CN106372412B - Interlocking equipment energy efficiency analysis method for air - Google Patents
Interlocking equipment energy efficiency analysis method for air Download PDFInfo
- Publication number
- CN106372412B CN106372412B CN201610776691.3A CN201610776691A CN106372412B CN 106372412 B CN106372412 B CN 106372412B CN 201610776691 A CN201610776691 A CN 201610776691A CN 106372412 B CN106372412 B CN 106372412B
- Authority
- CN
- China
- Prior art keywords
- equipment
- energy consumption
- relevant parameter
- frame
- data
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
-
- 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
- Y02P80/00—Climate change mitigation technologies for sector-wide applications
- Y02P80/10—Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
-
- 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
- Y02P90/80—Management or planning
- Y02P90/82—Energy audits or management systems therefor
Abstract
The present invention relates to a kind of interlocking equipment energy efficiency analysis method for air, comprising: data sampling equidistantly samples whens waiting the energy consumption relevant parameter of at least two equipment in interlocking equipment, obtains the sampled data in certain time；Data prediction carries out data prediction to actual measurement sampled data, removes abnormal data therein, completion missing data；Framing operation is divided into several frames to the pretreated sampled data of each equipment by the way of overlapping segmentation, and every frame corresponds to a duration equal period, and the frame of each interframe moves length and is less than frame length；Related coefficient calculates, and based on the sampled data contained by each frame, calculates correlation coefficient r of the energy consumption relevant parameter in each frame of at least two equipment in interlocking equipmentXY；Abnormal determination, the frame using related coefficient lower than decision threshold is as abnormal frame.The present invention can be used for the reason of assisting investigation to form linkage sexual abnormality, instructs equipment to be adjusted correspondingly, system overall operation can be made more stable and energy saving by the method.
Description
Technical field
The present invention relates to a kind of interlocking equipment energy efficiency analysis method for air.
Background technique
Carrying out the monitoring analysis of energy consumption relevant parameter to equipment each in production equipment in modern enterprise production is to understand and control
The effective ways of production equipment entirety energy consumption relevant parameter, but the prior art monitors often the energy consumption relevant parameter of production equipment
Be it is isolated, i.e., independent energy consumption relevant parameter monitoring analysis only is carried out to equipment and has ignored the interaction relation between equipment,
Although the whole energy consumption relevant parameter of equipment can be reduced to a certain extent using this method, due to not accounting for linking
Influence between when equipment is run, therefore its energy consumption relevant parameter effect of optimization is unsatisfactory.
Summary of the invention
In order to overcome the drawbacks described above under the prior art, the purpose of the present invention is to provide a kind of interlocking equipment Energy Efficiency Analysis
Method, energy consumption relevant parameter linkage anomalous identification and energy saving of system analysis for interlocking equipment in production system, analyzes institute
The energy consumption relevant parameter linkage exception obtained and the energy saving space amount calculated can be used for assisting the original of investigation formation linkage sexual abnormality
Cause, guidance are adjusted correspondingly equipment, system overall operation can be made more stable and energy saving by the method.
The technical scheme is that
A kind of interlocking equipment energy efficiency analysis method for air, comprising:
Data sampling, to the energy consumption relevant parameters of at least two equipment in interlocking equipment, (including voltage, electric current, power etc. can
For calculating the parameter of direct consumption energy consumption) equidistantly sampled, the sampling time point of each equipment is identical, obtains in certain time
Survey sampled data；
Data prediction carries out data prediction, including exceptional value cleaning and Supplementing Data to actual measurement sampled data, removes
Abnormal data therein, completion missing data form pretreated sampled data；
Framing operation is divided into several frames, every frame by the way of overlapping segmentation to the pretreated sampled data of each equipment
The period equal corresponding to a duration includes the pretreated sampled data of whole for (containing endpoint) in the corresponding period,
The frame of each interframe move length (consecutive frame originates the time interval between time point) be less than frame length (the every frame corresponding period when
It is long)；
Related coefficient calculates, and based on the sampled data contained by each frame, calculates the energy of at least two equipment in interlocking equipment
Consume relevant parameter each frame (each frame corresponding period in) Pearson correlation coefficient (Pearson product-moment correlation coefficient,
Pearson product-moment correlation coefficient) rXY, wherein X, Y respectively represent the energy consumption of two equipment
Relevant parameter；
Pearson correlation coefficient is lower than decision threshold according to the Pearson correlation coefficient decision threshold of setting by abnormal determination
The frame of value is as abnormal frame, and the corresponding period is as abnormal period.
