CN110263291A - A kind of industrial data trend recognition methods and system - Google Patents
A kind of industrial data trend recognition methods and system Download PDFInfo
- Publication number
- CN110263291A CN110263291A CN201910456660.3A CN201910456660A CN110263291A CN 110263291 A CN110263291 A CN 110263291A CN 201910456660 A CN201910456660 A CN 201910456660A CN 110263291 A CN110263291 A CN 110263291A
- Authority
- CN
- China
- Prior art keywords
- array
- data
- value
- slope
- minimum
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Analysis (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of industrial data trend recognition methods, comprising: obtains industrial data；The industrial data of acquisition is pre-processed, original array is obtained, and extracts characteristic value therein, generates several groups feature group array；Polynomial curve fitting is carried out to original array, characteristic value array, obtains slope value corresponding with each array；According to the analysis rule of setting, the trend type of the industrial data in this time segment limit is calculated in conjunction with the corresponding slope value of each array.The present invention obtains the variation tendency of this group of industrial data by extracting characteristic value, the changing rule of binding characteristic value array, original array, Conjoint Analysis, and due to introducing the variation tendency of characteristic value, analysis result accuracy is high, and calculation amount is few；The variation tendency for analyzing obtained industrial data can be efficiently used into production management process.In addition, newly generated data can be introduced directly into trend identification process, trend identification process is made to be provided with real-time.
Description
Technical field
The present invention relates to polynomial curve fit technique fields, in particular to a kind of industrial data trend recognition methods
And system.
Background technique
PolynonmiaCurveFitter is a polynomial curve fit technique, is established in Apache Commons
On the basis math3.This class libraries provides calculating API very rich, while providing the side of a variety of data Fitting Analysis
Method, so that the analysis of mass data becomes very easy.
The data application generated in industrial production business process has a high potential, but presently, and industrial big data is being applied
There are still some technology barriers in analysis and value excavation, identify that trend is effective data analysing method from industrial data.
Summary of the invention
It is an object of that present invention to provide a kind of industrial data trend recognition methods and systems, by extracting characteristic value, in conjunction with
The changing rule of characteristic value array, original array, Conjoint Analysis obtains the variation tendency of this group of industrial data, due to introducing
The variation tendency of characteristic value, analysis result accuracy is high, and calculation amount is few；The variation tendency for analyzing obtained industrial data can have
Effect uses in production management process.In addition, newly generated data can be introduced directly into trend identification process, identify trend
Process is provided with real-time.
To reach above-mentioned purpose, in conjunction with Fig. 1, the present invention proposes a kind of industrial data trend recognition methods, the identification side
Method includes:
S1: at the beginning of according to offer and the end time, the industrial data in this time segment limit, the work are obtained
Industry data include corresponding timestamp.
S2: pre-processing the industrial data of acquisition, obtains original array, and extracts characteristic value therein, generates
Several groups feature group array, the characteristic value array include at least maximum array, minimum array, maximum value minimum.
The maximum array includes the local crest numerical value in this time segment limit in industrial data, the minimum number
Group includes the local trough numerical value in this time segment limit in industrial data, and the maximum is industrial number in this time segment limit
Greatest measure in, the minimum are the minimum value in this time segment limit in industrial data.
S3: polynomial curve fitting is carried out to original array, characteristic value array, obtains slope corresponding with each array
Value.
S4: it according to the analysis rule of setting, is calculated in this time segment limit in conjunction with the corresponding slope value of each array
Industrial data trend type.
Based on preceding method, the present invention further mentions a kind of industrial data trend identifying system, the identifying system include with
Lower module:
(1) for inputting the input equipment of starting and end time.
(2) at the beginning of being used for according to input and the end time obtains the mould of the industrial data in this time segment limit
Block.
(3) for pre-processing to the industrial data of acquisition, the module of original array is obtained.
(4) for extracting characteristic value therein from the original array of acquisition, the mould of several groups feature group array is generated
Block.
(5) it for carrying out polynomial curve fitting to original array, characteristic value array, obtains corresponding with each array oblique
The module of rate value.
