CN117352094A - Physical property prediction analysis method and system for raw oil - Google Patents
Physical property prediction analysis method and system for raw oil Download PDFInfo
- Publication number
- CN117352094A CN117352094A CN202311640886.1A CN202311640886A CN117352094A CN 117352094 A CN117352094 A CN 117352094A CN 202311640886 A CN202311640886 A CN 202311640886A CN 117352094 A CN117352094 A CN 117352094A
- Authority
- CN
- China
- Prior art keywords
- data
- sequence
- anomaly
- analyzed
- density
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 55
- 230000000704 physical effect Effects 0.000 title claims abstract description 43
- 238000004519 manufacturing process Methods 0.000 claims abstract description 80
- 238000000034 method Methods 0.000 claims abstract description 70
- 230000002159 abnormal effect Effects 0.000 claims abstract description 39
- 230000035772 mutation Effects 0.000 claims abstract description 33
- 230000005856 abnormality Effects 0.000 claims description 18
- 238000010606 normalization Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 239000002994 raw material Substances 0.000 description 5
- 230000001419 dependent effect Effects 0.000 description 4
- 230000007423 decrease Effects 0.000 description 3
- MWPLVEDNUUSJAV-UHFFFAOYSA-N anthracene Chemical compound C1=CC=CC2=CC3=CC=CC=C3C=C21 MWPLVEDNUUSJAV-UHFFFAOYSA-N 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000005984 hydrogenation reaction Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000005477 standard model Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- UHOVQNZJYSORNB-UHFFFAOYSA-N benzene Substances C1=CC=CC=C1 UHOVQNZJYSORNB-UHFFFAOYSA-N 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data predictive analysis, in particular to a physical property predictive analysis method and system for raw oil. The method comprises the following steps: acquiring density data of raw oil to obtain a sequence to be analyzed; determining data to be detected, data with the same sequence and data at the same time, and further obtaining data anomaly degree; obtaining an anomaly correction coefficient and a process influence coefficient according to the data anomaly degree of the data to be detected, the same-sequence data and the same-time data; further determining a data mutation coefficient of the data to be detected; determining the sequence weight of each sequence to be analyzed according to the data mutation coefficient and the data anomaly degree; and obtaining a predicted sequence, and analyzing the density and physical properties of the raw oil according to the predicted sequence to obtain an analysis result. According to the method, the influence caused by abnormal mutation of data can be effectively eliminated by analyzing the same production batch and the same production time in two dimensions, the accuracy and objectivity of the raw oil density prediction are improved, the reliability of physical property analysis is improved, and the analysis effect is enhanced.
Description
Technical Field
The invention relates to the technical field of data predictive analysis, in particular to a physical property predictive analysis method and system for raw oil.
Background
The physical property prediction analysis method of the raw oil has been widely applied to the petroleum and chemical industries. The density of the raw oil is an important physical index during the processing of the raw oil, and the density index of each flow is usually predicted and analyzed during the processing of the raw oil.
In the related art, physical property indexes are obtained, then, a vector matrix of the physical property indexes is established, and the physical property prediction analysis of the raw oil is realized by comparing the vector matrix with a standard model.
In this way, the physical property index of the raw oil may cause abnormal mutation due to the abnormality of the transmission process or the collection process, and the existence of the abnormal mutation may affect the comparison result of the standard model, so as to affect the subsequent prediction analysis, so that the accuracy and objectivity of predicting the density of the raw oil are low, the reliability of the physical property analysis of the raw oil is insufficient, and the analysis effect is poor.
Disclosure of Invention
In order to solve the technical problems of low accuracy and objectivity of predicting the density of the raw oil and poor analysis effect of the physical property analysis of the raw oil in the related art, the invention provides a physical property prediction analysis method and system for the raw oil, and the adopted technical scheme is as follows:
the invention provides a physical property prediction analysis method for raw oil, which comprises the following steps:
periodically acquiring density data of raw oil of different production batches, and respectively arranging the density data of the same production batch according to a time sequence order to obtain a sequence to be analyzed;
taking the density data of any sequence to be analyzed at any production time as data to be analyzed, taking other density data of the data to be analyzed in the same sequence as the same sequence data, and taking the density data of the data to be analyzed at the same production time as the same time data except the other sequences to be analyzed of the sequence to be analyzed where the data to be analyzed is located; determining the data anomaly degree of the data to be tested according to the data to be tested, the same-sequence data and the same-time data;
determining an anomaly correction coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the same-sequence data, and determining a process influence coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the same-time data; determining a data mutation coefficient of the data to be detected according to the abnormal correction coefficient and the process influence coefficient; determining the sequence weight of each sequence to be analyzed according to the data mutation coefficients and the data anomaly degree of all density data in the same sequence to be analyzed;
and carrying out weighted fusion on all the sequences to be analyzed according to the sequence weight to obtain a predicted sequence, and analyzing the density physical properties of the raw oil according to the predicted sequence to obtain an analysis result.
