CN108596489A - A kind of experimental parameter clustering method towards dredging yield - Google Patents
A kind of experimental parameter clustering method towards dredging yield Download PDFInfo
- Publication number
- CN108596489A CN108596489A CN201810376685.8A CN201810376685A CN108596489A CN 108596489 A CN108596489 A CN 108596489A CN 201810376685 A CN201810376685 A CN 201810376685A CN 108596489 A CN108596489 A CN 108596489A
- Authority
- CN
- China
- Prior art keywords
- similarity
- yield
- experimental parameter
- attribute
- dredging
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of experimental parameter clustering methods towards dredging yield, include the following steps:(1) data information of yield and experimental parameter in dredging operation is collected, and establishes corresponding sample matrix;(2) parametric data in sample matrix is standardized;(3)The experimental parameter of selection and yield are carried out to the calculating of similarity;(4)The similarity result calculated is analyzed, determines the correlation between each experimental parameter and yield.The attainable advantageous effect of institute of the invention:A kind of method is provided for the optimizing research of dredging operation yield, achievees the purpose that high efficiency, high yield, is of great significance to the correlation research of yield in dredging operation and experimental parameter.
Description
Technical field
The present invention relates to the applications that the dredging yield based on clustering is determined with experimental parameter correlation, belong to dredger
Journey field.
Background technology
Dredging work is the big event of water conservancy marine traffic engineering, to progress of human society, enhancement of environment and economic development
Effect is very great.Modern dredging work relies primarily on dredger to carry out, and process control parameter is numerous, makees in dredging at this stage
In industry, mainly based on manual operation.Even veteran operating personnel lead to qualitative survey since influence factor is numerous
Relative difficulty is tried, dredging operation is constantly in high energy consumption, low yield state.Therefore, it searches and the relevant work of dredging operation yield
Skill parameter is particularly important.High yield, the low energy consumption for realizing dredging operation, reduce Operating Complexity, improve dredging operation
The degree of automation seems particularly urgent in today.
Invention content
It is an object of the invention to overcome drawbacks described above, using clustering, a kind of technique towards dredging yield is provided
Parameter clustering method, the correlation between the yield and experimental parameter of dredging operation are determined, and screening influences dredging
An important factor for operation yield, realizes high efficiency, the purpose of high yield, improves dredging the degree of automation.
To achieve the above object, the technical solution adopted in the present invention is:A kind of experimental parameter towards dredging yield is poly-
Alanysis method, includes the following steps:
Step (1):Collect dredging operation in yield and experimental parameter data information, determine p situational variables, list because
Variable and independent variable sample matrix;Wherein, p is positive integer;
Step (2):Associated parametric data is standardized;
Step (3):Calculating to each experimental parameter and yield progress similarity of selection;
Step (4):The similarity result calculated is analyzed, is determined related between each experimental parameter and yield
Property.
Sample matrix in above-mentioned steps (1) is as follows:
If to p independent variable x1,x2,…,xpWith 1 dependent variable y1It has gone n times observation, has remembered independent variable and dependent variable respectively
The data matrix of " sample point × variable " type be:
Remember that sample matrix is
Matrix standardization is as follows in above-mentioned steps (2):
The matrix that sample matrix is standardized after transformation is as follows,
Wherein, r=1,2 ..., n
HereFor independent variable xpObservation minimum value,For independent variable xpThe pole of observation
Difference;For dependent variable y1Observation minimum value,For dependent variable y1Observation it is very poor.Through
After crossing very poor normalization transformation, matrix FREach element value between 0~1.
The calculating of similarity is as follows in above-mentioned steps (3):
Any one independent variable after being standardized is chosen, such asIt is two n dimensions
Data object, then xjWith y1Between similarity calculating can divide three steps calculate:
The first step compares, and measures the separation degree between single attribute.It can be calculated by following formula:
Wherein, RjIndicate the variance of j-th of attribute;
Second step, it is comprehensive, the measurement results between each attribute are integrated, are formed to whole measurement results.It can be under
Formula calculates:
Wherein,Indicate the weight of j-th of attribute.
Third walks, and result is mapped between [0,1] by conversion.F indicates transfer function, can be calculated by following formula:
The first step, which calculates single attributes similarity, is laid the groundwork for second step synthesis, from single attributes similarity it can be seen that every
The similitude of group data, and comprehensive function is the synthesis to overall data, cannot therefrom find out the similitude between single group data.
Similarity result is analyzed as follows in the step (4):
In the present invention, single attributes similarity that step (3) is calculated is the separation degree for weighing single attribute.Work as calculating
Be separation degree when, single attributes similarity value range is entire set of real numbers, and two elements are more similar, then value is smaller.