(6) for the analysis rule according to setting, this period model is calculated in conjunction with the corresponding slope value of each array
The module of the trend type of industrial data in enclosing.
The industrial data generated in industrial production business process generally includes the attributes such as data type, numerical value, timestamp,
Such as the unit time output of production and operation result, unit time consumable quantity can be reacted etc., it can directly react production warp
The unit time yields of battalion's process, unit time discharge amount of exhaust gas etc., can electrically set in indirect reaction production management process
Voltage value, the performance number etc. of standby service condition.
By the trend of this kind of industrial data in analysis certain time period, can quickly and effectively understand in this period
Production and management.
First according to offer at the beginning of and the end time, obtain in this time segment limit want analysis industrial number
According to the industrial data includes corresponding timestamp.Preferably, the industrial data chosen at this time belongs to same class, with
Ensure precision of analysis.
Secondly, the industrial data to acquisition pre-processes, original array is obtained.Pretreatment includes data cleansing, data
Screening, Data Format Transform etc..The characteristic value in original array is extracted, several groups feature group array, the characteristic value are generated
Array includes at least maximum array, minimum array, maximum value minimum.
The maximum array includes the local crest numerical value in this time segment limit in industrial data, the minimum number
Group includes the local trough numerical value in this time segment limit in industrial data, and the maximum is industrial number in this time segment limit
Greatest measure in, the minimum are the minimum value in this time segment limit in industrial data.
Finally, carrying out polynomial curve fitting to original array, characteristic value array, slope corresponding with each array is obtained
Value.According to the analysis rule of setting, the industrial number in this time segment limit is calculated in conjunction with the corresponding slope value of each array
According to trend type.The trend type includes periodical, steady, the first rising, the second rising, third up and down.Specifically
, it is converted by analyzing maximum array, minimum array, the slope value of original array one by one, obtains local crest numerical value, office
The changing rule of portion's trough numerical value, raw value, then Conjoint Analysis judge that the variation of industrial data in this period becomes
Gesture.
When there are new data to generate, it can be introduced directly into original array tail portion, by judging former last bit data and newly drawing
The feature value attribute for entering data is conducted into corresponding characteristic value array, adjusts the slope of corresponding eigenvalue array, judges new
Influence of the data of generation to the variation tendency identified, the industrial data trend recognition methods for referring to the present invention are provided with reality
Shi Xing.
The above technical solution of the present invention, compared with existing, significant beneficial effect is:
1) present invention is obtained by extracting characteristic value, the changing rule of binding characteristic value array, original array, Conjoint Analysis
The variation tendency of this group of industrial data, due to introducing the variation tendency of characteristic value, analysis result accuracy is high, calculation amount
It is few.
2) variation tendency for the industrial data that analysis obtains can be efficiently used into production management process.
3) newly generated data can be introduced directly into trend identification process, and trend identification process is made to be provided with real-time.
It should be appreciated that as long as aforementioned concepts and all combinations additionally conceived described in greater detail below are at this
It can be viewed as a part of the subject matter of the disclosure in the case that the design of sample is not conflicting.In addition, required guarantor
All combinations of the theme of shield are considered as a part of the subject matter of the disclosure.
Can be more fully appreciated from the following description in conjunction with attached drawing present invention teach that the foregoing and other aspects, reality
Apply example and feature.The features and/or benefits of other additional aspects such as illustrative embodiments of the invention will be below
Description in it is obvious, or learnt in practice by the specific embodiment instructed according to the present invention.
Detailed description of the invention
Attached drawing is not intended to drawn to scale.In the accompanying drawings, identical or nearly identical group each of is shown in each figure
It can be indicated by the same numeral at part.For clarity, in each figure, not each component part is labeled.
Now, example will be passed through and the embodiments of various aspects of the invention is described in reference to the drawings, in which:
Fig. 1 is the flow chart of industrial data trend recognition methods of the invention.
Specific embodiment
In order to better understand the technical content of the present invention, special to lift specific embodiment and institute's accompanying drawings is cooperated to be described as follows.
In conjunction with Fig. 1, the present invention refers to that a kind of industrial data trend recognition methods, the recognition methods include:
S1: at the beginning of according to offer and the end time, the industrial data in this time segment limit, the work are obtained
Industry data include corresponding timestamp.