Further, the mode of the Supplementing Data are as follows: arrange sampled data according to the time that sampled data generates
Sequence has following two situation:
A) when in interlocking equipment with the presence of the sampled data shortage of data situation of an equipment, using model completion method pair
Missing data carries out completion, if the energy consumption relevant parameter of related two interlocking equipments is respectively the first equipment energy consumption relevant parameter Y
With the second equipment energy consumption relevant parameter X, the energy consumption relevant parameter of the first equipment exists in t moment to be lacked, and takes t moment nearby n
The data value that X, Y are not lacked, the formula that the model completion method carries out sampled data completion are as follows:
Wherein:It is the estimated value of the first equipment t moment energy consumption relevant parameter missing values；For the second equipment t moment
Energy consumption relevant parameter sampled data；For error term；、For related coefficient, calculated according to following equation:
In formula:The energy consumption relevant parameter mean value of respectively the first equipment and the second equipment, xiIt is the second equipment in i
The power consumption values of sampled point, yiIt is the first equipment in the power consumption values of i sampled point, i is positive integer, and i=1 ..., n, n are sampled point
Number,(generally n is determined by the minimum production period)；
B) when related two interlocking equipments are when synchronization exists simultaneously sampled data missing, distinguished using the method for moving average
Completion is carried out to interlocking equipment sampled data, the formula that the method for moving average carries out sampled data completion is as follows:
Wherein, m is integer；It is the corresponding equipment energy consumption relevant parameter of t moment；It is equipmentMoment energy consumption phase
The weighted value of parameter is closed,。
Further, it will be determined as exceptional value, the exception less than 0 or greater than the energy consumption relevant parameter of equipment rated value
Value cleaning method particularly includes: exceptional value or right is replaced with the mean value of multiple energy consumption relevant parameters near exceptional value
Exceptional value assigns null value, and the position for assigning null value carries out assignment to exceptional value position using the method for the Supplementing Data.
Further, the framing operation is realized with the method that moveable finite length window is weighted,
Here it is multiplying s (n) with certain window function w (n), so that windowing signal is formed, formula are as follows:
Wherein, wherein w is frame number,Energy consumption relevant parameter when for equipment w frame after windowing process；S (n) is the
The energy consumption relevant parameter sampled data for being included in n frame；W (n) is window function.
The calculation method of the window function w (n) is as follows:
Energy consumption relevant parameter sampled data of the equipment on a certain frame can be expressed as a dimension after windowing process
Group S, formula are as follows:
Wherein, P is the length of the one-dimension array.
Further, further include the steps that equipment energy consumption relevant parameter sampled data being converted to two-dimensional array, described
One-dimension array is converted in the transformation of the two-dimensional array, and remaining sampled data is not enough to fill up the two-dimensional array last line
When, then use the energy consumption relevant parameter of sampled data most endLast line is filled up, the expression formula of two-dimensional array is as follows after conversion:
Wherein, N is frame length, and M is frame shifting, M < N；W is frame number,。
Further, the related coefficientCalculation formula are as follows:
Wherein,、AndIt is that the criterion score of the second equipment energy consumption relevant parameter, energy consumption relevant parameter are average respectively
Value and energy consumption relevant parameter standard deviation；、AndIt is criterion score, the energy of the first equipment energy consumption relevant parameter respectively
Consume relevant parameter average value and energy consumption relevant parameter standard deviation；
It is calculated by related coefficient, the energy consumption relevant parameter sampled data of associate device is transformed to the one of one group of related coefficient
Dimension group, each related coefficientIndicate that interlocking equipment energy consumption in the i-th frame is related
The correlation of parameter, the related coefficientValue range be [- 1,1].
Further, interlocking equipment energy consumption relevant parameter correlation prediction method particularly includes: whenWhen indicate connection
Dynamic equipment energy consumption relevant parameter X and Y is positively correlated, whenWhen indicate that interlocking equipment energy consumption relevant parameter X and Y are negatively correlated, whenWhen indicate interlocking equipment energy consumption relevant parameter X and Y it is uncorrelated；WhenWhen indicate interlocking equipment energy consumption correlation ginseng
Number X and Y perfect positive correlation, whenWhen indicate energy consumption relevant parameter X and Y interlocking equipment perfect negative correlation.