Further, the determining the data anomaly degree of the data to be tested according to the data to be tested, the same-sequence data and the same-time data includes:
calculating the average value of all the same-sequence data as a sequence data average value, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the sequence data average value to obtain a sequence anomaly coefficient;
calculating the average value of the moment data as a moment data average value, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the moment data average value to obtain a moment anomaly coefficient;
and determining the data anomaly degree of the data to be detected according to the sequence anomaly coefficient and the moment anomaly coefficient.
Further, the determining the data anomaly degree of the data to be measured according to the sequence anomaly coefficient and the time anomaly coefficient includes:
and calculating a normalized value of the product of the sequence anomaly coefficient and the moment anomaly coefficient to obtain the data anomaly degree.
Further, the determining the anomaly correction coefficient of the data to be tested according to the data anomalies of the data to be tested and all the data with the same sequence includes:
calculating the average value of the data anomaly degree of all the same-sequence data to obtain the average value of the sequence anomaly degree;
and taking a normalized value of the absolute value of the difference between the data anomaly degree of the data to be detected and the average value of the sequence anomaly degree as an anomaly correction coefficient of the data to be detected.
Further, the determining the process influence coefficient of the data to be measured according to the data anomaly degree of the data to be measured and all the simultaneous data includes:
calculating the difference absolute value of the data anomaly degree of the data to be measured and the data anomaly degree of each moment data respectively as the moment anomaly degree difference;
and calculating the normalized value of the mean value of all the abnormal degree differences at the same time to obtain the process influence coefficient of the data to be detected.
Further, the determining the data mutation coefficient of the data to be measured according to the anomaly correction coefficient and the process influence coefficient includes:
and calculating a normalized value of the product of the abnormal correction coefficient and the process influence coefficient to obtain a data mutation coefficient of the data to be detected.
Further, the determining the sequence weight of each sequence to be analyzed according to the data mutation coefficient and the data anomaly degree of all the density data in the same sequence to be analyzed comprises the following steps:
calculating the product of the data abnormality degree and the data mutation coefficient of any density data in the same sequence to be analyzed to obtain a weight influence value;
and calculating an inverse proportion normalization value of the weight influence value mean value of all the density data in each sequence to be analyzed to obtain the sequence weight of the sequence to be analyzed, wherein the sum value of the sequence weights of all the sequences to be analyzed is 1.
Further, the step of performing weighted fusion on all the sequences to be analyzed according to the sequence weights to obtain a predicted sequence includes:
calculating the product of each density data in the sequence to be analyzed and the sequence weight to obtain a density adjustment value;
calculating the sum value of all the density adjustment values at the same production time to obtain a predicted density value;
and sequencing all the predicted density values according to the time sequence to obtain a predicted sequence.
Further, the analyzing the density and physical properties of the raw oil according to the predicted sequence to obtain an analysis result includes:
calculating the average value of the absolute values of the differences between the predicted density values at all production moments in the predicted sequence and the preset density standard values corresponding to each production moment to obtain standard differences;
normalizing the standard difference to obtain a raw oil density index;
and taking the raw oil density index as an analysis result.
The invention also provides a physical property prediction analysis system for the raw oil, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the physical property prediction analysis method for the raw oil.
The invention has the following beneficial effects:
the invention obtains the sequence to be analyzed and determines the data to be analyzed, then, based on the data to be analyzed at the same time sequence and the data of the same sequence in the same production batch, carries out data anomaly degree analysis, it can be understood that the invention determines the data anomaly degree of the data to be analyzed through two dimensions of the same production batch and the same production time, so that the obtaining of the data anomaly degree can more accurately and effectively represent the abnormal condition of the data to be analyzed, the abnormal condition can be influenced by various objective factors, such as raw material quality, production process and the like, in order to eliminate the objective factors, thus, the abnormal mutation density data can be obtained, the invention determines the abnormal correction coefficient of the data to be tested according to the data to be tested and the data anomaly degree of all the same sequence in the same production batch dimension, determines the process influence coefficient of the data to be tested according to the data to be tested and the data anomaly degree of all the same time sequence in the same production time dimension, and based on the abnormal correction coefficient and the process influence coefficient of the process, so as to eliminate the influence of the anomaly, analyzes the abnormal condition of each density data to be analyzed according to the abnormal condition of the data to the abnormal sequence, and the data to be analyzed in the same production batch dimension and the data to be analyzed by the same time sequence, and the data to be analyzed in the same time sequence can be accurately and the data to be analyzed by the corresponding quality factor in the same time sequence and the sequence, the prediction sequence with stronger reliability and objectivity is obtained, further, the density and physical properties of the raw oil can be analyzed according to the prediction sequence, an analysis result is obtained, and the accuracy and the reliability of the analysis result are improved. In conclusion, the method and the device can effectively eliminate the influence of abnormal mutation of data by analyzing the two dimensions of the same production batch and the same production time, improve the accuracy and objectivity of the raw oil density prediction, improve the reliability of physical property analysis and enhance the analysis effect.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for predicting and analyzing physical properties of a raw oil according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following description refers to the specific implementation, structure, characteristics and effects of a method and a system for predicting and analyzing physical properties of raw oil according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a physical property prediction analysis method for raw oil according to the present invention with reference to the accompanying drawings.