S2: pre-processing the industrial data of acquisition, obtains original array, and extracts characteristic value therein, generates
Several groups feature group array, the characteristic value array include at least maximum array, minimum array, maximum value minimum.
The maximum array includes the local crest numerical value in this time segment limit in industrial data, the minimum number
Group includes the local trough numerical value in this time segment limit in industrial data, and the maximum is industrial number in this time segment limit
Greatest measure in, the minimum are the minimum value in this time segment limit in industrial data.
Described to extract characteristic value therein in step S2, the process for generating several groups feature group array includes following step
It is rapid:
S21: maximum array, minimum array are generated, selects any one data in original array as maximum
With the initial value of minimum.
S22: by each data in original array, data adjacent thereto are made comparisons respectively, if the data are all larger than phase
Adjacent data, are added into maximum array, if the data are respectively less than adjacent data, are added into minimum array.
S23: by each data in original array respectively compared with maximum and minimum, if the data are greater than greatly
Value Data is updated to maximum, if the data are less than minimum Value Data, is updated to minimum.
The array data of acquisition maximum value maximum, minimum, minimum array data described in step S2 of the present invention,
Decomposition step is as follows:
If including n data in original array, each data are expressed as value [i], its corresponding timestamp is expressed as
Time [i],.
Putting in order for array is not limited to the processing of original array, it is assumed for convenience of description that from small to large according to i
Sequence original array is handled.
At that time, the Data Position in value was in 0, and the data of value [0] are assigned to maximum max, minimum
It is then compared by min in value [1], if value [0] is greater than value [1], value [0] is stored in maximum
In array maxvalue, and corresponding timestamp is stored in maxtimes, it is on the contrary by value [0] and its corresponding time
Stamp is stored in minimum array respectively in minvalue and mintimes.If it is greater than maximum max simultaneously, by its assignment
To max, if it is less than minimum min, it is assigned to minimum min.
At that time, the data in value are in 1 to the last one reciprocal position, allow value [i] and value [i-1] and
Value [i+1] compares, if value [i] is simultaneously greater than value [i-1] and value [i+1], value [i] is stored
It is stored in maximum array maxvalue, and corresponding time stamp data is stored in maxtimes, conversely, by value
[i] and its corresponding time stamp data are stored in minimum array respectively in minvalue and mintimes.If simultaneously
Value [i] is greater than maximum max, and value [i] is assigned to max, if value [i] is less than minimum min, by value
[i] is assigned to minimum min.
At that time, when the data in value are in last, value [n-1] and value [n-2] is allowed to be compared, if
When value [n-1] is greater than value [n-2], value [n-1] is stored in maximum array maxvalue, and will be corresponding
Time stamp data is stored in maxtimes, otherwise value [n-1] and its corresponding timestamp are stored in minimum number respectively
In group in minvalue and mintimes.If value [n-1] is greater than maximum max simultaneously, value [n-1] is assigned to
Value [n-1] is assigned to minimum min if value [n-1] is less than minimum min by max.
For example, original array is [1,2,3,4,5,4,3,4,3,6,5,6,2,4], 1 is set by max and min, according to preceding
Sequence is stated from front to back to handle the data in original array:
First data and second data are compared first, 1 less than 2, therefore first data 1 is put into minimum
In array, the value of max is updated to 2.
Second data is compared with first data, third data respectively again, 2 are greater than 1 less than 3, therefore second
A data are not classified as characteristic value.
Likewise, successively being handled the numerical value for coming non-first and last position in original array according to aforementioned process, can obtain
Out, maximum array is [5,4,6,6], and minimum array is [3,3,5,2], and in the process, max is updated to 6, min maintenance 1 not
Become.
Finally judge the size of last bit data 4 Yu preceding a data 2,4 are greater than 2, therefore are put into maximum array for 4, most
Whole maximum array is [5,4,6,6,4], and minimum array is [3,3,5,2], and max 6, min maintenance 1 are constant.
