021036546 A1 WO2021036546 A1 WO 2021036546A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- samples
- model
- sample
- biased
- init
- Prior art date
Links
- 238000004445 quantitative analysis Methods 0.000 title claims abstract description 14
- 238000010276 construction Methods 0.000 title claims abstract description 5
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 10
- 238000012360 testing method Methods 0.000 claims abstract description 10
- 238000005457 optimization Methods 0.000 claims description 7
- 230000002068 genetic effect Effects 0.000 claims description 5
- 238000002790 cross-validation Methods 0.000 claims description 2
- 238000007781 pre-processing Methods 0.000 claims description 2
- 230000035945 sensitivity Effects 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000000694 effects Effects 0.000 abstract description 2
- 239000003921 oil Substances 0.000 description 15
- 238000002156 mixing Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 5
- TVMXDCGIABBOFY-UHFFFAOYSA-N octane Chemical compound CCCCCCCC TVMXDCGIABBOFY-UHFFFAOYSA-N 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010986 on-line near-infrared spectroscopy Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Definitions
- the invention belongs to the technical field of oil product detection, and relates to an online detection method of oil product attributes in the process of oil product blending quality feedback control.
- Gasoline blending is the last process before finished gasoline leaves the factory, and it is an important link to realize economic benefits.
- timely and accurate online detection of oil properties is required; this is an important means to ensure product qualification and reduce excess quality.
- the key attributes of gasoline such as research octane number, motor octane number, etc.
- On-line near-infrared technology can realize real-time detection of key attributes based on the near-infrared spectrum of oil products. Therefore, the feedback control of oil quality based on this technology is one of the hot issues in current research.
- the selection of training samples for near-infrared quantitative analysis models often first obtains the product label based on production scheduling information, initially selects modeling samples that are closely related to the target working conditions, and then repeats trial and error, and finally determines the construction Model samples.
- the analysis model is used in the quality feedback control system, in order to avoid the unqualified product quality caused by the high model prediction value, the model expert needs to correct the prediction model in real time. Because the existing methods do not analyze the selection of modeling samples from the inherent perspective of the data, and fail to consider the impact of prediction deviations on production from the nature of modeling, it is difficult to obtain prediction results suitable for quality feedback control.
- the purpose of the present invention is to provide a method for constructing a near-infrared quantitative analysis model based on biased estimation.
- the beneficial effect of the present invention is to consider the impact of prediction deviation on production from the essence of modeling, thereby obtaining prediction results suitable for quality feedback control. .
- O init ⁇ X ⁇ R N ⁇ m , Y ⁇ R N ⁇ 1 ⁇ O
- N represents the number of samples, m represents the sample dimension
- the method of preliminary selection of training samples select y low ⁇ y ⁇ y up from the data set O init to form a test set sample, where,
- stdY is the standard deviation of the attribute value in O init
- k is the sensitivity parameter
- y obj is the factory index of the oil product attribute of the target order.
- n is the total number of samples in the O init data set
- nl is the number of modeling samples selected from O init
- w is the weighting factor
- S23 Use the genetic algorithm to iteratively select part of the samples to form a training set, and calculate the fitness function value corresponding to the training set.
- S24 Select the solution corresponding to the minimum fitness function value to form the optimal training set O opt .
- the weighting factor w in step S21 is:
- ⁇ is a positional parameter.
- step S22 the samples are encoded in binary.
- step S23 a genetic algorithm is used for optimization.
- the regression model is established using the biased minimum maximum probability machine, and the near-infrared spectrum data X and attribute data Y are processed as follows:
- ⁇ is the supremum of the given absolute error.
- Step S3 is the most biased
- the small maximum probability machine model is:
- ⁇ is the lower bound of the correct classification probability of u class
- ⁇ 0 is the lower bound of the given correct classification probability of v class.
- y i is the reference attribute
- n is the number of samples.
- the method of the present invention first uses the biased minimum maximum probability regression algorithm to construct the fitness function, completes the selection of training samples, reduces the number of modeling samples, and can maximize the representativeness of the selected samples to the target operating conditions; reuse Biased minimum and maximum probability regression constructs a near-infrared quantitative analysis model to provide probabilistic biased prediction output.