Further, the calculation formula of the decision threshold T are as follows:
Wherein, whereinFor the first quartile of related coefficient data in box figure,It is related equal to by all energy consumptions
Number after the ascending arrangement of parameter values at a quarter；IQR is quartile range；
The calculation formula of the range IQR are as follows:
Wherein,For third quartile,Equal to by four points after all ascending arrangements of energy consumption relevant parameter numerical value
Three at number.
Further, further include the steps that energy saving space amount calculates, calculate and obtain in abnormal time section relative to reasonable
The energy saving space amount of energy consumption.
Further, the energy saving space amount calculates method particularly includes: related with the energy consumption of the lower equipment of relative value
Based on parameter, reasonable energy consumption of the sum of the reasonable energy consumption of abnormal period equipment as the abnormal period equipment is calculated, with different
The actual measurement energy consumption of normal period equipment and the difference of reasonable energy consumption, as the energy saving space amount of the abnormal period, when certain time memory
In one or more abnormal periods, with the energy saving space amount of an abnormal period or the energy saving space of multiple abnormal period
The sum of amount is total energy saving space amount of the time.
The invention has the benefit that
1. the framing operation that analysis model of the present invention uses considers the timing and production system mechanism of production process, in point
Frame shifting appropriate is carried out in frame operation, not only reduces the computation complexity of correlation analysis, and such operation has been stabilized point
Result is analysed to the susceptibility of abnormal data, there is stronger robustness；
2. method of the invention is comprehensively described the linkage of interlocking equipment, not only the connection in terms of correlation
Dynamic property has carried out mathematics measurement, and determines linkage sexual abnormality using box primitive reason in linkage abnormal determination, effectively
It ensure that the validity and reliability of analysis result；
3. the energy saving space amount that method of the invention is analyzed resulting energy consumption relevant parameter linkage exception and calculated is available
The reason of forming linkage sexual abnormality is checked in auxiliary, guidance is adjusted correspondingly equipment, and interlocking equipment can be made to run shape
State be improved significantly, energy consumption relevant parameter level is substantially reduced, achieve the purpose that system steadily with optimization run.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is the schematic diagram of the overlapping segmentation method framing operation of the present invention；
Fig. 3 is box figure schematic illustration.
Specific embodiment
Referring to Fig. 1, the invention discloses a kind of interlocking equipment energy efficiency analysis method for air, comprising:
(1) data acquisition and pretreatment
Due to production scene work condition environment complexity, it is different more or less to there are shortage of data, data in data acquisition
Situations such as normal.Therefore, it needs to carry out the live initial data of interlocking equipment in production system before analyzing data
Pretreatment operation.The method of the pretreatment operation is according to history energy consumption relevant parameter sampled data combination production system mechanism
Supplementing Data and exceptional value cleaning are carried out to original energy consumption relevant parameter sampled data.
1) Supplementing Data
Supplementing Data of the present invention is ranked up according to the time that energy consumption relevant parameter sampled data generates, using rolling average
Method and model completion method.
A) when in interlocking equipment with the presence of the sampled data shortage of data situation of an equipment, using model completion method pair
Missing data carries out completion, if the energy consumption relevant parameter of related two interlocking equipments is respectively the first equipment energy consumption relevant parameter Y
With the second equipment energy consumption relevant parameter X, the energy consumption relevant parameter of the first equipment exists in t moment to be lacked, and takes t moment nearby n
The data value that X, Y are not lacked, the formula that the model completion method carries out sampled data completion are as follows:
Wherein:It is the estimated value of the first equipment t moment energy consumption relevant parameter missing values；For the second equipment t moment
Energy consumption relevant parameter sampled data；For error term；、For related coefficient, calculated according to following equation:
In formula:The energy consumption relevant parameter mean value of respectively the first equipment and the second equipment, xiIt is the second equipment in i
The power consumption values of sampled point, yiIt is the first equipment in the power consumption values of i sampled point, i is positive integer, and i=1 ..., n, n are sampled point
Number,(generally n is determined by the minimum production period)；
B) when related two interlocking equipments are when synchronization exists simultaneously sampled data missing, distinguished using the method for moving average
Completion is carried out to interlocking equipment sampled data, the formula that the method for moving average carries out sampled data completion is as follows:
Wherein, m is integer；It is the corresponding equipment energy consumption relevant parameter of t moment；It is equipmentMoment energy consumption phase
The weighted value of parameter is closed,。
2) exceptional value is cleared up
Since the generation of most energy consumption relevant parameter exceptional values is to show as energy due to caused by acquisition equipment fault
There are maximum or minimum, (exceptional value directly reads the ammeter table truth of a matter by collector and obtains in consumption relevant parameter historical data
Out).