Referring to fig. 1, a flow chart of a method for predicting and analyzing physical properties of a raw oil according to an embodiment of the invention is shown, where the method includes:
s101: and periodically acquiring the density data of the raw oil of different production batches, and respectively arranging the density data of the same production batch according to a time sequence order to obtain a sequence to be analyzed.
The specific implementation scene of the invention is as follows: and periodically acquiring the density data of the raw oil of different production batches, and predicting and analyzing the data of the subsequent production batches according to the change of the density data of the raw oil of different production batches.
The density of the raw oil is an important physical property index during the processing of the raw oil, and the density index of each flow is usually predicted during the processing of the raw oil. In the processing process of the raw oil, a plurality of different processes are needed, for example, an anthracene oil hydrogenation process comprises a plurality of processing flows of a two-benzene tower, a hydrogenation reactor, a light component removal tower, a rectifying tower and the like.
The acquisition period is a raw oil density acquisition period, and the embodiment of the invention can take the time sequence starting time of each production batch as a starting point, and then acquire the density data of the raw oil in different production batches based on the same acquisition period. Alternatively, the acquisition period may be, for example, 30 seconds, or the present invention may be specifically adjusted according to the production condition of the actual production line, which is not limited.
After the density data is obtained, the density data of the same production batch are arranged according to the time sequence to obtain the data to be analyzed, wherein the time sequence can also be used as the flow sequence of a production line, and the production line obtains the sequence to be analyzed by tracking the production and processing conditions of a batch of raw oil when the raw oil is processed.
In the embodiment of the invention, the production of the raw oil is predicted and analyzed according to the sequences to be analyzed of different production batches, and the specific prediction and analysis process is referred to in the subsequent embodiment.
S102: taking the density data of any sequence to be analyzed at any production moment as data to be analyzed, taking other density data of the data to be analyzed in the same sequence as the data to be analyzed, and taking the density data of the data to be analyzed at the same production moment as data at the same moment except the other sequences to be analyzed of the sequence to be analyzed where the data to be analyzed is located; and determining the data anomaly degree of the data to be tested according to the data to be tested, the same-sequence data and the same-time data.
In the embodiment of the invention, the density data of any sequence to be analyzed at any production time can be used as the data to be detected, and then, the data to be detected is subjected to double analysis of the same production batch and the same production time.
In the embodiment of the invention, other density data in the same sequence to be analyzed with the data to be analyzed are taken as the same-sequence data. That is, the same-sequence data is other data in the same production lot as the data to be measured, for example, there are 5 density data {1,2,3,4,5} in the same production lot, the corresponding 5 composition sequences to be analyzed are "3" as the data to be measured, and then the corresponding same-sequence data are "1", "2", "4", and "5".
It can be understood that in the embodiment of the invention, the density data of the sequence to be analyzed except the sequence to be analyzed where the data to be analyzed is located and the data to be measured at the same production time are taken as the data at the same time. Because the data are collected relatively synchronously in different production batches, that is, the corresponding time nodes in the process flow are collected once, that is, the production moments in different sequences to be analyzed can be matched with each other, for example, the fifth data in any sequence to be analyzed is used as the data to be tested, and then the fifth data in all other sequences to be analyzed can be used as the data to be tested at the same time.
Therefore, after the same-sequence data and the same-time data are determined, the data to be detected is specifically analyzed based on the data values of the data to be detected, the same-sequence data and the same-time data.
Further, in some embodiments of the present invention, determining the data anomaly of the data to be measured according to the data to be measured, the co-sequential data, and the simultaneous data includes: calculating the average value of all the same-sequence data as the average value of the sequence data, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the average value of the sequence data to obtain the sequence anomaly coefficient; calculating the average value of the data at the same time as the average value of the data at the same time, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the average value of the data at the same time to obtain a time anomaly coefficient; and determining the data anomaly degree of the data to be tested according to the sequence anomaly coefficient and the moment anomaly coefficient.
In one embodiment of the present invention, the normalization process may specifically be, for example, maximum and minimum normalization processes, and the normalization in the subsequent steps may be performed by using the maximum and minimum normalization processes, and in other embodiments of the present invention, other normalization methods may be selected according to a specific range of values, which will not be described herein.
In the embodiment of the invention, the average value of all the data in the same sequence is calculated as the average value of the sequence data, namely the average value of the sequence data can represent the overall level of the density data in the sequence to be analyzed, the difference value between the data to be tested and the average value of the sequence data is calculated, the difference of the data to be tested in the whole sequence to be analyzed can be represented, and the sequence anomaly coefficient is obtained.