- the advantage of this method is that genetic algorithms can be used to select training samples, which greatly improves the quality and efficiency of establishing a near-infrared quantitative analysis model; the prediction model is established through biased minimum and maximum probability regression, which can better process non-Gaussian data.
- the given probability biased prediction value can greatly improve the control effect of oil blending quality feedback control.
- Figure 1 is a preferred training sample.
- the process mainly includes three parts: component oil and blended product oil attribute detection, real-time optimization of blending formula, and pipeline valve proportional control.
- component oil pipeline valve The opening degree is determined by the blending formula, and the optimization server performs online optimization based on the real-time properties of blended oils and component oils.
- the gasoline data of the blended product is collected from monitoring data during the gasoline pipeline blending process of a domestic refinery.
- the wavelength range of the gasoline near-infrared spectrum is 1100nm-1300nm, and the wavelength accuracy is 1nm; the reference value of the sample research method octane number adopts ASTM Standard motor machine measurement.
- the historical data set contains 350 sets of samples, and the test set contains 250 sets of samples.
- Step 2 Use the method in S21 to construct the fitness function, and use the method in S22-S24 to use the GA algorithm to optimize the training samples. Finally, 120 training samples were selected.
- the fourth step use the above training samples and model parameters to establish a biased minimum and maximum probability regression model for the online process.
- the gradient descent method is used to solve the minimum maximum probability model, and the regression equation obtained after the solution is:
Abstract
一种基于有偏估计的近红外定量分析模型构建方法,首先从历史数据集中初选训练样本;根据目标工况,从样本集中选择合适的建模样本组成子集,优选后的样本作为近红外定量分析模型的建模样本;利用建模样本建立属性与近红外光谱之间的有偏最小最大概率回归模型:通过误差下确界的选取使模型的预测偏差以最大概率在期望的方向上;将测试集近红外光谱带入模型进行预测,根据输出的预测值与参考值计算出模型对应的均方根误差对比,选择最佳模型参数。有益效果是能够从建模本质考虑预测偏差对生产的影响,从而获取适合用于质量反馈控制的预测结果。
Description
本发明属于油品检测技术领域,涉及油品调合质量反馈控制过程中油品属性的在线检测方法。