In the present invention, such exceptional value is handled using threshold method.When energy consumption relevant parameter is less than 0 or greater than setting
When standby rated value (rated power etc.), exceptional value is replaced with the mean value of multiple energy consumption relevant parameters near exceptional value；
In addition, also directly can assign null value to exceptional value in the present invention, the position for assigning null value is mended using the data in above-mentioned 1) step
Full method to carry out assignment to exceptional value position.
(2) framing operates
According to the data prediction result of each input variable of production system according to working system minimum production in production process
Period carries out framing operation to the energy consumption relevant parameter sampled data of interlocking equipment in system.The purpose of the framing operation is not only
It is to reduce computation complexity, it is often more important that for the ease of extracting data characteristics or specifically being calculated.The framing behaviour
Work is that equipment energy consumption relevant parameter historical data is divided into several isometric " segments ", and one " segment " is referred to as one
Frame.The length of one frame is known as frame length, and the length difference of former frame and a later frame is known as frame shifting.The framing operation is first according to energy consumption
Relevant parameter historical data granularity and minimum duty cycle determine the frame length of framing operation, then determine that frame moves again, general frame moves
It is 0-1/2 times of frame length.In order to make to seamlessly transit (keeping its continuity) between frame and frame, framing operation generallys use overlapping point
The method of section.In general, framing is realized with the method that moveable finite length window is weighted, that is, use window letter
Number w (n) Lai Chengyi frame signal s (n) (s (n) indicates the energy consumption relevant parameter sampled data for being included in this frame), adds to be formed
Window signal
Wherein, w is frame number,Energy consumption relevant parameter when for equipment w frame after windowing process；S (n) is in n-th frame
The energy consumption relevant parameter sampled data for being included；W (n) is window function.
The calculation method of the window function w (n) is as follows:
Fig. 2 show the schematic diagram of overlapping segmentation method framing operation.As seen from Figure 2: frame length frameLen；From
M frame is frameInc to m-1 frame movable length, and it is frameInc that frame, which moves,.In fact, framing operation is one by one-dimension array
Convert the process embarked on journey and have overlapping two-dimensional array with row.Its process is as follows:
Interlocking equipment framing operation energy consumption relevant parameter sampled data is usedIt indicates, wherein S is linkage
A certain equipment in equipment, P are the length of the equipment energy consumption relevant parameter sampled data.
Framing is operated after missing values and exceptional value are carried out completion and replaced using the method for above-mentioned (1) data prediction
Energy consumption relevant parameter sampled data S-transformation is following two-dimensional array:
Wherein, N is frame length, and M is frame shifting, M < N；W is frame number,；If in transformation, remaining energy consumption correlation ginseng
Number sampled data is not enough to fill up last line (i.e. P cannot be divided exactly by M), then with energy consumption relevant parameter sampled data most end
NumberFill up last line.
(3) related coefficient calculating and abnormal determination
Related coefficient of the interlocking equipment energy consumption relevant parameter sampled data on corresponding frame is calculated, is judged according to related coefficient
The linkage of equipment room, and determine the decision threshold of linkage sexual abnormality.
1) related coefficient calculates
For pretreated data after above-mentioned (2) framing operation, length is the energy consumption relevant parameter sampled data of P
It is converted intoTwo-dimensional array.For interlocking equipment, the energy consumption relevant parameter sampled data of each equipment can
It is transformed into the identical two-dimensional array of ranks number.
For two interlocking equipments, it can be operated by framing and be converted to one-dimensional energy consumption relevant parameter sampled data
Identical two-dimensional array.Further, using related coefficient calculation formula, two two-dimensional arrays are calculated separately on corresponding row
Related coefficient.The related coefficient calculation formula is as follows:
Wherein,、AndIt is that the criterion score of the second equipment energy consumption relevant parameter, energy consumption relevant parameter are average respectively
Value and energy consumption relevant parameter standard deviation；、AndIt is criterion score, the energy of the first equipment energy consumption relevant parameter respectively
Consume relevant parameter average value and energy consumption relevant parameter standard deviation.