From the foregoing it will be appreciated that the present invention proposes for 2-3 numerical value to be set to a partial analysis array, in maximum array
Numerical value is local crest numerical value, and the numerical value in minimum array is local trough numerical value, in the feelings of not impact analysis result
Under condition, preliminary screening has been done to data, has had chosen the characteristic value Cooperative Analysis for capableing of reaction tendency variation, on the one hand, from complexity
Industrial array in filtered out wherein characteristic value, make analyze result be provided with feasibility and characteristic, on the other hand, reduce
Operand.
S3: polynomial curve fitting is carried out to original array, characteristic value array, obtains slope corresponding with each array
Value.
It is described that polynomial curve fitting is carried out to original array, characteristic value array in step S3, it obtains and each array pair
The process of the slope value answered the following steps are included:
In conjunction with maximum array, minimum array, original array and corresponding time list, using numerical value value as x-axis,
It using corresponding timestamp as y-axis, carries out curve fitting, seeks corresponding slope value.
Preferably, the curve matching includes univerible curve fitting.More preferred, using polynomial curve fit technique
Carry out the disposable curve matching of unitary.For example, carrying out the disposable curve matching of unitary using PolynonmiaCurveFitter.
S4: it according to the analysis rule of setting, is calculated in this time segment limit in conjunction with the corresponding slope value of each array
Industrial data trend type.
In some instances, the setting analysis rule includes:
If maximum array slope, minimum array slope, original array slope are respectively maxR, minR, medR.
Calculate the slope absolute value maxRV of maximum array, the slope absolute value minRV of minimum array, original array
Slope absolute value medRV.
Threshold value h is compared according to industrial data type set.
By comparing maximum array slope maxR, minimum array slope minR, original array slope medR, maximum
The slope absolute value maxRV of array, the slope absolute value minRV of minimum array, original array slope absolute value medRV with
The comparison threshold value h of setting, with the trend type of the determination industrial data.
It should be appreciated that comparing threshold value h is not fixed value, according to different analysis purposes and different data types, take
It is worth different, such as unit time output, h can be set to 0.1, when slope absolute value is less than 0.1, the oblique line
Variation tendency is extremely gentle, it can be assumed that at unit time output without significant change.
Further, the trend type includes periodical, steady, the first rising, the second rising, third up and down.
The division methods of trend type include a variety of, aforementioned trends type only wherein one according to the type of industrial data and analysis purpose
Kind.
Based on aforementioned trends Type division method, the present invention provides one of which to set analysis rule.
(1) if the slope absolute value minRV of the slope absolute value maxRV of maximum array, minimum array are small simultaneously
In the comparison threshold value h of setting, determine that the trend type of the industrial data is periodicity.
When the slope absolute value maxRV of maximum array, the slope absolute value minRV of minimum array are less than setting simultaneously
Comparison threshold value h, when such as 0.1, the numerical value change in maximum array and minimum array is little, i.e., the data in original array
The cyclic fluctuation between maximum and minimum determines that the trend type of the industrial data is periodicity at this time.
For example, original array be [1,2,6,2, Isosorbide-5-Nitrae, 6,2,1,6,3,1,2,6], maximum 6, minimum 1, greatly
Data in value array are 6, and the data in minimum array are 1, the slope of maximum array and minimum array it is oblique
Rate is 0.It can be seen that data multiplicity in original array, directly observation are difficult to judge variation tendency, by extracting maximum number
Group and minimum array can effectively exclude intermediate fluctuation data, allow users to clearly know the number in original array
It is in periodically variable according to being.
(2) if the slope absolute value medRV of maximum array slope maxR, original array are less than the comparison threshold value of setting
H, and minimum array slope minR is more than or equal to h, determines that the trend type of the industrial data is steady.
Maximum array slope, which is less than, compares threshold value h, illustrates that maximum array is in steady or downward trend, raw value
The slope absolute value of original array illustrates the variation of raw value array in moderate tone, and minimum array slope minR is greater than
Equal to h, illustrate minimum array in steady or ascendant trend, but the numerical value after rising is no more than maximum, former under such situation
Beginning array determines that the variation tendency of industrial data is that numerical value reaches unanimity, trend type is steady in convergence shape.