汽油调合是成品汽油出厂前的最后一道工序,是实现经济效益的重要环节。在油品质量反馈控制系统中,需要对油品属性进行及时、准确的在线检测;这是保证产品合格、减少质量过剩的重要手段。目前,汽油的关键属性,如研究法辛烷值、马达法辛烷值等主要通过人工采样并利用ASTM标准的马达机才可获取;此类方法成本过高、检测周期长不宜用于在线质量反馈控制中。在线近红外技术可根据油品的近红外光谱实现对关键属性的实时检测,因此基于该技术的油品质量反馈控制是当前研究的热点问题之一。当前,在油品调合领域,近红外定量分析模型训练样本的选择往往先根据生产调度信息获知产品标号,初选出与目标工况密切相关的建模样本,然后反复试差,最终确定建模样本。当分析模型用于质量反馈控制系统中时,为了避免因模型预测值偏高而引发的产品质量不合格,需要模型专家实时校正预测模型。由于现有方法没有从数据内在角度分析建模样本的选择,且未能从建模本质考虑预测偏差对生产的影响,较难获取适合用于质量反馈控制的预测结果。
发明内容
本发明的目的在于提供基于有偏估计的近红外定量分析模型构建方法,本发明的有益效果是能从够建模本质考虑预测偏差对生产的影响,从而获取适合用于质量反馈控制的预测结果。
本发明所采用的技术方案是按照以下步骤进行:
S1:数据预处理:首先从历史数据集O中初选训练样本
O
init={X∈R
N×m,Y∈R
N×1}∈O
(N代表样本数,m代表样本维度);
训练样本进行初选的方法:从数据集O
init中选取y
low≤y≤y
up组成测试集小 样,式中,
其中,stdY为O
init中属性值的标准差,k为灵敏度参数,y
obj为目标定单的油品属性出厂指标。遍历O数据集,选择出满足y
low≤y≤y
up的所有样本组成O
init。
S2:训练样本优化选择:根据目标工况,从O
init样本集中选择合适的建模样本组成子集O
opt∈O
init,优选后的样本作为近红外定量分析模型的建模样本;S21:构造适应度函数
S22:采用二进制编码方法,对O
init数据集中的样本进行编码,若所述数据集中某个样本被选为建模样本,则其编码值为1,否则为0。
S23:利用遗传算法迭代选择部分样本构成训练集,并计算所述训练集对应的适应度函数值。S24:选择最小适应度函数值对应的解组成最优训练集O
opt。
步骤S21中的权重因子w为:
其中,γ为位置参数。
步骤S22中采用二进制对样本编码。
步骤S23中采用遗传算法进行优化求解。
S3:利用所述建模样本建立属性Y与近红外光谱X之间的有偏最小最大概率回归模型:通过误差下确界的选取使模型的预测偏差以最大概率在期望的方向上;
采用有偏最小最大概率机建立回归模型,对近红外光谱数据X和属性数据Y做如下处理:
u
i=(Y
i+ε,X
i,1,X
i,2,...,X
i,j,...,X
i,m),U=(u
1,u
2,...,u
n)
T
v
i=(Y
i-ε,X
i,1,X
i,2,...,X
i,j,...,X
i,m),V=(v
1,v
2,...,v
n)
T
其中,ε为给定的绝对误差上确界。
步骤S3中需要指定绝对误
差上确界。步骤S3有偏最
小最大概率机模型为:
其中,α为u类的正确分类概率下确界,η
0为给定的v类正确分类概率下确界。
S4:完成对所述模型的参数调优:将测试集近红外光谱带入模型进行预测,根据输出的预测值与参考值计算出模型对应的均方根误差对比,选择最佳模型参数。
包括:
S41:从工业现场采集样本构建测试集;
S42:给定v类正确分类概率下确界η
0的搜索范围以及步长;
S42:遍历η
0,并根据所述有偏最小最大概率回归模型在测试集上输出 的预测值和参考属性计算均方根误差RMSE:
S43:选定使RMSE最低的η
0为模型参数。
本发明方法首先采用有偏最小最大概率回归算法构建适应度函数,完成对训练样本的选取,在降低建模样本数量的同时能够最大程度的提高所选样本对目标工况的代表性;再利用有偏最小最大概率回归构建近红外定量分析模型,以此提供概率有偏预测输出。
该方法的优势在于可利用遗传算法对训练样本进行选择,极大地提高了建立近红外定量分析模型的质量和效率;通过有偏最小最大概率回归建立预测模型,可以较好的处理非高斯数据,给出的概率有偏预测值可极大提高油品调合质量反馈控制的控制效果。
图1是优选的训练样本。
下面结合具体实施方式对本发明进行详细说明。
以实际汽油汽油调合过程为例,该过程主要包含组分油及调合成品油属性检测、调合配方实时优化、管道阀门比例控制3大部分,如图1所示,组分油管道阀门的开度由调合配方决定,优化服务器依据调合成品油以及组分油的实时属性进行在线优化。
调合成品汽油数据采自国内某炼油厂汽油管道调合过程过程中的监测数据,汽油近红外光谱的波长范围为1100nm‐1300nm,波长精度为1nm;样本研究法辛烷值的参考值采用ASTM标准的马达机测定。历史数据集中共包含350组样本,测试集中包含250组样本。
通过MATLAB对上述算法进行仿真,对本发明做进一步详述:
第一步:根据生产工况要求及从历史数据求出历史数据的stdY=0.49, 工况要求调合成品汽油RON≥93.8,依据经验最终取k=1;那么,从历史数据集中选取93.31≤y≤94.29的样本组成初始训练集O
init;
第二步:利用S21中的方法构造适应度函数,并利用S22‐S24中所述方法采用GA算法进行训练样本优选。最终选择出120个训练样本。
第三步:利用所选出的训练样本构建有偏最小最大概率回归模型,并利用S41‐S42所述方法确定最优η
0=0.29;
第四步:利用上述训练样本以及模型参数,建立有偏最小最大概率回归模型,用于在线过程。
选用梯度下降法求解所述最小最大概率模型,求解后获得回归方程为:
以上所述仅是对本发明的较佳实施方式而已,并非对本发明作任何形式上的限制,凡是依据本发明的技术实质对以上实施方式所做的任何简单修改,等同变化与修饰,均属于本发明技术方案的范围内。
Claims (5)
