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

Oil-gas-water three-phase flow phase state critical point identification algorithm based on clustering analysis

A technology of identification algorithm and cluster analysis, applied in the field of petroleum engineering, can solve problems such as affecting the results of three-phase flow measurement of oil, gas and water

Active Publication Date: 2019-12-27
华运隆腾机械制造有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The most common clustering analysis algorithm is the k-means algorithm, which is simple in principle and has the advantages of fast calculation speed. However, if there are abnormal data or noisy data, the abnormal data or noisy data will have a great impact on the processing of the overall data. Applying it to the detection of the critical point of the phase state of the oil-gas-water three-phase flow will highlight the defects of the k-means algorithm and affect the results of the flow measurement of each phase of the oil-gas-water three-phase flow

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
  • Oil-gas-water three-phase flow phase state critical point identification algorithm based on clustering analysis
  • Oil-gas-water three-phase flow phase state critical point identification algorithm based on clustering analysis
  • Oil-gas-water three-phase flow phase state critical point identification algorithm based on clustering analysis

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0062] A cluster analysis-based identification algorithm for phase state critical points of oil-gas-water three-phase flow, the identification algorithm comprising the following steps:

[0063] Step 1: Select 3 data from all current measurement data values ​​of oil-gas-water three-phase flow, define it as the initial center M, and divide all data into 3 categories with M as the center;

[0064] Step 2: According to the Euclidean distance, classify other data except the initial center M into the class represented by the nearest initial center point to itself, and calculate the difference of the whole set of data in this classification case sum of absolute values ​​E pre ;

[0065] Step 3: Extract the data of the center position in the two types of data in step 2 as the new center M new , and reclassified to calculate the sum of the absolute values ​​of the corresponding differences E new , using the principle that the total cost of exchange TP is negative, replace the initia...

specific Embodiment approach 2

[0068] This embodiment is a further description of the first embodiment. In the first step, the current measurement data of the oil-gas-water three-phase flow is represented by a set X, and the set X is expressed as:

[0069] X={a 1 ,a 2 ,...,a i ,...,a j ,...,a s}

[0070] In the formula, i and j both represent the serial number of the current data value, a i ,a j Obtain the converted i-th and j-th current data values ​​for the multi-phase flow detected by the differential pressure sensor respectively, and s is the number of all current measurement data;

[0071] Because the data acquired by the differential pressure sensor for each phase of the oil-gas-water three-phase flow will change, therefore, the set X is classified and integrated, and the set X is expressed as:

[0072] X={A 1 ,A 2 ,A 3}

[0073] where A 1 ,A 2 ,A 3 Three sets of data sets are obtained for the differential pressure sensor detecting oil, water, and gas phases respectively.

specific Embodiment approach 3

[0075] This embodiment is a further description of Embodiment 2. The method of selecting 3 initial centers described in step 1 is:

[0076] First calculate the distance D between all current data values ​​according to the Euclidean distance:

[0077]

[0078] In the formula Obtain the converted current data value a for the differential pressure sensor to detect multiphase flow i and current data value a j the distance between

[0079] Then calculate the current data value a i and current data value a j (j≠i) distance accounted for current data value a j The sum V of distance ratios to all data:

[0080]

[0081] In the formula, l represents the serial number of the current data value. For the convenience of subsequent description, the matrix V is rewritten, and the matrix V is expressed as:

[0082]

[0083] In the formula, u represents the number of elements in the matrix V, arrange all the elements in the matrix V in ascending order, and the arranged matrix ...

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 belongs to the technical field of petroleum engineering, and particularly relates to an oil-gas-water three-phase flow phase state critical point recognition algorithm based on clustering analysis. The algorithm has the following steps: selecting three pieces of data from the current measurement data values, defining the data as an initial center M, and dividing the data into three classes by taking the M as the center; classifying the data except the M into a class represented by an initial central point closest to the data except the M, and calculating the sum Epre of absolutevalues of difference values of the whole group of data; extracting the data of the center position in the two types of data as a new center Mnew, re-classifying and calculating the sum Enew of the absolute values of the corresponding difference values, replacing M by utilizing the principle that the total exchange cost is negative, and naming the updated initial center as Mproc; and repeating thetwo steps until the updated initial center Mproc does not change any more or the number k of iterations reaches the maximum number kmax of iterations. The influence of critical point identification onthe measurement precision during oil-gas-water three-phase flow split-phase flow measurement is improved, and the metering precision of the oil-gas-water three-phase flow is improved.

Description

Technical field: [0001] The invention belongs to the technical field of petroleum engineering, and in particular relates to an identification algorithm of a phase state critical point of oil-gas-water three-phase flow based on k-medoids clustering analysis. Background technique: [0002] The flow measurement of each phase of oil-gas-water three-phase flow is an important parameter for evaluating oil well production in oilfields. The traditional multiphase flow measurement method is based on the single-phase flow measurement method, and there are many defects and disadvantages, which make the traditional multiphase flow measurement method low in accuracy and cannot provide effective support for subsequent oilfield production work. Separate measurement of oil-gas-water three-phase flow. This method separates oil-gas-water three-phase flow from oil-water and oil-gas, and uses the volumetric method to measure the flow of each phase. The operation is safe and reliable and the cos...

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
IPC IPC(8): G06F17/50G06F16/28G01F5/00
CPCG01F5/005G06F16/285
Inventor 韩连福王海霞付长凤刘兴斌邵丽艳杨学梅
Owner 华运隆腾机械制造有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com