(3) if maximum array slope maxR is less than h, minimum array slope minR is more than or equal to h, and original array is oblique
Rate medR is greater than h, determines that the trend type of the industrial data is the first rising.
Maximum array slope maxR is less than h, illustrates maximum array in steady or downward trend, minimum array slope
MinR is more than or equal to h, illustrates that minimum array is in rising trend, and original array slope medR is greater than h, illustrates that original array is whole
It is in rising trend, determine that the trend type of the industrial data is the first rising, that is, the variation tendency of industrial data is wave at this time
Valley is gradually increasing, but the greatest measure after rising has boundary, is no more than former local crest value, and the first propradation is mainly
Refer to the variation tendency of valley value.
(4) if maximum array slope maxR is more than or equal to h, the slope absolute value minRV of minimum array is less than h,
The trend type for determining the industrial data is the second rising.
Maximum array slope maxR is more than or equal to h, illustrates that maximum array is in rising trend, the slope of minimum array
Absolute value minRV is less than h, illustrates minimum in moderate tone, third rises the variation tendency for being primarily referred to as crest value.
(5) if maximum array slope maxR is more than or equal to h, minimum array slope minR is more than or equal to h, determines institute
The trend type of industrial data is stated as third rising.
Maximum array slope maxR is more than or equal to h, illustrates that maximum array is in rising trend, minimum array slope
MinR is more than or equal to h, illustrates that minimum array is in rising trend, and industrial data is one apparent ascendant trend of whole presentation,
That is the second propradation refers to the unification and variety trend of wave crest and trough.
(6) if maximum array slope maxR is less than h, the slope absolute value minRV of minimum array is more than or equal to h,
And minimum array slope minR determines the trend type of the industrial data for decline less than 0.
Maximum array slope maxR is less than h, illustrate maximum array in steady or downward trend, minimum array it is oblique
Rate absolute value minRV is more than or equal to h, and minimum array slope minR illustrates that minimum is in be decreased obviously trend less than 0, whole
Downward trend is presented in a industrial data.
Previous example is one of analysis rule, and the setting of analysis rule is that binding analysis purpose and data type are determined
Fixed, for example, it is also possible to multiple comparison threshold ranges are set, keep analysis result more diversified etc..
When there are new data to generate, original array tail portion is introduced, the new feature value attribute for introducing data is judged, is led
Enter corresponding characteristic value array, adjusts the slope of corresponding eigenvalue array, judge that newly generated data become to the variation identified
The influence of gesture.
If original array of this analysis is the industrial data currently produced in the previous period, when there is new industry
When data generate, newly generated industrial data directly can be introduced into original array tail portion, judge former last bit data and new introducing
The two is directed respectively into corresponding characteristic value array by the feature value attribute of data, then adjusts the slope of corresponding eigenvalue array,
Influence of the newly generated data to the variation tendency identified is judged in conjunction with slope adjusted, due to only needing to rejudge original
Last bit data and the new feature value attribute for introducing data, calculation amount is small, and operation time is short, and the analysis of industrial data is made to be provided with reality
Shi Xing.
Based on preceding method, the present invention further mentions a kind of industrial data trend identifying system, the identifying system include with
Lower module:
(1) for inputting the input equipment of starting and end time.
(2) at the beginning of being used for according to input and the end time obtains the mould of the industrial data in this time segment limit
Block.
(3) for pre-processing to the industrial data of acquisition, the module of original array is obtained.
(4) for extracting characteristic value therein from the original array of acquisition, the mould of several groups feature group array is generated
Block.
(5) it for carrying out polynomial curve fitting to original array, characteristic value array, obtains corresponding with each array oblique
The module of rate value.
(6) for the analysis rule according to setting, this period model is calculated in conjunction with the corresponding slope value of each array
The module of the trend type of industrial data in enclosing.
Various aspects with reference to the accompanying drawings to describe the present invention in the disclosure, shown in the drawings of the embodiment of many explanations.
Embodiment of the disclosure need not be defined on including all aspects of the invention.It should be appreciated that a variety of designs and reality presented hereinbefore
Those of apply example, and describe in more detail below design and embodiment can in many ways in any one come it is real
It applies, this is because conception and embodiment disclosed in this invention are not limited to any embodiment.In addition, disclosed by the invention one
A little aspects can be used alone, or otherwise any appropriately combined use with disclosed by the invention.