- 基于有偏估计的近红外定量分析模型构建方法,其特征在于按照以下步骤进行:S1:数据预处理:首先从历史数据集O中初选训练样本O init={X∈R N×m,Y∈R N×1}∈O,N代表样本数,m代表样本维度;S2:训练样本优化选择:根据目标工况,从O init样本集中选择合适的建模样本组成子集O opt∈O init,优选后的样本作为近红外定量分析模型的建模样本;S3:利用所述建模样本建立属性Y与近红外光谱X之间的有偏最小最大概率回归模型:通过误差下确界的选取使模型的预测偏差以最大概率在期望的方向上;S4:完成对所述模型的参数调优:将测试集近红外光谱带入模型进行预测,根据输出的预测值与参考值计算出模型对应的均方根误差对比,选择最佳模型参数。
- 按照权利要求1所述基于有偏估计的近红外定量分析模型构建方法,其特征在于:所述步骤S2包括S21:构造适应度函数S22:采用二进制编码方法,对O init数据集中的样本进行编码,若所述数据集中某个样本被选为建模样本,则其编码值为1,否则为0;S23:利用遗传算法迭代选择部分样本构成训练集,并计算所述训练集对应的适应度函数值;S24:选择最小适应度函数值对应的解组成最优训练集O opt;步骤S21中的权重因子w为:其中,γ为位置参数;步骤S22中采用二进制对样本编码;步骤S23中采用遗传算法进行优化求解。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
ZA2021/08529A ZA202108529B (en) | 2019-08-29 | 2021-11-02 | Near-infrared quantitative analysis model construction method based on biased estimation |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910806825.5A CN110514619B (zh) | 2019-08-29 | 2019-08-29 | 基于有偏估计的近红外定量分析模型构建方法 |
CN201910806825.5 | 2019-08-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021036546A1 true WO2021036546A1 (zh) | 2021-03-04 |
Family
ID=68627794
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/101622 WO2021036546A1 (zh) | 2019-08-29 | 2020-07-13 | 基于有偏估计的近红外定量分析模型构建方法 |
Country Status (3)
Country | Link |
---|---|
CN (1) | CN110514619B (zh) |
WO (1) | WO2021036546A1 (zh) |
ZA (1) | ZA202108529B (zh) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113125377A (zh) * | 2021-03-30 | 2021-07-16 | 武汉理工大学 | 一种基于近红外光谱检测柴油性质的方法及装置 |
CN113821934A (zh) * | 2021-09-30 | 2021-12-21 | 国网青海省电力公司电力科学研究院 | 一种工况参数的预测方法、装置、设备及存储介质 |
CN113868597A (zh) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | 一种用于年龄估计的回归公平性度量方法 |
CN114062306A (zh) * | 2021-10-25 | 2022-02-18 | 华东理工大学 | 一种近红外光谱数据分段预处理方法 |
CN114219157A (zh) * | 2021-12-17 | 2022-03-22 | 西南石油大学 | 一种基于最优决策和动态分析的烷烃气体红外光谱测量方法 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110514619B (zh) * | 2019-08-29 | 2021-01-29 | 山东科技大学 | 基于有偏估计的近红外定量分析模型构建方法 |
CN112509643B (zh) * | 2021-02-03 | 2021-07-09 | 蓝星安迪苏南京有限公司 | 一种定量分析模型构建方法、定量分析方法、装置及系统 |
CN113569951B (zh) * | 2021-07-29 | 2023-11-07 | 山东科技大学 | 一种基于生成对抗网络的近红外定量分析模型构建方法 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104376325A (zh) * | 2014-10-30 | 2015-02-25 | 中国科学院半导体研究所 | 一种近红外定性分析模型的建立方法 |
CN104462751A (zh) * | 2014-10-29 | 2015-03-25 | 温州大学 | 一种基于多元高斯拟合的近红外光谱建模方法 |
CN105548027A (zh) * | 2015-12-09 | 2016-05-04 | 湖南省农产品加工研究所 | 基于近红外光谱测定调和油中茶油含量的分析模型及方法 |