In the embodiment of the invention, the average value of the data at the same time is calculated as the average value of the data at the same time, the average value of the data at the same time can represent the integral level of the density data in all the sequences to be analyzed at the same production time, and the normalized value of the absolute value of the difference value between the data to be measured and the average value of the data at the same time is calculated to obtain the abnormal coefficient at the time.
Further, in some embodiments of the present invention, determining the data anomaly degree of the data to be measured according to the sequence anomaly coefficient and the time anomaly coefficient includes: and calculating the normalized value of the product of the sequence anomaly coefficient and the moment anomaly coefficient, and obtaining the data anomaly degree.
In the embodiment of the invention, as the change of the raw oil in the same production batch is more linear, the average value of the sequence data can represent the average level of the raw oil density corresponding to the same production batch, and the larger the average value difference between the data to be tested and the sequence data is, the more likely to be nonlinear abnormal data is indicated, so that the larger the corresponding data anomaly degree is, namely the sequence anomaly coefficient and the data anomaly degree are in positive correlation.
The positive correlation relationship indicates that the dependent variable increases along with the increase of the independent variable, the dependent variable decreases along with the decrease of the independent variable, and the specific relationship can be multiplication relationship, addition relationship, idempotent of an exponential function and is determined by practical application; in contrast, the negative correlation represents that the dependent variable decreases with increasing independent variable, and the dependent variable increases with decreasing independent variable, and may be a subtraction relationship, a division relationship, or the like, which is determined by the actual application.
In the embodiment of the invention, because the numerical value difference at the same production time is smaller in different production batches, the more abnormal the corresponding data to be measured is also represented when the difference is larger, so that the time abnormality coefficient and the data abnormality are in positive correlation. And calculating the normalized value of the product of the sequence anomaly coefficient and the moment anomaly coefficient, and obtaining the data anomaly degree.
S103: determining an abnormal correction coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the data with the same sequence, and determining a process influence coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the data with the same time; determining a data mutation coefficient of the data to be tested according to the abnormal correction coefficient and the process influence coefficient; and determining the sequence weight of each sequence to be analyzed according to the data mutation coefficients and the data anomaly degree of all the density data in the same sequence to be analyzed.
In the embodiment of the invention, since the data anomaly degree is an anomaly analysis result obtained in two dimensions of the same sequence to be analyzed and the same production time, namely, the reliability of the data anomaly degree is higher, after the data anomaly degree is determined, the data prediction analysis can be performed on the follow-up data based on the data anomaly degree of each density data.
Further, in some embodiments of the present invention, determining an anomaly correction coefficient of the data to be measured according to the data anomalies of the data to be measured and all the same-sequence data includes: calculating the average value of the data anomaly degree of all the data in the same sequence to obtain the average value of the sequence anomaly degree; and taking the normalized value of the difference absolute value of the data anomaly degree of the data to be detected and the sequence anomaly degree mean value as an anomaly correction coefficient of the data to be detected.
The embodiment of the invention analyzes the same-sequence data, obtains the sequence anomaly average value by calculating the average value of the data anomalies of all the same-sequence data, and calculates the difference absolute value of the data anomalies of the data to be tested and the sequence anomaly average value after calculating the sequence anomaly average value of the same-sequence data of the sequence to which the data to be tested belongs.
It will be appreciated that anomalies in the production process of the raw oil may be the influence of the quality of the raw oil itself, and that under the influence of the quality of the raw oil itself, the overall density data analysis anomalies should be similar, i.e. the corresponding anomalies should extend through the whole production batch, i.e. the data anomalies of all density data in the whole production batch should be similar.
In the embodiment of the invention, the sequence anomaly mean value characterizes the overall anomaly characteristic of the sequence to be analyzed, and when the data anomaly of the data to be tested is larger than the difference, the data anomaly of the data to be tested is more likely to be different from the data anomaly of all density data in the production batch, namely the numerical value of the data anomaly of the data to be tested is less likely to be anomalies caused by the quality of the raw oil, and the corresponding data to be tested is more likely to be sudden data anomalies, so that the anomaly correction coefficient is larger.
Further, in some embodiments of the present invention, determining a process influence coefficient of the data to be measured according to the data anomalies of the data to be measured and all the simultaneous data includes: calculating the difference absolute value of the data anomaly degree of the data to be measured and the data anomaly degree of each moment data respectively as the moment anomaly degree difference; and calculating the normalized value of the mean value of all the abnormal degree differences at the same time to obtain the process influence coefficient of the data to be detected.
In the embodiment of the present invention, the process influence coefficient is a difference generated by the process distinction, and it can be understood that in the same production time, the processes used should be consistent, that is, under a normal production batch, the data anomaly degree of all density data at the same production time should be kept consistent.