It is calculated by related coefficient, the energy consumption relevant parameter sampled data of two associate devices is transformed to one group of phase relation
Several one-dimension arrays, the length of the array is moved by frame length and frame to be determined.Each phase relation
NumberTwo interlocking equipments are indicated in the correlation of i-th of period energy consumption relevant parameter, its linkage is quantified with this.Phase
Relationship numberValue range be [- 1,1], whenWhen indicate that energy consumption relevant parameter X and Y are positively correlated, whenWhen
Indicate that energy consumption relevant parameter X and Y is negatively correlated, whenWhen indicate energy consumption relevant parameter X and Y it is uncorrelated；WhenWhen table
Show energy consumption relevant parameter X and Y perfect positive correlation, whenWhen indicate energy consumption relevant parameter X and Y perfect negative correlation.It is related
CoefficientCloser to 1, show that energy consumption relevant parameter X and Y correlation is stronger, otherwise correlation is weaker.Energy consumption in the present invention
Relevant parameter X and Y positive correlation is stronger, then it represents that the linkage between equipment is better；Conversely, linkage is poorer.
2) decision threshold calculates
Obtaining related coefficient arrayLater, it needs to determine abnormal related coefficient, that is, determines abnormal linkage
Property.The present invention determines abnormal related coefficient using threshold method, and calculates decision threshold T using box primitive reason.
As shown in figure 3, including most data between limit in two, box figure or so.It is generally acknowledged that in the two
Data except limit are abnormal data.
In the present invention, the first of related coefficient data, third quartile are calculated separately using box primitive reason、
With quartile range IQR.Calculation formula is as follows:
In formula, first quartileEqual to by a quarter after the ascending arrangements of numerical value all in sample (if
At a quarter without specific value can be used its front and back two numerical value average value) number；Third quartileEqual to all
(if being averaged for its two numerical value of front and back can be used without specific value at 3/4ths at 3/4ths after the ascending arrangement of numerical value
Value) number.
The calculation formula of decision threshold T are as follows:
WhenWhen, it is believed that in the i-th frame position, the correlation of interlocking equipment energy consumption relevant parameter is poor, that is, joins
Dynamic property is poor.It can use threshold value thus to determine the operating status of interlocking equipment linkage difference.
3) abnormal determination
Linkage abnormal determination threshold value is calculated to restore the data after framing using linkage abnormal determination threshold value,
The part for the sexual abnormality that links during positioning energy consumption.
Firstly, positioning abnormal related coefficient according to linkage abnormal determination threshold value；
Further, frame corresponding to abnormal related coefficient is found on the basis of framing operating result, so that positioning is asked
Inscribe frame；
Finally, determining the start-stop boundary that linkage sexual abnormality occurs in problem frame；
By comparing the size of related coefficient and threshold value, it is known that at the corresponding moment with the presence or absence of linkage sexual abnormality.
When the corresponding related coefficient of certain frame be less than threshold value, and the corresponding related coefficient of frame consecutive frame be greater than threshold value, then
Think interlocking equipment in the corresponding time slice of the frame (time range of the row matrix after the corresponding framing operation of target frame)
Linkage occurs abnormal；
When there is the related coefficient of continuous several frames to be both less than threshold value, then it is assumed that interlocking equipment is in these frames corresponding period
(union of the time slice of multiple row matrixs after the corresponding framing operation of target frame) interior linkage occurs abnormal.
(4) energy saving space amount calculates
In the present invention, the mean value of the sum of interlocking equipment energy consumption relevant parameter for excluding linkage unusual part is calculated, and will
The mean value is regarded as reasonable energy consumption relevant parameter.Then reasonable energy consumption relevant parameter and interlocking equipment linkage unusual part energy are calculated
The difference between relevant parameter is consumed, which is denoted as energy saving space amount.
It is disclosed by the invention it is each preferably with optional technological means, unless otherwise indicated and one preferably or can selecting technology hand
Section is that further limiting for another technological means is outer, can form several different technical solutions in any combination.