CN107356556A (zh) * | 2017-07-10 | 2017-11-17 | 天津工业大学 | 一种近红外光谱定量分析的双集成建模方法 |
CN107357269A (zh) * | 2017-06-17 | 2017-11-17 | 湖州师范学院 | 基于混合mpls的多阶段过程质量预报方法 |
US20180259445A1 (en) * | 2012-06-14 | 2018-09-13 | Ramot At Tel-Aviv University Ltd. | Quantitative assessment of soil contaminants, particularly hydrocarbons, using reflectance spectroscopy |
CN110514619A (zh) * | 2019-08-29 | 2019-11-29 | 山东科技大学 | 基于有偏估计的近红外定量分析模型构建方法 |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1703272A1 (en) * | 2005-03-16 | 2006-09-20 | BP Chemicals Limited | Measuring near infra-red spectra using a demountable NIR transmission cell |
CN103440425B (zh) * | 2013-09-05 | 2016-07-06 | 中国石油化工股份有限公司 | 一种辛烷值回归模型的建立方法 |
CN103528990B (zh) * | 2013-10-31 | 2017-07-28 | 天津工业大学 | 一种近红外光谱的多模型建模方法 |
KR20180030414A (ko) * | 2015-07-30 | 2018-03-22 | 바이오마린 파머수티컬 인크. | 골 이형성증 치료를 위한 c―형 나트륨이뇨 펩타이드 변이체의 용도 |
CN107748146A (zh) * | 2017-10-20 | 2018-03-02 | 华东理工大学 | 一种基于近红外光谱检测的原油属性快速预测方法 |
-
2019
- 2019-08-29 CN CN201910806825.5A patent/CN110514619B/zh active Active
-
2020
- 2020-07-13 WO PCT/CN2020/101622 patent/WO2021036546A1/zh active Application Filing
-
2021
- 2021-11-02 ZA ZA2021/08529A patent/ZA202108529B/en unknown
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180259445A1 (en) * | 2012-06-14 | 2018-09-13 | Ramot At Tel-Aviv University Ltd. | Quantitative assessment of soil contaminants, particularly hydrocarbons, using reflectance spectroscopy |
CN104462751A (zh) * | 2014-10-29 | 2015-03-25 | 温州大学 | 一种基于多元高斯拟合的近红外光谱建模方法 |
CN104376325A (zh) * | 2014-10-30 | 2015-02-25 | 中国科学院半导体研究所 | 一种近红外定性分析模型的建立方法 |
CN105548027A (zh) * | 2015-12-09 | 2016-05-04 | 湖南省农产品加工研究所 | 基于近红外光谱测定调和油中茶油含量的分析模型及方法 |
CN107357269A (zh) * | 2017-06-17 | 2017-11-17 | 湖州师范学院 | 基于混合mpls的多阶段过程质量预报方法 |
CN107356556A (zh) * | 2017-07-10 | 2017-11-17 | 天津工业大学 | 一种近红外光谱定量分析的双集成建模方法 |
CN110514619A (zh) * | 2019-08-29 | 2019-11-29 | 山东科技大学 | 基于有偏估计的近红外定量分析模型构建方法 |
Non-Patent Citations (1)
Title |
---|
HE KAIXUN; CAO PENGFEI: "Optimization of Soft-sensing Modeling Based on Intelligent Optimization Algorithm and Application", CHEMICAL INDUSTRY AND ENGINEERING PROGRESS, vol. 37, no. 7, 5 July 2018 (2018-07-05), pages 2516 - 2523, XP009526489, ISSN: 1000-6613, DOI: 10.16085/j.issn.1000-6613.2017-1846 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113125377A (zh) * | 2021-03-30 | 2021-07-16 | 武汉理工大学 | 一种基于近红外光谱检测柴油性质的方法及装置 |