In the embodiment of the invention, the absolute value of the difference between the data anomaly degree of the data to be measured and the data anomaly degree of each moment data is calculated as the difference of the data anomaly degree of each moment, the difference of the data to be measured and the data value of the data at the moment is represented by the difference of the anomaly degree of the moment, the larger the difference value is, the more unlikely the difference between the data to be measured and the data at the moment is represented by the difference value, the difference is the difference generated by the process, therefore, the normalized value of the average value of the differences of the anomaly degree of all the moments is obtained, the process influence coefficient of the data to be measured is the larger the process influence coefficient is, the abnormal condition of the data to be measured is the abnormal condition caused by the process is represented by the less likely to be the abnormal condition of the data to be measured.
According to the embodiment of the invention, the abnormal correction coefficient and the process influence coefficient are obtained, so that the abnormal influence of the material and the abnormal influence of the process on the data to be detected are analyzed, and the reliability of the abnormal data analysis is ensured.
Further, in some embodiments of the present invention, determining a data mutation coefficient of the data to be measured according to the anomaly correction coefficient and the process influence coefficient includes: and calculating a normalized value of the product of the abnormal correction coefficient and the process influence coefficient to obtain a data mutation coefficient of the data to be detected.
In the embodiment of the invention, after the anomaly correction coefficient and the process influence coefficient are determined, the data mutation coefficient can be calculated by combining the anomaly coefficient and the process influence coefficient, wherein the data mutation coefficient is the influence value generated by the anomaly of the data to be measured, and the influence is not influenced by objective factors of the process and the raw materials, so that the data mutation coefficient is used as the anomaly degree of the data to be measured to have higher accuracy.
The above steps can easily learn that the larger the abnormality correction coefficient is, the less likely the value of the data abnormality of the data to be detected is the abnormality caused by the quality of the raw oil, and the more likely the corresponding data to be detected is the sudden data abnormality; the larger the process influence coefficient is, the less likely the abnormal condition of the data to be measured is the abnormality caused by the process, namely the more likely the abnormal condition is the abnormality of the data to be measured, therefore, the embodiment of the invention calculates the normalized value of the product of the abnormality correction coefficient and the process influence coefficient to obtain the data mutation coefficient of the data to be measured.
Further, in some embodiments of the present invention, determining the sequence weight of each sequence to be analyzed according to the data mutation coefficients and the data anomalies of all density data within the same sequence to be analyzed includes: calculating the product of the data abnormality degree and the data mutation coefficient of any density data in the same sequence to be analyzed to obtain a weight influence value; and calculating an inverse proportion normalization value of the weight influence value mean value of all the density data in each sequence to be analyzed to obtain the sequence weight of the sequence to be analyzed.
The sequence weight is a weight value for predicting the sequence of the sequence to be analyzed. In the embodiment of the invention, the data mutation coefficient represents an abnormal influence coefficient caused by mutation of the density data, and the larger the value is, the lower the reliability of the corresponding density data on subsequent prediction analysis is represented, so that the smaller the corresponding weight is, the larger the data anomaly degree is, the higher the anomaly degree of the representing density data is, and the lower the reliability of the corresponding density data on subsequent prediction analysis is, namely the lower the weight is.
In the embodiment of the invention, the weight influence value is obtained by calculating the product of the data anomaly degree and the data mutation coefficient, and the larger the weight influence value is, the smaller the corresponding influence on the sequence weight is, the inverse proportion normalization processing is carried out on the weight influence value mean value of all density data in the sequence to be analyzed to obtain the sequence weight of the sequence to be analyzed, and the larger the sequence weight is, the larger the corresponding predicted influence on the follow-up sequence to be analyzed is. Thus, predictive analysis based on sequence weights is facilitated.
In the embodiment of the invention, the average value of the weight influence values of all density data in the sequence to be analyzed is subjected to inverse proportion normalization processing to obtain the sequence weight of the sequence to be analyzed, and the sum value of all the sequence weights can be set to be 1 for facilitating subsequent statistics and subsequent calculation.
S104: and carrying out weighted fusion on all sequences to be analyzed according to the sequence weight to obtain a predicted sequence, and analyzing the density physical properties of the raw oil according to the predicted sequence to obtain an analysis result.
In the embodiment of the invention, each sequence to be analyzed has the corresponding sequence weight, and the weighting fusion can be carried out according to the sequence weight of each sequence to be analyzed.
Further, in some embodiments of the present invention, the weighted fusion is performed on all the sequences to be analyzed according to the sequence weights, so as to obtain a predicted sequence, including: calculating the product of each density data in the sequence to be analyzed and the sequence weight to obtain a density adjustment value; calculating the sum value of all density adjustment values at the same production time to predict a density value; and sequencing all the predicted density values according to the time sequence to obtain a predicted sequence.