It is similar with list attributes similarity since synthesized attribute is held to the comprehensive of each attribute, when single attributes similarity
When what is calculated is separation degree, two elements are more similar, then value is smaller.
The similarity finally calculated is the mapping to synthesized attribute.The value range of similarity is [0,1], the size of value
Indicate the similarity degree between element.I.e. when two elements are similar samples, the result of calculation of similarity more levels off to 1;When
When two element dissmilarities, the result of calculation of similarity more tends to 0.
The beneficial effects of the invention are as follows:The present invention utilizes the similarity in clustering to define, by calculating dredging operation
Yield and experimental parameter between similarity, determine the correlation between yield and experimental parameter.Can be dredging operation yield
Optimizing research a kind of method is provided, achieve the purpose that high efficiency, high yield, to the phase of yield in dredging operation and experimental parameter
Closing property is determined and is of great significance.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Specific implementation mode
The invention will be further described with example below in conjunction with the accompanying drawings.Following instance is only used for clearly illustrating this
The technical solution of invention, and not intended to limit the protection scope of the present invention.
A kind of experimental parameter clustering method towards dredging yield, includes the following steps:
Step (1):Collect dredging operation in yield and experimental parameter data information, determine p situational variables, list because
Variable and independent variable sample matrix;Wherein, p is positive integer;
Step (2):Associated parametric data is standardized;
Step (3):Calculating to each experimental parameter and yield progress similarity of selection;
Step (4):The similarity result calculated is analyzed, is determined related between each experimental parameter and yield
Property.
Sample matrix in above-mentioned steps (1) is as follows:
If to p independent variable x1,x2,…,xpWith 1 dependent variable y1It has gone n times observation, has remembered independent variable and dependent variable respectively
The data matrix of " sample point × variable " type be:
Remember that sample matrix is
Matrix standardization is as follows in above-mentioned steps (2):
The matrix that sample matrix is standardized after transformation is as follows,
Wherein, r=1,2 ..., n
HereFor independent variable xpObservation minimum value,For independent variable xpThe pole of observation
Difference;For dependent variable y1Observation minimum value,For dependent variable y1Observation it is very poor.Through
After crossing very poor normalization transformation, matrix FREach element value between 0~1.
The calculating of similarity is as follows in above-mentioned steps (3):
Any one independent variable after being standardized is chosen, such asIt is two n dimensions
Data object, then xjWith y1Between similarity calculating can divide three steps calculate:
The first step compares, and measures the separation degree between single attribute.It can be calculated by following formula:
Wherein, RjIndicate the variance of j-th of attribute;
Second step, it is comprehensive, the measurement results between each attribute are integrated, are formed to whole measurement results.It can be under
Formula calculates:
Wherein,Indicate the weight of j-th of attribute.
Third walks, and result is mapped between [0,1] by conversion.F indicates transfer function, can be calculated by following formula:
The first step, which calculates single attributes similarity, is laid the groundwork for second step synthesis, from single attributes similarity it can be seen that every
The similitude of group data, and comprehensive function is the synthesis to overall data, cannot therefrom find out the similitude between single group data.
Similarity result is analyzed as follows in the step (4):
In the present invention, single attributes similarity that step (3) is calculated is the separation degree for weighing single attribute.Work as calculating
Be separation degree when, single attributes similarity value range is entire set of real numbers, and two elements are more similar, then value is smaller.
It is similar with list attributes similarity since synthesized attribute is held to the comprehensive of each attribute, when single attributes similarity
When what is calculated is separation degree, two elements are more similar, then value is smaller.
The similarity finally calculated is the mapping to synthesized attribute.The value range of similarity is [0,1], the size of value
Indicate the similarity degree between element.I.e. when two elements are similar samples, the result of calculation of similarity more levels off to 1;When
When two element dissmilarities, the result of calculation of similarity more tends to 0.
It is illustrated below using " day lion number " cutter suction dredger site operation real-time data collection as data source
Step (1):Collect the data information of yield and each experimental parameter in dredging operation, in this example, the work of selection
Skill parameter has pipeline mean concentration, flow rate of pipelines, exit velocity, reamer to cut mud thickness, reamer forward travel distance, transverse moving speed, strand
Knife depth, totally 7 parameters.(selected parameter is only for reference)
Array data is randomly selected from " day lion " number mass data collected, and establishes corresponding data matrix F, such as table
Shown in 1;
1 yield of table and some processes parametric data
Step (2):Sample matrix is standardized.
By step (2), the matrix in step (1) is standardized, this step can utilize MATLAB to call
Rscore functions are calculated.
Matrix after standardization is as shown in table 2.
2 yield of table and experimental parameter standardized data
Step (3):Carry out the calculating of similarity with yield respectively to the experimental parameter of selection.
A. the variance of selected each experimental parameter is calculated, as shown in table 3:
The variance of 3 each experimental parameter of table
B. single attributes similarity is calculated, can be obtained by formula (5):(only enumerating one of experimental parameter herein), obtains
The results are shown in Table 4:
Single attributes similarity of 4 pipeline mean concentration of table
C. weight is calculatedIt can be calculated by formula (7), the results are shown in Table 5.
The weight of 5 each experimental parameter of table
D. comprehensive function Mn is calculated.It can be calculated by formula (6).Acquired results are as shown in table 6.
The synthesized attribute of 6 each experimental parameter of table
E. the similarity between each experimental parameter is calculated.Formula (8) calculates.Acquired results are as shown in table 7.
Similarity between 7 experimental parameter of table and yield
Step (4) analyzes the similarity result calculated, determines the correlation between each experimental parameter and yield.
As can be seen from Table 4, what single attributes similarity calculated in this example is separation degree, each technique after standardization
Parameter and the numerical value of yield closer to when, single attributes similarity value is smaller.
By the result and claims 6 of 6 calculated synthesized attribute of table, it can be deduced that conclusion:The calculating of synthesized attribute
As a result value is bigger, and two element sepatation degree are bigger, then two element similitudes are smaller.
What the similarity that table 7 is calculated indicated is the similarity degree between two elements, the more similar then value of two elements
It is bigger.Therefore, by the result in table 7 it is found that in the experimental parameter chosen above, the similarity of transverse moving speed and yield is most
Height is followed successively by pipeline mean concentration, reamer forward travel distance, exit velocity, flow rate of pipelines, reamer depth, reamer and cuts mud thickness.
The above is only a preferred embodiment of the present invention, it should be pointed out that:For the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvement and deformations are also answered
It is considered as protection scope of the present invention.
Claims (6)
1. a kind of experimental parameter clustering method towards dredging yield, which is characterized in that include the following steps:
Step (1):The data information for collecting yield and experimental parameter in dredging operation, determines p situational variables, lists dependent variable
With independent variable sample matrix;Wherein, p is positive integer;
Step (2):Associated parametric data is standardized;
Step (3):Calculating to each experimental parameter and yield progress similarity of selection;
Step (4):The similarity result calculated is analyzed, determines the correlation between each experimental parameter and yield.
2. a kind of experimental parameter clustering method towards dredging yield according to claim 1, which is characterized in that institute
It states and lists dependent variable in step (1) and be specially with independent variable sample matrix:If to p independent variable x1,x2,…,xpWith 1 because becoming
Measure y1It has gone n times observation, has remembered that the data matrix of independent variable and " sample point × variable " type of dependent variable is as follows respectively:
Remember that sample matrix is
Wherein, aijFor xjIth measured value, br1For y1The r times measured value.
3. a kind of experimental parameter clustering method towards dredging yield according to claim 2, which is characterized in that institute
Step (2) is stated to be standardized specially associated parametric data:
Sample matrix in step (1) is standardized transformation, the matrix after transformation is as follows,
Wherein,
For independent variable xpObservation minimum value,For independent variable xpObservation
It is very poor;For dependent variable yqObservation minimum value,For dependent variable y1Observation it is very poor.
4. a kind of experimental parameter clustering method towards dredging yield according to claim 3, which is characterized in that institute
Stating calculating of each experimental parameter in step (3) to selection respectively with yield progress similarity is specially:
Any one independent variable after being standardized is chosen,It is two n dimension datas pair
As calculating xjWith y1Between similarity.
5. a kind of experimental parameter clustering method towards dredging yield according to claim 4, which is characterized in that institute
It states and calculates x in step (3)jWith y1Between similarity be divided into three steps, it is specific as follows:
The first step compares, and measures the separation degree between single attribute, is calculated by formula (5):
Wherein, RjIndicate the variance of j-th of attribute;For by xjWith y1Between separation degree;
Second step, it is comprehensive, the measurement results between each attribute are integrated, are calculated by formula (6):
Wherein,Indicate the weight of j-th of attribute, Mn(x) it is synthesis result
Third walks, conversion, and between synthesis result is mapped to [0,1], f indicates transfer function, is calculated by formula (8):
6. a kind of experimental parameter clustering method towards dredging yield according to claim 5, which is characterized in that institute
It states and similarity result is analyzed specially in step (4):
In the step (3),
For similarity analysis between single attribute, it is the separation degree for weighing single attribute to measure single attributes similarity, works as calculating
Be separation degree when, single attributes similarity value range is entire set of real numbers, and two elements are got over mutually from then value is bigger;
Synthesis result is analyzed, since synthesized attribute is held to the comprehensive of each attribute, with list attributes similarity alanysis
Method is identical, when single attributes similarity calculate be separation degree when, two elements more detach, then value is bigger;
To synthesis result map analysis, the value range of similarity is [0,1], and the size of value indicates the similar journey between element
Degree, when two elements are similar samples, the result of calculation of similarity more levels off to 1;When two element dissmilarities, similarity
Result of calculation more tend to 0.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810376685.8A CN108596489A (en) | 2018-04-25 | 2018-04-25 | A kind of experimental parameter clustering method towards dredging yield |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810376685.8A CN108596489A (en) | 2018-04-25 | 2018-04-25 | A kind of experimental parameter clustering method towards dredging yield |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108596489A true CN108596489A (en) | 2018-09-28 |
Family
ID=63609599
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810376685.8A Pending CN108596489A (en) | 2018-04-25 | 2018-04-25 | A kind of experimental parameter clustering method towards dredging yield |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108596489A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104850910A (en) * | 2015-05-25 | 2015-08-19 | 河海大学常州校区 | Dredging technical parameter multi-attribute decision analysis method based on OWA |
CN105718426A (en) * | 2016-01-22 | 2016-06-29 | 河海大学常州校区 | Dredging output mathematical model building method based on multiple linear regression analysis |
CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
CN107391939A (en) * | 2017-07-25 | 2017-11-24 | 河海大学 | A kind of basin similitude comprehensive evaluation index computational methods of quantization |
-
2018
- 2018-04-25 CN CN201810376685.8A patent/CN108596489A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104850910A (en) * | 2015-05-25 | 2015-08-19 | 河海大学常州校区 | Dredging technical parameter multi-attribute decision analysis method based on OWA |
CN105718426A (en) * | 2016-01-22 | 2016-06-29 | 河海大学常州校区 | Dredging output mathematical model building method based on multiple linear regression analysis |
CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
CN107391939A (en) * | 2017-07-25 | 2017-11-24 | 河海大学 | A kind of basin similitude comprehensive evaluation index computational methods of quantization |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104239970B (en) | A kind of conductor galloping method for prewarning risk based on Adaboost | |
CN104899437B (en) | Heavy rainfall type landslide disaster method for early warning | |
CN107341497A (en) | The unbalanced weighting data streams Ensemble classifier Forecasting Methodology of sampling is risen with reference to selectivity | |
CN108550259A (en) | Congestion in road judgment method, terminal device and computer readable storage medium | |
CN108573285A (en) | A kind of experimental parameter clustering method towards dredging energy consumption | |
CN105241465B (en) | A kind of method of road renewal | |
CN110059963A (en) | A kind of tunnel risk evaluating method based on fuzzy polymorphism Bayesian network | |
CN108985380B (en) | Point switch fault identification method based on cluster integration | |
CN108492557A (en) | Highway jam level judgment method based on multi-model fusion | |
CN110285877B (en) | Train real-time positioning tracking and speed calculating method based on Spark Streaming | |
CN110222701A (en) | A kind of bridge defect automatic identifying method | |
CN104807589B (en) | A kind of ONLINE RECOGNITION method collecting flow pattern of gas-liquid two-phase flow in defeated-riser systems | |
CN109343154A (en) | Landing Tropical Cyclone Precipitation forecast system | |
CN110188797A (en) | A kind of intelligent automobile method for rapidly testing based on Bayes's optimization | |
CN117057616B (en) | Water conservancy monitoring method and system based on digital twin | |
CN103903441A (en) | Road traffic state distinguishing method based on semi-supervised learning | |
CN107230350A (en) | A kind of urban transportation amount acquisition methods based on bayonet socket Yu mobile phone flow call bill data | |
CN103927547B (en) | Method for rapidly analyzing rock debris fluorescence image | |
CN105869398A (en) | Road traffic open degree judging method based on K-means cluster | |
CN103278616A (en) | Multi-factor method for rapidly evaluating soil corrosivity | |
CN114550498B (en) | Intelligent early warning system for ship deviation channel driven by massive AIS data | |
CN106327802A (en) | Fatigue driving identification device, system and method | |
CN114003623A (en) | Similar path typhoon retrieval method | |
CN108596489A (en) | A kind of experimental parameter clustering method towards dredging yield | |
CN109543144A (en) | Dredging yield or energy consumption experimental parameter screening technique based on main substrate analytic approach |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |
|
RJ01 | Rejection of invention patent application after publication |