Although the present invention has been disclosed as a preferred embodiment, however, it is not to limit the invention.Skill belonging to the present invention
Has usually intellectual in art field, without departing from the spirit and scope of the present invention, when can be used for a variety of modifications and variations.Cause
This, the scope of protection of the present invention is defined by those of the claims.
Claims (10)
1. a kind of industrial data trend recognition methods, which is characterized in that the recognition methods includes:
S1: at the beginning of according to offer and the end time, the industrial data in this time segment limit, the industry number are obtained
According to including corresponding timestamp；
S2: pre-processing the industrial data of acquisition, obtains original array, and extracts characteristic value therein, generates several
Group feature group array, the characteristic value array include at least maximum array, minimum array, maximum value minimum；
The maximum array includes the local crest numerical value in this time segment limit in industrial data, the minimum array packet
The local trough numerical value in this time segment limit in industrial data is included, the maximum is in industrial data in this time segment limit
Greatest measure, the minimum is the minimum value in this time segment limit in industrial data；
S3: polynomial curve fitting is carried out to original array, characteristic value array, obtains slope value corresponding with each array；
S4: according to the analysis rule of setting, the work in this time segment limit is calculated in conjunction with the corresponding slope value of each array
The trend type of industry data.
2. industrial data trend recognition methods according to claim 1, which is characterized in that described to extract in step S2
Characteristic value therein, generate several groups feature group array process the following steps are included:
Maximum array, minimum array are generated, selects any one data in original array as maximum and minimum
Initial value；
By each data in original array, data adjacent thereto are made comparisons respectively, if the data are all larger than adjacent number
According to being added into maximum array, if the data are respectively less than adjacent data, be added into minimum array；
By each data in original array respectively compared with maximum and minimum, if the data are greater than very big Value Data,
It is updated to maximum, if the data are less than minimum Value Data, is updated to minimum.
3. industrial data trend recognition methods according to claim 1, which is characterized in that described to original in step S3
Array, characteristic value array carry out polynomial curve fitting, and the process for obtaining slope value corresponding with each array includes following step
It is rapid:
In conjunction with maximum array, minimum array, original array and corresponding time list, using numerical value value as x-axis, with right
The timestamp answered is y-axis, carries out curve fitting, seeks corresponding slope value.
4. industrial data trend recognition methods according to claim 3, which is characterized in that the curve matching includes unitary
Curve matching.
5. industrial data trend recognition methods according to claim 4, which is characterized in that use polynomial curve fitting skill
Art carries out the disposable curve matching of unitary.
6. industrial data trend recognition methods according to claim 1, which is characterized in that the setting analysis rule packet
It includes:
If maximum array slope, minimum array slope, original array slope are respectively maxR, minR, medR；
Calculate the slope absolute value maxRV of maximum array, the slope absolute value minRV of minimum array, original array it is oblique
Rate absolute value medRV；
Threshold value h is compared according to industrial data type set, h is greater than 0.
By comparing maximum array slope maxR, minimum array slope minR, original array slope medR, maximum array
Slope absolute value maxRV, the slope absolute value minRV of minimum array, the slope absolute value medRV of original array and setting
Comparison threshold value h, with the trend type of the determination industrial data.
7. industrial data trend recognition methods according to claim 1, which is characterized in that the trend type includes the period
Property, it is steady, first rise, second rise, third up and down.
8. industrial data trend recognition methods according to claim 6, which is characterized in that the setting analysis rule also wraps
It includes:
If the slope absolute value minRV of the slope absolute value maxRV of maximum array, minimum array is less than setting simultaneously
Threshold value h is compared, determines that the trend type of the industrial data is periodicity；
If the slope absolute value medRV of maximum array slope maxR, original array is less than the comparison threshold value h of setting, and pole
Small value array slope minR is more than or equal to h, determines that the trend type of the industrial data is steady；
If maximum array slope maxR is less than h, minimum array slope minR is more than or equal to h, original array slope medR
Greater than h, determine that the trend type of the industrial data is the first rising；
If maximum array slope maxR is more than or equal to h, the slope absolute value minRV of minimum array is less than h, described in judgement
The trend type of industrial data is the second rising；
If maximum array slope maxR is more than or equal to h, minimum array slope minR is more than or equal to h, determines the industry
The trend type of data is third rising；
If maximum array slope maxR is less than h, the slope absolute value minRV of minimum array is more than or equal to h, and minimum
Array slope minR determines the trend type of the industrial data for decline less than 0.
9. industrial data trend recognition methods according to claim 1, which is characterized in that when there is new data to generate,
Original array tail portion is introduced, former last bit data and the new feature value attribute for introducing data is judged, is conducted into corresponding characteristic value
Array adjusts the slope of corresponding eigenvalue array, judges influence of the newly generated data to the variation tendency identified.
10. a kind of industrial data trend identifying system using industrial data trend recognition methods described in claim 1, special
Sign is that the identifying system includes:
For inputting the input equipment of starting and end time；
For at the beginning of according to input and the end time obtains the module of the industrial data in this time segment limit；
For pre-processing to the industrial data of acquisition, the module of original array is obtained；
For extracting characteristic value therein from the original array of acquisition, the module of several groups feature group array is generated；
For carrying out polynomial curve fitting to original array, characteristic value array, slope value corresponding with each array is obtained
Module；
For the analysis rule according to setting, the work in this time segment limit is calculated in conjunction with the corresponding slope value of each array
The module of the trend type of industry data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910456660.3A CN110263291B (en) | 2019-05-29 | 2019-05-29 | Industrial data trend identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910456660.3A CN110263291B (en) | 2019-05-29 | 2019-05-29 | Industrial data trend identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110263291A true CN110263291A (en) | 2019-09-20 |
| CN110263291B CN110263291B (en) | 2020-03-31 |
Family
ID=67915890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910456660.3A Active CN110263291B (en) | 2019-05-29 | 2019-05-29 | Industrial data trend identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110263291B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132722A (en) * | 2020-08-20 | 2020-12-25 | 彭涛 | Method and device for determining abnormal trend of number of government affair hotlines, electronic equipment and medium |
| CN112562039A (en) * | 2020-12-23 | 2021-03-26 | 平安银行股份有限公司 | Method and device for determining extreme value of longitudinal axis in trend graph |
| CN118013259A (en) * | 2024-04-09 | 2024-05-10 | 中国人民解放军海军工程大学 | Data analysis method based on non-contact measurement and related equipment thereof |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101223070B1 (en) * | 2011-06-10 | 2013-01-17 | 주식회사 에스비아이에스 | Apparatus for providing information of industry trend using weather information |
| CN103414409A (en) * | 2013-08-14 | 2013-11-27 | 浙江工业大学 | Stepping motor pre-subdivision driving method based on curve fitting |
| CN105573279A (en) * | 2015-12-31 | 2016-05-11 | 广州中国科学院先进技术研究所 | Method of monitoring industrial process based on sensor data |
| CN109118009A (en) * | 2018-08-24 | 2019-01-01 | 山东师范大学 | Time Series Forecasting Methods, system and medium based on polar coordinates fuzzy information granule |
| CN109410052A (en) * | 2018-09-28 | 2019-03-01 | 阿里巴巴集团控股有限公司 | A kind of data predication method and device |
-
2019
- 2019-05-29 CN CN201910456660.3A patent/CN110263291B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101223070B1 (en) * | 2011-06-10 | 2013-01-17 | 주식회사 에스비아이에스 | Apparatus for providing information of industry trend using weather information |
| CN103414409A (en) * | 2013-08-14 | 2013-11-27 | 浙江工业大学 | Stepping motor pre-subdivision driving method based on curve fitting |
| CN105573279A (en) * | 2015-12-31 | 2016-05-11 | 广州中国科学院先进技术研究所 | Method of monitoring industrial process based on sensor data |
| CN109118009A (en) * | 2018-08-24 | 2019-01-01 | 山东师范大学 | Time Series Forecasting Methods, system and medium based on polar coordinates fuzzy information granule |
| CN109410052A (en) * | 2018-09-28 | 2019-03-01 | 阿里巴巴集团控股有限公司 | A kind of data predication method and device |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112132722A (en) * | 2020-08-20 | 2020-12-25 | 彭涛 | Method and device for determining abnormal trend of number of government affair hotlines, electronic equipment and medium |
| CN112132722B (en) * | 2020-08-20 | 2023-12-26 | 彭涛 | Government hot line quantity trend abnormity determining method and device, electronic equipment and medium |
| CN112562039A (en) * | 2020-12-23 | 2021-03-26 | 平安银行股份有限公司 | Method and device for determining extreme value of longitudinal axis in trend graph |
| CN112562039B (en) * | 2020-12-23 | 2023-08-22 | 平安银行股份有限公司 | Method and device for determining extreme value of vertical axis in trend chart |
| CN118013259A (en) * | 2024-04-09 | 2024-05-10 | 中国人民解放军海军工程大学 | Data analysis method based on non-contact measurement and related equipment thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110263291B (en) | 2020-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110263291A (en) | A kind of industrial data trend recognition methods and system | |
| US20180260726A1 (en) | Analysis apparatus, analysis method, and non-transitory computer readable medium | |
| CN107357764B (en) | Data analysis method, electronic device, and computer storage medium | |
| CN106776566B (en) | Method and device for recognizing emotion vocabulary | |
| CN109891508A (en) | Single cell type detection method, device, equipment and storage medium | |
| CN110991739A (en) | Construction method and prediction method of industry electric quantity demand prediction model | |
| CN110647995A (en) | Rule training method, device, equipment and storage medium | |
| CN112364014A (en) | Data query method, device, server and storage medium | |
| CN117132135A (en) | Urban informatization management system and method based on digital twinning | |
| CN113723747A (en) | Analysis report generation method, electronic device and readable storage medium | |
| CN110032585B (en) | Time sequence double-layer symbolization method and device | |
| CN110263344B (en) | Text emotion analysis method, device and equipment based on hybrid model | |
| CN114858958B (en) | Method and device for analyzing mass spectrum data in quality evaluation and storage medium | |
| CN104615681B (en) | Text selection method and device | |
| CN111160393A (en) | Data-driven modularized modeling method for carrier rocket health assessment model | |
| CN115599842A (en) | Time series prediction system based on neural network method | |
| CN115409541A (en) | Cigarette brand data processing method based on data blood relationship | |
| CN110472292B (en) | Industrial equipment data simulation configuration system and method | |
| CN109284354B (en) | Script searching method and device, computer equipment and storage medium | |
| CN116523039B (en) | Continuous casting knowledge graph generation method and device, storage medium and terminal | |
| CN110941607A (en) | Dirty data identification method, device, equipment and storage medium | |
| US20230280318A1 (en) | Learning data producing method, waveform analysis device, waveform analysis method, and recording medium | |
| JP7258253B1 (en) | Normal model generation program, normal model generation device, and normal model generation method | |
| CN113822257B (en) | Abnormal point detection method based on combination of dimensionless features and virtual samples | |
| CN117952311A (en) | Quality prediction method, device, equipment and storage medium for injection molding machine |
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 | ||
| CP02 | Change in the address of a patent holder |
Address after: Room a808, World Trade Center building, 67 Shanxi Road, Gulou District, Nanjing, Jiangsu 210000 Patentee after: LUCULENT SMART TECHNOLOGIES Co.,Ltd. Address before: 211100 langkun wisdom Park, No.23 Changqing street, Jiangning District, Nanjing City, Jiangsu Province Patentee before: LUCULENT SMART TECHNOLOGIES Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: Floor 31, Asia Pacific business building, No. 2 Hanzhong Road, Gulou District, Nanjing, Jiangsu 210005 Patentee after: LUCULENT SMART TECHNOLOGIES CO.,LTD. Address before: Room a808, World Trade Center building, 67 Shanxi Road, Gulou District, Nanjing, Jiangsu 210000 Patentee before: LUCULENT SMART TECHNOLOGIES CO.,LTD. |
|
| CP02 | Change in the address of a patent holder |