In the embodiment of the invention, the sequence weight can be used as the weight value corresponding to the density data in the sequence to be analyzed, and the product of each density data in the sequence to be analyzed and the sequence weight is calculated to obtain the density adjustment value. Because the sum of all the sequence weights is 1, the embodiment of the invention calculates the product of each density data in the sequence to be analyzed and the sequence weights to obtain a density adjustment value, calculates the sum of all the density adjustment values at the same production time to obtain a predicted density value, and the process uses the sequence weights as the weights in the weighting process, so that the weighting calculation is carried out on all the density data at the same production time to obtain the predicted density value at the corresponding production time.
The invention can obtain the predicted sequence by combining the predicted density values at all production moments according to time sequence.
Further, in some embodiments of the present invention, the density physical properties of the raw oil are analyzed according to a predicted sequence, and analysis results are obtained, including: calculating the average value of the absolute values of the differences between the predicted density values at all production moments in the predicted sequence and the preset density standard values corresponding to each production moment to obtain standard differences; normalizing the standard difference to obtain a raw oil density index; the raw oil density index was used as the analysis result.
According to the method, the standard difference is obtained by calculating the average value of the absolute value of the difference between the predicted density value at all production moments in the prediction sequence and the preset density standard value corresponding to each production moment, namely, the standard difference is obtained by comparing the absolute value with the preset standard value, the raw oil density index is obtained by analyzing the standard difference, namely, the difference index between the predicted raw oil density and the standard value is obtained, the greater the raw oil density index is, the greater the predicted density and the standard difference of the corresponding raw oil are, namely, the more the conditions such as raw material abnormality and process abnormality in a production batch are, the smaller the raw oil density index is, namely, the smaller the predicted density and the standard difference of the corresponding raw oil are, namely, the raw material abnormality and process abnormality in the production batch are.
The invention obtains the sequence to be analyzed and determines the data to be analyzed, and then, based on the data to be analyzed, the data are simultaneously time-stamped on the time sequence and the data with the same sequence in the same production batch are analyzed for data abnormality degree. It can be understood that the abnormal condition of the data to be detected is determined through the two dimensions of the same production batch and the same production time, so that the acquisition of the abnormal condition of the data to be detected can be more accurately and effectively represented, the abnormal condition can be influenced by various objective factors, such as raw material quality, production process and the like. In conclusion, the method and the device can effectively eliminate the influence of abnormal mutation of data by analyzing the two dimensions of the same production batch and the same production time, improve the accuracy and objectivity of the raw oil density prediction, improve the reliability of physical property analysis and enhance the analysis effect.
The invention also provides a physical property prediction analysis system for the raw oil, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the physical property prediction analysis method for the raw oil.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. A method for predictive analysis of physical properties of a raw oil, the method comprising:
periodically acquiring density data of raw oil of different production batches, and respectively arranging the density data of the same production batch according to a time sequence order to obtain a sequence to be analyzed;
taking the density data of any sequence to be analyzed at any production time as data to be analyzed, taking other density data of the data to be analyzed in the same sequence as the same sequence data, and taking the density data of the data to be analyzed at the same production time as the same time data except the other sequences to be analyzed of the sequence to be analyzed where the data to be analyzed is located; determining the data anomaly degree of the data to be tested according to the data to be tested, the same-sequence data and the same-time data;
determining an anomaly correction coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the same-sequence data, and determining a process influence coefficient of the data to be detected according to the data anomaly degree of the data to be detected and all the same-time data; determining a data mutation coefficient of the data to be detected according to the abnormal correction coefficient and the process influence coefficient; determining the sequence weight of each sequence to be analyzed according to the data mutation coefficients and the data anomaly degree of all density data in the same sequence to be analyzed;
and carrying out weighted fusion on all the sequences to be analyzed according to the sequence weight to obtain a predicted sequence, and analyzing the density physical properties of the raw oil according to the predicted sequence to obtain an analysis result.
2. The method for predicting analysis of physical properties of a raw oil according to claim 1, wherein determining the degree of data anomaly of the data to be measured based on the data to be measured, the co-sequential data, and the simultaneous data comprises:
calculating the average value of all the same-sequence data as a sequence data average value, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the sequence data average value to obtain a sequence anomaly coefficient;
calculating the average value of the moment data as a moment data average value, and calculating the normalized value of the absolute value of the difference value between the data to be detected and the moment data average value to obtain a moment anomaly coefficient;
and determining the data anomaly degree of the data to be detected according to the sequence anomaly coefficient and the moment anomaly coefficient.
3. The method for predicting analysis of physical properties of raw oil according to claim 2, wherein determining the degree of data anomaly of the data to be measured based on the sequence anomaly coefficient and the time anomaly coefficient comprises:
and calculating a normalized value of the product of the sequence anomaly coefficient and the moment anomaly coefficient to obtain the data anomaly degree.