Claims (11)
1. a kind of interlocking equipment energy efficiency analysis method for air, characterized by comprising:
Data sampling equidistantly samples the energy consumption relevant parameter of at least two equipment in interlocking equipment, the sampling of each equipment
Time point is identical, obtains the actual measurement sampled data in certain time；
Data prediction carries out data prediction, including exceptional value cleaning and Supplementing Data to actual measurement sampled data, removes wherein
Abnormal data, completion missing data forms pretreated sampled data；
Framing operation is divided into several frames to the pretreated sampled data of each equipment by the way of overlapping segmentation, and every frame is corresponding
In a duration equal period, comprising the pretreated sampled data of whole in the corresponding period, the frame of each interframe is moved
Length is less than frame length；
Related coefficient calculates, and based on the sampled data contained by each frame, calculates the energy consumption phase of at least two equipment in interlocking equipment
Parameter is closed in the Pearson correlation coefficient r of each frameXY, wherein X, Y respectively represent the energy consumption relevant parameter of two equipment；
Abnormal determination, according to the Pearson correlation coefficient decision threshold of setting, by Pearson correlation coefficient lower than decision threshold
Frame is as abnormal frame, and the corresponding period is as abnormal period.
2. interlocking equipment energy efficiency analysis method for air as described in claim 1, it is characterised in that the mode of the Supplementing Data are as follows: press
Sampled data is ranked up by the time generated according to sampled data, there is following two situation:
A) when in interlocking equipment with the presence of the sampled data shortage of data situation of an equipment, using model completion method to missing
Data carry out completion, if the energy consumption relevant parameter of related two interlocking equipments is respectively the first equipment energy consumption relevant parameter Y and the
Two equipment energy consumption relevant parameter X, the energy consumption relevant parameter of the first equipment exists in t moment to be lacked, and taking t moment, nearby n X, Y are equal
The data value not lacked, the formula that the model completion method carries out sampled data completion are as follows:
Wherein: YtIt is the estimated value of the first equipment t moment energy consumption relevant parameter missing values；XtFor the energy consumption phase of the second equipment t moment
Close parameter sampling data；eiFor error term；For related coefficient, calculated according to following equation:
In formula:The energy consumption relevant parameter mean value of respectively the first equipment and the second equipment, xiIt is the second equipment in i sampled point
Power consumption values, yi be the first equipment in the power consumption values of i sampled point, i is positive integer, and i=1 ..., n, n is number of sampling points, 0≤n
≤ (N-1), N are frame length；
B) when related two interlocking equipments are when synchronization exists simultaneously sampled data missing, using method of moving average difference distich
Dynamic equipment sampled data carries out completion, and the formula that the method for moving average carries out sampled data completion is as follows:
xc=w1xt-1+w2xt-2+…+wmxt-m
Wherein, m is integer；xtIt is the corresponding equipment energy consumption relevant parameter of t moment；wmIt is equipment xtht-mMoment energy consumption relevant parameter
Weighted value,
3. interlocking equipment energy efficiency analysis method for air as claimed in claim 2, it is characterised in that will be less than 0 or specified greater than equipment
The energy consumption relevant parameter of value is determined as exceptional value, the exceptional value cleaning method particularly includes: with more near exceptional value
For the mean value of a energy consumption relevant parameter to replace exceptional value or to exceptional value tax null value, the position for assigning null value uses the number
Come to carry out assignment to exceptional value position according to the method for completion.
4. interlocking equipment energy efficiency analysis method for air as claimed in claim 3, it is characterised in that the framing operation is with removable
The method that is weighted of finite length window come what is realized, i.e., multiply s (n) with certain window function w (n) and form adding window letter
Number, formula are as follows:
sw(n)=w (n) * s (n)
Wherein, w is frame number, sw(n) be equipment w frame when windowing process after energy consumption relevant parameter；S (n) in n-th frame by wrapping
The energy consumption relevant parameter sampled data contained；W (n) is window function；
The calculation method of the window function w (n) is as follows:
Sampled data of the equipment on a certain frame can be expressed as an one-dimension array S, formula after windowing process are as follows:
S=(s1, s2... sp)
Wherein, P is the length of the one-dimension array.
5. interlocking equipment energy efficiency analysis method for air as claimed in claim 4, it is characterised in that further include joining equipment energy consumption correlation
The step of number sampled data is converted to two-dimensional array, it is remaining in the transformation that the one-dimension array is converted to the two-dimensional array
When sampled data is not enough to fill up the two-dimensional array last line, then the energy consumption relevant parameter S of sampled data most end is usedpIt fills out
Full last line, the expression formula of two-dimensional array is as follows after conversion:
Wherein, N is frame length, and M is frame shifting, M < N；W is frame number,
6. interlocking equipment energy efficiency analysis method for air as claimed in claim 5, it is characterised in that the correlation coefficient rXYCalculating it is public
Formula are as follows:
Wherein,And σXBe respectively the criterion score of the second equipment energy consumption relevant parameter, energy consumption relevant parameter average value and
Energy consumption relevant parameter standard deviation；And σYIt is criterion score, the energy consumption correlation of the first equipment energy consumption relevant parameter respectively
Mean parameter and energy consumption relevant parameter standard deviation.
7. interlocking equipment energy efficiency analysis method for air as claimed in claim 6, it is characterised in that interlocking equipment energy consumption relevant parameter phase
Close sex determination method particularly includes: work as rXYIndicate that interlocking equipment energy consumption relevant parameter X and Y are positively correlated, and work as r when > 0XYTable when < 0
Show that interlocking equipment energy consumption relevant parameter X and Y are negatively correlated, works as rXYIndicate that interlocking equipment energy consumption relevant parameter X and Y are uncorrelated when=0；
Work as rXYInterlocking equipment energy consumption relevant parameter X and Y perfect positive correlation is indicated when=1, works as rXYEnergy consumption relevant parameter X is indicated when=- 1
With Y interlocking equipment perfect negative correlation.
8. interlocking equipment energy efficiency analysis method for air as claimed in claim 7, it is characterised in that the calculation formula of the decision threshold T
Are as follows:
T=Q1-1.5*IQR
Wherein, Q1For the first quartile of related coefficient data in box figure, Q1Equal to by all energy consumption relevant parameter numerical value by
The small number at a quarter after longer spread；IQR is quartile range；
The calculation formula of the range IQR are as follows:
IQR=Q3-Q1
Wherein, Q3For third quartile, Q3Equal to will be at after all ascending arrangements of energy consumption relevant parameter numerical value 3/4ths
Number.
9. interlocking equipment energy efficiency analysis method for air as claimed in claim 8, it is characterised in that further include what energy saving space amount calculated
Step calculates and obtains the energy saving space amount in abnormal time section relative to reasonable energy consumption.
10. interlocking equipment energy efficiency analysis method for air as claimed in claim 9, it is characterised in that the tool that the energy saving space amount calculates
Body method are as follows: based on the energy consumption relevant parameter of the lower equipment of relative value, calculate the reasonable energy consumption of abnormal period equipment
The sum of reasonable energy consumption as the abnormal period equipment, with the difference of the actual measurement energy consumption of abnormal period equipment and reasonable energy consumption, as
The energy saving space amount of the abnormal period, when there is one or more abnormal periods in certain time, with an abnormal period
Energy saving space amount or multiple abnormal period the sum of energy saving space amount be the time total energy saving space amount.
11. interlocking equipment energy efficiency analysis method for air as claimed in claim 2, it is characterised in that n is determined by the minimum production period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610776691.3A CN106372412B (en) | 2016-08-31 | 2016-08-31 | Interlocking equipment energy efficiency analysis method for air |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610776691.3A CN106372412B (en) | 2016-08-31 | 2016-08-31 | Interlocking equipment energy efficiency analysis method for air |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106372412A CN106372412A (en) | 2017-02-01 |
| CN106372412B true CN106372412B (en) | 2019-03-26 |
Family
ID=57900979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610776691.3A Active CN106372412B (en) | 2016-08-31 | 2016-08-31 | Interlocking equipment energy efficiency analysis method for air |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106372412B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830510B (en) * | 2018-07-16 | 2021-09-21 | 国网上海市电力公司 | Power data preprocessing method based on mathematical statistics |