CN113125377B (zh) * | 2021-03-30 | 2024-02-23 | 武汉理工大学 | 一种基于近红外光谱检测柴油性质的方法及装置 |
CN113868597A (zh) * | 2021-09-27 | 2021-12-31 | 电子科技大学 | 一种用于年龄估计的回归公平性度量方法 |
CN113821934A (zh) * | 2021-09-30 | 2021-12-21 | 国网青海省电力公司电力科学研究院 | 一种工况参数的预测方法、装置、设备及存储介质 |
CN113821934B (zh) * | 2021-09-30 | 2024-01-19 | 国网青海省电力公司电力科学研究院 | 一种工况参数的预测方法、装置、设备及存储介质 |
CN114062306A (zh) * | 2021-10-25 | 2022-02-18 | 华东理工大学 | 一种近红外光谱数据分段预处理方法 |
CN114062306B (zh) * | 2021-10-25 | 2024-04-05 | 华东理工大学 | 一种近红外光谱数据分段预处理方法 |
CN114219157A (zh) * | 2021-12-17 | 2022-03-22 | 西南石油大学 | 一种基于最优决策和动态分析的烷烃气体红外光谱测量方法 |
CN114219157B (zh) * | 2021-12-17 | 2023-10-17 | 西南石油大学 | 一种基于最优决策和动态分析的烷烃气体红外光谱测量方法 |
Also Published As
Publication number | Publication date |
---|---|
ZA202108529B (en) | 2022-04-28 |
CN110514619B (zh) | 2021-01-29 |
CN110514619A (zh) | 2019-11-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021036546A1 (zh) | 基于有偏估计的近红外定量分析模型构建方法 | |
CN109783906B (zh) | 一种管道内检测漏磁数据智能分析系统及方法 | |
CN112101480B (zh) | 一种多变量聚类与融合的时间序列组合预测方法 | |
CN105740984A (zh) | 一种基于性能预测的产品概念性能评价方法 | |
CN103543719B (zh) | 一种基于工况的流程行业操作模式自适应调整方法 | |
CN112100745B (zh) | 基于lda理论的汽车大梁钢力学性能预测方法 | |
CN108875118B (zh) | 一种高炉铁水硅含量预测模型准确度评价方法和设备 | |
CN110046377B (zh) | 一种基于异构相似度的选择性集成即时学习软测量建模方法 | |
CN115860211A (zh) | 一种基于局部在线建模的铸坯质量预测方法 | |
CN109241077A (zh) | 基于相似性的生产指标变化趋势可视化查询系统及方法 | |
CN115497574A (zh) | 一种基于模型融合的hpc抗压强度预测方法和系统 | |
CN116468160A (zh) | 基于生产大数据的铝合金压铸件质量预测方法 | |
CN116502455A (zh) | 一种激光选区熔化技术的工艺参数确定方法及系统 | |
Vahdani et al. | A neural network model based on support vector machine for conceptual cost estimation in construction projects | |
CN109858707A (zh) | 一种基于abc-anfis-ctf的乙烯裂解炉炉管智能结焦诊断方法 | |
CN116825253B (zh) | 基于特征选择的热轧带钢力学性能预测模型的建立方法 | |
CN117139380A (zh) | 一种基于调控经验自学习的镰刀弯控制方法 | |
Yan et al. | Estimation of copper concentrate grade for copper flotation | |
CN113723015A (zh) | 一种基于机理模型与大数据技术的催化裂化装置优化方法 | |
CN115186584A (zh) | 一种融合注意力机制和自适应构图的宽度学习半监督软测量建模方法 | |
CN113569951B (zh) | 一种基于生成对抗网络的近红外定量分析模型构建方法 | |
CN112925202B (zh) | 基于动态特征提取的发酵过程阶段划分方法 | |
CN116595889B (zh) | 基于peek材料的薄型筋体均布结构加工方法及系统 | |
CN115576288B (zh) | 一种过滤材料的生产控制方法和系统 | |
CN111781824B (zh) | 一种基于vine copula分位数回归的自适应软测量方法及系统 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20859289 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20859289 Country of ref document: EP Kind code of ref document: A1 |