4. The method for predicting analysis of physical properties of raw oil according to claim 1, wherein determining an anomaly correction coefficient of the data to be measured based on data anomalies of the data to be measured and all the same-sequence data comprises:
calculating the average value of the data anomaly degree of all the same-sequence data to obtain the average value of the sequence anomaly degree;
and taking a normalized value of the absolute value of the difference between the data anomaly degree of the data to be detected and the average value of the sequence anomaly degree as an anomaly correction coefficient of the data to be detected.
5. The method for predicting analysis of physical properties of raw oil according to claim 1, wherein determining the process influence coefficient of the data to be measured according to the data anomaly degree of the data to be measured and all the simultaneous data comprises:
calculating the difference absolute value of the data anomaly degree of the data to be measured and the data anomaly degree of each moment data respectively as the moment anomaly degree difference;
and calculating the normalized value of the mean value of all the abnormal degree differences at the same time to obtain the process influence coefficient of the data to be detected.
6. The method for predicting analysis of physical properties of raw oil according to claim 1, wherein said determining a data mutation coefficient of the data to be measured based on the anomaly correction coefficient and the process influence coefficient comprises:
and calculating a normalized value of the product of the abnormal correction coefficient and the process influence coefficient to obtain a data mutation coefficient of the data to be detected.
7. The method for predicting analysis of physical properties of raw oil according to claim 1, wherein determining the sequence weight of each sequence to be analyzed based on the data mutation coefficients and the data anomaly degree of all density data in the same sequence to be analyzed comprises:
calculating the product of the data abnormality degree and the data mutation coefficient of any density data in the same sequence to be analyzed to obtain a weight influence value;
and calculating an inverse proportion normalization value of the weight influence value mean value of all the density data in each sequence to be analyzed to obtain the sequence weight of the sequence to be analyzed, wherein the sum value of the sequence weights of all the sequences to be analyzed is 1.
8. The method for predicting and analyzing physical properties of raw oil according to claim 1, wherein the step of weighting and fusing all the sequences to be analyzed according to the sequence weights to obtain predicted sequences comprises the steps of:
calculating the product of each density data in the sequence to be analyzed and the sequence weight to obtain a density adjustment value;
calculating the sum value of all the density adjustment values at the same production time to obtain a predicted density value;
and sequencing all the predicted density values according to the time sequence to obtain a predicted sequence.
9. The method for predicting analysis of physical properties of a raw oil according to claim 8, wherein the analyzing of the density physical properties of the raw oil according to the prediction sequence to obtain the analysis result comprises:
calculating the average value of the absolute values of the differences between the predicted density values at all production moments in the predicted sequence and the preset density standard values corresponding to each production moment to obtain standard differences;
normalizing the standard difference to obtain a raw oil density index;
and taking the raw oil density index as an analysis result.
10. A physical property prediction analysis system for a raw oil, the system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640886.1A CN117352094B (en) | 2023-12-04 | 2023-12-04 | Physical property prediction analysis method and system for raw oil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311640886.1A CN117352094B (en) | 2023-12-04 | 2023-12-04 | Physical property prediction analysis method and system for raw oil |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117352094A true CN117352094A (en) | 2024-01-05 |
| CN117352094B CN117352094B (en) | 2024-03-08 |
Family
ID=89363504
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311640886.1A Active CN117352094B (en) | 2023-12-04 | 2023-12-04 | Physical property prediction analysis method and system for raw oil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117352094B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040045049A1 (en) * | 1998-09-22 | 2004-03-04 | James Zhang | Polynucleotides and polypeptides in plants |
| CN104766175A (en) * | 2015-04-16 | 2015-07-08 | 东南大学 | Power system abnormal data identifying and correcting method based on time series analysis |
| CN110852527A (en) * | 2019-11-20 | 2020-02-28 | 成都理工大学 | Reservoir physical property parameter prediction method combining deep learning |
| US20210240154A1 (en) * | 2020-01-31 | 2021-08-05 | Keyence Corporation | Programmable logic controller and analyzer |
| JP2022155382A (en) * | 2021-03-30 | 2022-10-13 | 一般財団法人石油エネルギー技術センター | Method for estimating yield of produced oil in fluidized catalytic cracking device |