| CN109190803B (en) * | 2018-08-14 | 2020-08-25 | 北京猿力未来科技有限公司 | Prediction method, prediction device, computing equipment and storage medium |
| CN109460398B (en) * | 2018-10-12 | 2021-03-12 | 深圳绿米联创科技有限公司 | Time series data completion method and device and electronic equipment |
| CN110263846B (en) * | 2019-06-18 | 2021-12-28 | 华北电力大学 | Fault diagnosis method based on fault data deep mining and learning |
| CN111162968A (en) * | 2019-12-26 | 2020-05-15 | 上海云海万邦数据科技有限公司 | Cloud resource use efficiency metering method and device, electronic equipment and storage medium |
| CN116205467B (en) * | 2023-04-28 | 2023-07-11 | 扬州市职业大学(扬州开放大学) | High-efficiency consumption management system and method based on industrial Internet of things |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799682A (en) * | 2012-05-10 | 2012-11-28 | 中国电力科学研究院 | Massive data preprocessing method and system |
| CN105203876A (en) * | 2015-09-15 | 2015-12-30 | 云南电网有限责任公司电力科学研究院 | Transformer on-line monitoring state assessment method utilizing support vector machine and correlation analysis |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10241887B2 (en) * | 2013-03-29 | 2019-03-26 | Vmware, Inc. | Data-agnostic anomaly detection |
-
2016
- 2016-08-31 CN CN201610776691.3A patent/CN106372412B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799682A (en) * | 2012-05-10 | 2012-11-28 | 中国电力科学研究院 | Massive data preprocessing method and system |
| CN105203876A (en) * | 2015-09-15 | 2015-12-30 | 云南电网有限责任公司电力科学研究院 | Transformer on-line monitoring state assessment method utilizing support vector machine and correlation analysis |
Non-Patent Citations (1)
| Title |
|---|
| 基于能效分析的水泥厂变频优化技术及应用;邓全亮 等;《2016第五届水泥工业节能环保技术高峰论坛文集》;20160712;第23-26页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106372412A (en) | 2017-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106372412B (en) | Interlocking equipment energy efficiency analysis method for air | |
| US8909999B2 (en) | Dynamic voltage scaling system based on on-chip monitoring and voltage prediction | |
| US20160239592A1 (en) | Data-driven battery aging model using statistical analysis and artificial intelligence | |
| US20130082724A1 (en) | Pv panel diagnosis device, diagnosis method and diagnosis program | |
| Mitrovic et al. | Low-dimensional models for compressed sensing and prediction of large-scale traffic data | |
| Qin et al. | Adaptive and robust prediction for the remaining useful life of electrolytic capacitors | |
| CN108616377B (en) | Service chain virtual machine control method and system | |
| Conti et al. | Blind maximum likelihood estimation of traffic matrices under long-range dependent traffic | |
| KR102624950B1 (en) | System for detecting abnormal value with periodicity using time-series data | |
| Chandra et al. | Model-free angle stability assessment using wide area measurements | |
| CN108108837B (en) | Regional new energy power supply structure optimization prediction method and system | |
| CN101860025A (en) | Predictor-corrector technology-based power loss calculation method of grid in future operation mode | |
| KR20200039877A (en) | Apparatus for detecting change of electric load and method thereof | |
| EP4142088A1 (en) | Predicting grid frequency | |
| Maillet et al. | Dynamic state estimation in distributed aircraft electric control systems via adaptive submodularity | |
| CN113533950B (en) | BiLSTM-based high-voltage circuit breaker operation voltage sampling method | |
| Mahmood et al. | A method for extracting steady state components from Syncrophasor data using Kalman Filters | |
| JP2010204784A (en) | Disturbance control apparatus, disturbance control method, disturbance control program and recording medium | |
| CN110866652B (en) | Online PMU data error correction method and system based on LSTM model | |
| CN101877543A (en) | Switching power supply control method and device | |
| Andrade | On-line estimation of electrolytic capacitor parameters using recursive least squares and sliding mode differentiators | |
| Buiatti et al. | An unified method for estimating the parameters of non-isolated DC/DC converters using continuous time models | |
| CN109274970A (en) | A kind of rapid scene switching detection method and system | |
| CN116223044B (en) | Bearing residual life prediction system and method based on multi-sensor fusion and Bi-TACN | |
| CN110968502B (en) | System and method for performing traversal test based on MFAC algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room 406A, 4th floor, 106 Lize Zhongyuan Building, Chaoyang District, Beijing Patentee after: Beijing visualization intelligent Polytron Technologies Inc Address before: Room 406A, 4th floor, 106 Lize Zhongyuan Building, Chaoyang District, Beijing Patentee before: BEIJING AVATER ENERGY SAVING TECHNOLOGY CO., LTD. |