| CN116029434A (en) * | 2023-01-04 | 2023-04-28 | 华东理工大学 | Method and system for predicting hydrogen content in raw oil and heavy fraction oil |
| CN116361695A (en) * | 2023-03-31 | 2023-06-30 | 支付宝实验室(新加坡)有限公司 | Abnormal data detection method and device |
| CN116659589A (en) * | 2023-07-25 | 2023-08-29 | 澳润(山东)药业有限公司 | Donkey-hide gelatin cake preservation environment monitoring method based on data analysis |
| CN116862912A (en) * | 2023-09-04 | 2023-10-10 | 山东恒信科技发展有限公司 | Raw oil impurity detection method based on machine vision |
| CN117132844A (en) * | 2023-10-27 | 2023-11-28 | 江苏惠汕新能源集团有限公司 | Classifying method for cracks and scratches of photovoltaic panel based on image processing |
-
2023
- 2023-12-04 CN CN202311640886.1A patent/CN117352094B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040045049A1 (en) * | 1998-09-22 | 2004-03-04 | James Zhang | Polynucleotides and polypeptides in plants |
| CN104766175A (en) * | 2015-04-16 | 2015-07-08 | 东南大学 | Power system abnormal data identifying and correcting method based on time series analysis |
| CN110852527A (en) * | 2019-11-20 | 2020-02-28 | 成都理工大学 | Reservoir physical property parameter prediction method combining deep learning |
| US20210240154A1 (en) * | 2020-01-31 | 2021-08-05 | Keyence Corporation | Programmable logic controller and analyzer |
| JP2022155382A (en) * | 2021-03-30 | 2022-10-13 | 一般財団法人石油エネルギー技術センター | Method for estimating yield of produced oil in fluidized catalytic cracking device |
| CN116029434A (en) * | 2023-01-04 | 2023-04-28 | 华东理工大学 | Method and system for predicting hydrogen content in raw oil and heavy fraction oil |
| CN116361695A (en) * | 2023-03-31 | 2023-06-30 | 支付宝实验室(新加坡)有限公司 | Abnormal data detection method and device |
| CN116659589A (en) * | 2023-07-25 | 2023-08-29 | 澳润(山东)药业有限公司 | Donkey-hide gelatin cake preservation environment monitoring method based on data analysis |
| CN116862912A (en) * | 2023-09-04 | 2023-10-10 | 山东恒信科技发展有限公司 | Raw oil impurity detection method based on machine vision |
| CN117132844A (en) * | 2023-10-27 | 2023-11-28 | 江苏惠汕新能源集团有限公司 | Classifying method for cracks and scratches of photovoltaic panel based on image processing |
Non-Patent Citations (3)
| Title |
|---|
| NGOOT-CHIN: "Detection of ploidy and chromosomal aberrations in commercial oil palm using high-throughput SNP markers", 《PLANTA》, vol. 253, no. 2, 18 March 2021 (2021-03-18) * |
| 李辉: "基于随钻录井工程参数的变质岩潜山储层物性预测方法及应用", 《特种油气藏》, 16 August 2023 (2023-08-16), pages 8 * |
| 郑文龙;都金康;: "基于BP神经网络的玛纳斯河流域雪盖率预测", 南京大学学报(自然科学), no. 05, 30 September 2015 (2015-09-30), pages 112 - 119 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117352094B (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112527788B (en) | Method and device for detecting and cleaning abnormal value of transformer monitoring data | |
| CN115933787B (en) | Indoor multi-terminal intelligent control system based on indoor environment monitoring | |
| CN109376892B (en) | Equipment state prediction method based on life cycle stage of equipment | |
| CN110070202A (en) | A method of economic output is predicted by electricity consumption data | |
| CN113935535A (en) | Principal component analysis method for medium-and-long-term prediction model | |
| CN117352094B (en) | Physical property prediction analysis method and system for raw oil | |
| CN112685857B (en) | Bearing residual life prediction method based on transfer migration | |
| CN114139446A (en) | On-line detection soft measurement method for components in special rectification process | |
| CN113869642A (en) | Enterprise activity determination method and device, electronic equipment and storage medium | |
| CN116258087B (en) | Matte grade soft measurement method and device, electronic equipment and storage medium | |
| CN113158448A (en) | Self-adaptive calculation method for equipment threshold of ship system | |
| CN116306322B (en) | Water total phosphorus concentration inversion method and device based on hyperspectral data | |
| CN112418522A (en) | Industrial heating furnace steel temperature prediction method based on three-branch integrated prediction model | |
| CN112580254A (en) | Multi-dimensional analysis and early warning method for converter transformer based on multi-level decision fusion | |
| CN114814419A (en) | Electric power cabinet heat dissipation performance detection method based on big data | |
| CN114819743B (en) | Energy consumption diagnosis and analysis method for chemical enterprises | |
| CN114386196B (en) | Method for evaluating mechanical property prediction accuracy of plate strip | |
| CN117710008B (en) | Ecological product sales information management system suitable for karst region | |
| CN117192063B (en) | Water quality prediction method and system based on coupled Kalman filtering data assimilation | |
| Agu et al. | Comparative Study of Least Squares, Ridge, Lasso and Principal Component Regression Algorithms | |
| CN117764247A (en) | Method and device for predicting power consumption of whole industry based on industrial elements | |
| CN116090784A (en) | Regional water resource balance evaluation diagnosis regulation integrated system | |
| Topalova et al. | Automated classification of heat resistant steel structures based on neural networks | |
| CN117216489A (en) | Waterproof coating quality inspection analysis method and system based on Internet | |
| CN116793979A (en) | Method for detecting chemical oxygen demand in water sample |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |