CN107622452A - Method and apparatus for estimating uncertainty of model related to wind turbine generator set - Google Patents
Method and apparatus for estimating uncertainty of model related to wind turbine generator set Download PDFInfo
- Publication number
- CN107622452A CN107622452A CN201710841534.0A CN201710841534A CN107622452A CN 107622452 A CN107622452 A CN 107622452A CN 201710841534 A CN201710841534 A CN 201710841534A CN 107622452 A CN107622452 A CN 107622452A
- Authority
- CN
- China
- Prior art keywords
- model
- parameter
- output
- parameters
- estimation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000012937 correction Methods 0.000 claims description 9
- 241000208340 Araliaceae Species 0.000 description 12
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 12
- 235000003140 Panax quinquefolius Nutrition 0.000 description 12
- 235000008434 ginseng Nutrition 0.000 description 12
- 230000006870 function Effects 0.000 description 8
- 238000013507 mapping Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Wind Motors (AREA)
Abstract
A method and a device for estimating the uncertainty of a model relating to a wind park are provided, the method comprising the following estimation steps performed each time the model is used: obtaining the output of the model when the model is used at the current time; estimating parameters of a distribution of errors of the output based on the output; a first uncertainty of an output of the model is derived based on the distributed parameters. According to the invention, the uncertainty of the model related to the wind generating set can be accurately evaluated.
Description
Technical field
The present invention relates to wind power generation field.More particularly, it is related to a kind of estimation mould relevant with wind power generating set
Probabilistic method and apparatus of type.
Background technology
Wind energy is increasingly taken seriously, installation amount is also continuously increased as a kind of regenerative resource of cleaning.With wind-force
The continuous development of generation technology, the various researchs of wind power generating set are also increasingly deep, various uses and wind power generating set
Relevant model is suggested.
Because the application of these models is increasingly extensive, it is related to the various aspects of wind power generating set, even relates to
The safe operation of wind power generating set.Therefore, how the uncertainty (degree) of the output of accurate evaluation these models be one urgently
The problem of to be solved.
The content of the invention
It is an object of the invention to provide a kind of probabilistic method for estimating the model relevant with wind power generating set
And equipment.
According to an aspect of the present invention, there is provided a kind of probabilistic side for estimating the model relevant with wind power generating set
Method, methods described, which is included in when the model is used every time, performs following estimating step：The model is obtained when previous
Output when being used；The parameter of distribution based on the error exported described in the output estimation；Parameter based on the distribution
Obtain the output of the model first is uncertain.
Alternatively, first it is uncertain for first predetermined value and the parameters of the distribution and.
Alternatively, the first predetermined value represents predetermined uncertainty.
Alternatively, the step of parameter for estimating the distribution of the error, includes：Based on the output and the model upper
The parameter of the distribution of the error of output when once being used, estimate the error of output of model when when previous used
Distribution parameter.
Alternatively, the ginseng of the distribution of the error of output of model when when previous used is estimated by recursive algorithm
Number.
Alternatively, described to be distributed as normal distribution, the parameter of the distribution includes average and standard deviation.
Alternatively, the estimating step also includes：Determine the type of the distribution of the error of the output of the model as institute
State predefined type.
Alternatively, the estimating step also includes：Obtain in the input that the model receives when when previous used
At least one parameter；It is determined that the pre-set interval that at least one parameter obtained is each fallen into；Respectively for described at least
A kind of every kind of parameter in parameter, count to each time of the model so far by using when every kind of parameter respectively fall in determination
The pre-set interval each fallen into total degree；Determine that the second of the output of the model does not know based on the total degree
Property.
Alternatively, the second uncertain step of the output of the model is determined based on the total degree to be included：It is determined that
The output of the corresponding instruction model of the pre-set interval of type, the total degree and determination with least one parameter
Probabilistic value.
Alternatively, at least one parameter is many kinds of parameters, and the estimating step also includes：According to many kinds of parameters
In the pre-set interval that is fallen into of predefined parameter to determine the parameter of predetermined quantity from many kinds of parameters, based on described total time
Number determines that the second uncertain step of the output of the model includes：It is total corresponding to parameter based on the predetermined quantity
Number determines the second uncertainty of the output of the model.
Alternatively, the second uncertain step of the output of the model is determined based on the total degree to be included：Work as institute
When to state pre-set interval that the predefined parameter in many kinds of parameters is fallen into be the first predetermined pre-set interval, it is based only upon and the predetermined ginseng
Total degree corresponding to number determines the second uncertainty of the output of the model；When the predefined parameter in many kinds of parameters is fallen
When the pre-set interval entered is the second predetermined pre-set interval different from the first predetermined pre-set interval, based on many kinds of parameters
Total degree corresponding to all parameters determines the second uncertainty of the output of the model.
Alternatively, be based only upon total degree corresponding with the predefined parameter determine the model output it is second uncertain
The step of property, includes：It is determined that type, total degree corresponding with the predefined parameter with the predefined parameter and described predetermined
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that parameter is fallen into；Based on many kinds of parameters
All parameters corresponding to total degree determine that the second uncertain step of the output of the model includes：It is determined that with it is described more
The pre-set interval pair that type, total degree corresponding with various parameters and the various parameters of various parameters in kind parameter are fallen into
Probabilistic value of the output for the instruction model answered.
Alternatively, the of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Two uncertain steps include：When total degree corresponding with any one parameter in many kinds of parameters is appointed no more than described
During frequency threshold value corresponding to a kind of parameter of anticipating, it is determined that type and various parameters pair with the various parameters in many kinds of parameters
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that the total degree and various parameters answered are fallen into.
Alternatively, the of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Two uncertain steps also include：When total degree corresponding with every kind of parameter is both greater than or equal to number corresponding to every kind of parameter
During threshold value, the second of the output of the model probabilistic value is defined as second predetermined value.
Alternatively, the second predetermined value represents that the output of the model is completely believable.
Alternatively, at least one parameter includes wind speed and/or turbulence intensity.
Alternatively, the predefined parameter is wind speed, and the lower limit of the first predetermined pre-set interval is more than the second predetermined pre-set interval
The upper limit.
Alternatively, the estimating step also includes：The second uncertainty is calculated with first probabilistic product as institute
State the output of model the 3rd is uncertain.
Another aspect of the present invention provides a kind of method for the output for correcting the model relevant with wind power generating set, described
Method includes：The first uncertain, Huo Zhe of the output of the model relevant with wind power generating set is estimated by the above method
Two is uncertain, or the 3rd uncertain；Use the Uncertainty correction of the estimation model relevant with wind power generating set
Output.
Alternatively, wrapped using the step of output of the Uncertainty correction of the estimation model relevant with wind power generating set
Include：Calculate the uncertain and product of the output of estimation.
Another aspect of the present invention provides a kind of probabilistic equipment for estimating the model relevant with wind power generating set,
The equipment includes：Acquiring unit is exported, obtains output of model when when previous used；Estimation of distribution parameters list
Member, the parameter of the distribution based on the error exported described in the output estimation；First estimation unit, the parameter based on the distribution
Obtain the output of the model first is uncertain.
Alternatively, first it is uncertain for first predetermined value and the parameters of the distribution and.
Alternatively, the first predetermined value represents predetermined uncertainty.
Alternatively, estimation of distribution parameters unit based on the output and the model it is upper once used when output
The parameter of the distribution of error, estimate the parameter of the distribution of the error of output of model when when previous used.
Alternatively, estimation of distribution parameters unit estimates output of model when when previous used by recursive algorithm
Error distribution parameter.
Alternatively, described to be distributed as normal distribution, the parameter of the distribution includes average and standard deviation.
Alternatively, the equipment also includes：Estimation unit is distributed, determines the class of the distribution of the error of the output of the model
Type is as the predefined type.
According to another aspect of the present invention, there is provided a kind of output for correcting the model relevant with wind power generating set is set
Standby, the equipment includes：Device described above；Estimation unit, estimated using the equipment relevant with wind power generating set
Model output it is first uncertain, second uncertain or the 3rd is uncertain as the uncertainty estimated, school
The output of model just relevant with wind power generating set.
Alternatively, estimation unit calculates the uncertain and product of the output of estimation.
Alternatively, the equipment also includes：Input parameter acquiring unit, in the model every time by use, obtaining institute
State at least one of the input that model receives when when previous used parameter；Interval judgement unit, it is determined that what is obtained is described
The pre-set interval that at least one parameter is each fallen into；Counting unit, respectively for every seed ginseng at least one parameter
Number, count to each time of the model so far by using when every kind of parameter respectively fall in determination each fallen into preset
The total degree in section；Second estimation unit, determine that the second of the output of the model is uncertain based on the total degree.
Alternatively, the second estimation unit determines and the type of at least one parameter, the total degree and determination
Probabilistic value of the output of the instruction model corresponding to pre-set interval.
Alternatively, at least one parameter is many kinds of parameters, and the equipment also includes：Selecting unit, according to described more
The pre-set interval that predefined parameter in kind parameter is fallen into determine the parameter of predetermined quantity from many kinds of parameters, and second estimates
Total degree corresponding to meter parameter of the unit based on the predetermined quantity determines the second uncertainty of the output of the model.
Alternatively, when the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is the first predetermined pre-set interval
When, the second estimation unit be based only upon total degree corresponding with the predefined parameter determine the model output it is second uncertain
Property；When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is pre- for second different from the first predetermined pre-set interval
When determining pre-set interval, the second estimation unit determines the model based on total degree corresponding with all parameters of many kinds of parameters
Output it is second uncertain.
Alternatively, be based only upon total degree corresponding with the predefined parameter determine the model output it is second uncertain
The processing of property includes：It is determined that type, total degree corresponding with the predefined parameter with the predefined parameter and described predetermined
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that parameter is fallen into；Based on many kinds of parameters
All parameters corresponding to total degree determine the model output second it is probabilistic processing include：It is determined that with it is described more
The pre-set interval pair that type, total degree corresponding with various parameters and the various parameters of various parameters in kind parameter are fallen into
Probabilistic value of the output for the instruction model answered.
Alternatively, the of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Two probabilistic processing include：When total degree corresponding with any one parameter in many kinds of parameters is appointed no more than described
During frequency threshold value corresponding to a kind of parameter of anticipating, it is determined that type and various parameters pair with the various parameters in many kinds of parameters
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that the total degree and various parameters answered are fallen into.
Alternatively, the of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Two probabilistic processing also include：When total degree corresponding with every kind of parameter is both greater than or equal to number corresponding to every kind of parameter
During threshold value, the second of the output of the model probabilistic value is defined as second predetermined value.
Alternatively, the second predetermined value represents that the output of the model is completely believable.
Alternatively, at least one parameter includes wind speed and/or turbulence intensity.
Alternatively, the predefined parameter is wind speed, and the lower limit of the first predetermined pre-set interval is more than the second predetermined pre-set interval
The upper limit.
Alternatively, the equipment also includes：3rd estimation unit, calculate second and uncertain probabilistic multiply with first
Threeth uncertainty of the product as the output of the model.
Another aspect of the present invention provides a kind of system for the output for correcting the model relevant with wind power generating set, and it is special
Sign is that the system includes：Processor；Memory, computer-readable code is stored with, when the computer-readable code quilt
During computing device, methods described is performed.
Another aspect of the present invention provides a kind of computer-readable recording medium for being wherein stored with computer-readable code,
Methods described is performed when the computer-readable code is performed.
According to probabilistic method and apparatus of the estimation of the present invention model relevant with wind power generating set, Ke Yizhun
Really evaluate the uncertainty of the model relevant with wind power generating set.In addition, by using the uncertainty according to the present invention
Evaluation method and equipment, the output accuracy or accuracy of model with the output of calibration model, can be improved.
Brief description of the drawings
By the detailed description carried out below in conjunction with the accompanying drawings, above and other objects of the present invention, feature and advantage will
Become more fully apparent, wherein：
Fig. 1 shows the uncertainty according to the estimation of the first embodiment of the present invention model relevant with wind power generating set
Method flow chart.
Fig. 2 shows the uncertainty of the estimation according to the second embodiment of the present invention model relevant with wind power generating set
Method flow chart.
Fig. 3 shows the uncertainty of the estimation according to the fourth embodiment of the invention model relevant with wind power generating set
Equipment block diagram.
Fig. 4 shows the uncertainty of the estimation according to the fifth embodiment of the invention model relevant with wind power generating set
Equipment block diagram.
Embodiment
Now, different example embodiments is more fully described with reference to the accompanying drawings.
The present invention provides the probabilistic method and apparatus for estimating the model relevant with wind power generating set.Here, with
The relevant model of wind power generating set can be the mould of the parts for wind power generating set entirety or wind power generating set
Type.These models can be used for various uses, such as, estimation or prediction load, estimation or predicted fatigue life, estimation or prediction
Failure etc., estimation or prediction operational factor etc., it should be appreciated that the model not limited to this relevant with wind power generating set.
In one embodiment, the uncertainty of the model relevant with wind power generating set can be understood as the defeated of the model
The uncertainty gone out.
In the present invention, probabilistic estimation is performed when the model is used every time, it is every to assess the model
The uncertainty of secondary output when being used.It should be understood that refer to input corresponding input to the model using the model
To be exported accordingly.
With reference to Fig. 1 show according to another embodiment of the invention when the model is used every time assess not
Deterministic method.
Fig. 1 shows the uncertainty according to the estimation of the first embodiment of the present invention model relevant with wind power generating set
Method flow chart.The model performs the method shown in Fig. 1 when being used every time.
In step S110, output of model when when previous used is obtained.In other words, the model is when previous
By using when based on input and exported.
In step S120, the parameter of the distribution based on the error exported described in the output estimation.
Specifically, can based on it is described output and the model it is upper once used when output error distribution (example
Such as, the distribution of predefined type) parameter, estimate the ginseng of the distribution of the error of output of model when when previous used
Number.For example, can by recursive algorithm, based on the output and the model it is upper once used when output error point
The parameter of cloth estimates the parameter of the distribution of the error of output of model when when previous used.Using existing suitable
Estimated for the various recursive algorithms of the parameter of the distribution of the error of the output of the model.
Here, the distribution of error refers to the distribution that the error of the output of the model is obeyed.For example, the distribution of error takes
It can be certainly normal distribution, Poisson distribution or Weibull distribution etc. in the characteristics of output of the model, but be not limited to
This.The distribution that the error of the output of the model is obeyed can be predefined.Now, it is determined that distribution type as described pre-
Determine type.
For example, error be distributed as normal distribution in the case of, the parameter of the distribution includes average and standard deviation.Under
The method that face describes the average based on the error exported described in the output estimation and standard deviation proposed by the invention.
Average is calculated by following equation (1)：
Wherein, μ∈, kRepresent the ginseng of distribution of the model in the error as the output when previous kth time is used
Average in number, μ∈, k-1Represent the distribution of the error of output of the model when being used for -1 time as upper kth once
Average in parameter,Output for kth time using model during the model.
Standard deviation is calculated by following equation (2)：
Wherein, σ∈, kRepresent the ginseng of distribution of the model in the error as the output when previous kth time is used
Standard deviation in number, σ∈, k-1Represent the distribution of the error of output of the model when being used for -1 time as upper kth once
Parameter in standard deviation,Output for kth time using model during the model.
It should be understood that the estimation average of the present invention and the method not limited to this of standard deviation, other method is also feasible.Separately
Outside, the method for estimation average of the invention and standard deviation is not limited to be applied to the situation for being distributed as normal distribution, other
It is also feasible using average and standard deviation as the distribution of parameter.
In step S130, the parameter based on the distribution obtains the uncertainty (hereinafter referred to as of the output of the model
One is uncertain).
In one embodiment, first it is uncertain for first predetermined value and the parameters of the distribution and.First
Predetermined value can represent predetermined uncertainty.For example, can be according to the difference of model, use environment, and/or occupation mode etc. come really
Determine first predetermined value.For example, in one example, the first predetermined value can be 1.
For example, it is described be distributed as normal distribution in the case of, the first uncertainty may be expressed as first predetermined value, estimate
The average of meter, the sum of the standard deviation of estimation this three.
Show that the assessment according to an embodiment of the invention when the model is used every time is not true with reference to Fig. 2
Qualitatively method.
Fig. 2 shows the uncertainty of the estimation according to the second embodiment of the present invention model relevant with wind power generating set
Method flow chart.The model performs the method shown in Fig. 2 when being used every time.
In step S210, at least one of the input that the model receives when when previous used parameter is obtained (i.e.,
One or more parameters).At least one parameter can be all parameters or partial parameters in the input received.It is described extremely
A kind of few parameter is determined in advance, and same type of ginseng is all obtained during probabilistic estimation of the model so as to perform every time
Number.
In step S220, it is determined that the pre-set interval that at least one parameter obtained is each fallen into.
In the present invention, for every kind of parameter at least one parameter, division is on the more of every kind of parameter in advance
Individual pre-set interval.When getting at least one parameter, it is multiple pre- corresponding to this kind of parameter to determine that every kind of parameter is fallen into
If which of section pre-set interval.It should be understood that the section that parameter is fallen into refers to the section that the value of parameter is fallen into.
For example, in the case where at least one parameter includes the first parameter and the second parameter, drawn for the first parameter
Divide the pre-set interval of the first quantity, the pre-set interval of the second quantity is divided for the second parameter.Getting the first parameter and
After two parameters, it is determined that the first parameter obtained falls into which of the pre-set interval of the first quantity section, it is determined that obtain second
Parameter falls into which of the pre-set interval of the second quantity section.It should be understood that the invention is not restricted to two kinds of parameters, retouching above
It is only exemplary for stating.
In step S230, respectively for every kind of parameter at least one parameter, count to so far described
Each time of model by using when every kind of parameter respectively fall in determination the pre-set interval each fallen into total degree.
In other words, for each pre-set interval determined in step S220, count including the not true of this all times
The total degree of the corresponding parameter of each pre-set interval of determination is fallen into qualitative estimation.That is, it is directed in step
The section that S220 is determined determines to have in history the parameter of corresponding types fall into total degree therein.
For example, the section fallen into for the first parameter determined in step S220, it is determined that including this all times
Uncertainty estimation in, the first parameter drops into the total degree in the section.If for example, the section only this and upper one
There is the first parameter to fall into secondary uncertainty estimation, then total degree is 2.
In step S240, the uncertainty of the output of the model is determined based on the total degree, and (hereinafter referred to as second is not
Certainty).
Specifically, it is determined that the pre-set interval pair of the type, the total degree and determination with least one parameter
Probabilistic value of the output for the instruction model answered.For example, can pre-establish on it is uncertain with the type of parameter,
The function of pre-set interval, total degree, or pre-establish the mapping of uncertain type, pre-set interval, total degree with parameter
Table.So, can the type based at least one parameter, determine pre-set interval (e.g., can by pre-set interval identifier,
Label etc. represent), statistics total degree, uncertainty is determined by function or mapping table.
Although obtaining at least one parameter in step S210, in the situation that at least one parameter is many kinds of parameters
Under, for some models, not all situation will will now consider predetermined ginseng therein all using many kinds of parameters
The section fallen into that number determines in step S220 is not true to determine to be used for using which of many kinds of parameters parameter
Qualitatively estimate.That is, the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is come from many kinds of parameters
The parameter of predetermined quantity is determined, total degree (that is, the parameter of described predetermined quantity corresponding to the parameter based on the predetermined quantity
The total degree in step S230 corresponding to the section fallen into determined in step S220) determine the output of the model
It is second uncertain.In an example embodiment, the second not true of the output of the model is determined based on the total degree
Qualitative step includes：When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is the first predetermined pre-set interval
When, it is based only upon the second uncertainty that total degree corresponding with the predefined parameter determines the output of the model；When described more
The pre-set interval that predefined parameter in kind parameter is fallen into is the second predetermined pre-set interval different from the first predetermined pre-set interval
When, determine that the second of the output of the model is uncertain based on total degree corresponding with all parameters of many kinds of parameters.
It should be understood that the first predetermined pre-set interval or the quantity of the second predetermined pre-set interval here can be one or more.
It is based only upon second probabilistic step that total degree corresponding with the predefined parameter determines the output of the model
Suddenly include：It is determined that type, total degree corresponding with the predefined parameter and predefined parameter institute with the predefined parameter
Probabilistic value of the output of the instruction model corresponding to the pre-set interval fallen into.For example, can by the type of predefined parameter,
Corresponding to pre-set interval that predefined parameter is fallen into (for example, can be represented as the identifier of pre-set interval, label etc.), predefined parameter
Total degree as input, based on pre-establish on it is uncertain with the type of parameter, the function of pre-set interval, total degree,
Or the uncertain and mapping table of the type of parameter, pre-set interval, total degree pre-established indicates the model to obtain
Output probabilistic value.
The second not true of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Qualitative step includes：It is determined that type, total degree corresponding with various parameters with the various parameters in many kinds of parameters with
And probabilistic value of the output of the instruction model corresponding to the pre-set interval that is fallen into of various parameters.For example, can by institute
Have pre-set interval that the type of parameter, all parameters fallen into (for example, can be represented by the identifier of pre-set interval, label etc.),
Total degree corresponding to all parameters as input, based on pre-establish on it is uncertain with the type of parameter, pre-set interval,
The function of total degree, or the mapping table of the uncertainty pre-established and the type of parameter, pre-set interval, total degree obtain
Indicate probabilistic value of the output of the model.
In addition, the second of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Uncertain step also includes：When total degree corresponding with any one parameter in many kinds of parameters is appointed no more than described
During frequency threshold value corresponding to a kind of parameter of anticipating, it is determined that type and various parameters pair with the various parameters in many kinds of parameters
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that the total degree and various parameters answered are fallen into.
When total degree corresponding with every kind of parameter is both greater than or during equal to frequency threshold value corresponding to every kind of parameter, by the output of the model
Second probabilistic value be defined as second predetermined value.Second predetermined value can represent that the output of the model is completely believable.
For example, according to probabilistic occupation mode, the second predetermined value can be 1.
In other words, frequency threshold value is set for all or part of pre-set intervals of every kind of parameter, according to uncertain at this
Property estimation in total degree corresponding to the pre-set interval that determines it is whether uncertain to further determine that beyond corresponding frequency threshold value
Property.
In one embodiment, at least one parameter includes wind speed and/or turbulence intensity.In at least one ginseng
In the case that number includes wind speed and turbulence intensity, predefined parameter above-mentioned is wind speed, the lower limit of the first predetermined pre-set interval
More than the upper limit of the second predetermined pre-set interval.It should be understood that this is only exemplary, depending on the difference of model, it is described at least
A kind of parameter is also different.
According to one embodiment of present invention, the present invention also provides estimation and wind-force according to the third embodiment of the invention
Probabilistic method of the relevant model of generating set.Side of the methods described shown in including first embodiment and second embodiment
Method, and methods described also includes the model and uncertain probabilistic multiplied with first in use, calculating second every time
Threeth uncertainty of the product as the output of the model.
According to one embodiment of present invention, the present invention also provides a kind of correction model relevant with wind power generating set
The method of output.In the method, the mould relevant with wind power generating set is estimated by the method shown in first embodiment first
The first of the output of type is uncertain, or the mould relevant with wind power generating set is estimated by the method shown in second embodiment
The second of the output of type is uncertain, or the method shown in logical 3rd embodiment estimates the model relevant with wind power generating set
Output it is the 3rd uncertain；Then using the defeated of the Uncertainty correction of the estimation model relevant with wind power generating set
Go out.Specifically, the uncertain and product of the output of estimation can be calculated to correct original output, use the product
Result update original output.
The estimation according to an embodiment of the invention model relevant with wind power generating set is shown with reference to Fig. 3
Probabilistic equipment.
Fig. 3 shows the uncertainty of the estimation according to the fourth embodiment of the invention model relevant with wind power generating set
Equipment block diagram.
As shown in figure 3, the estimation according to the fourth embodiment of the invention model relevant with wind power generating set is not true
Qualitatively equipment 300 includes output acquiring unit 310, estimation of distribution parameters unit 320, the first estimation unit 330.
The equipment 300 the model every time by using when operated to estimate the uncertainty.In other words, exist
The model is every time by use, output acquiring unit 310, estimation of distribution parameters unit 320, the first estimation unit 330 are carried out
Operation, to estimate the uncertainty.
Output acquiring unit 310 obtains output of model when when previous used.In other words, the model is being worked as
It is previous by using when based on input and exported.
The parameter of distribution of the estimation of distribution parameters unit 320 based on the error exported described in the output estimation.
Specifically, estimation of distribution parameters unit 320 can based on it is described output and the model it is upper once used when
The parameter of the distribution of the error of output, estimate the parameter of the distribution of the error of output of model when when previous used.
For example, can by recursive algorithm, based on the output and the model it is upper once used when output error distribution
Parameter estimate the parameter of the distribution of the error of output of model when when previous used.It is applicable using existing
Estimated in the various recursive algorithms of the parameter of the distribution.
Here, distribution refers to the distribution that the error of the output of the model is obeyed.For example, distribution depends on the model
Output the characteristics of can be normal distribution, Poisson distribution or Weibull distribution etc., but not limited to this.It can predefine
The distribution that the error of the output of the model is obeyed.Now, it is determined that distribution type as the predefined type.Now,
Shown equipment 300 also includes distribution estimation unit, and distribution estimation unit determines the class of the distribution of the error of the output of the model
Type is as the predefined type.
For example, it is described be distributed as normal distribution in the case of, the parameter of the distribution includes average and standard deviation.Example
Such as, the average and standard deviation of the error of the output can be estimated by above equation (1) and equation (2).
First parameter of the estimation unit 330 based on the distribution obtains the first uncertainty of the output of the model.
In one embodiment, first it is uncertain for first predetermined value and the parameters of the distribution and.First
Predetermined value can represent predetermined uncertainty.For example, can be according to the difference of model, use environment, and/or occupation mode etc. come really
Determine first predetermined value.For example, in one example, the first predetermined value can be 1.
For example, it is described be distributed as normal distribution in the case of, the first uncertainty may be expressed as first predetermined value, estimate
The average of meter, the sum of the standard deviation of estimation this three.
The estimation according to an embodiment of the invention model relevant with wind power generating set is shown with reference to Fig. 4
Probabilistic equipment.
Fig. 4 shows the uncertainty of the estimation according to the fifth embodiment of the invention model relevant with wind power generating set
Equipment block diagram.
As shown in figure 4, the estimation according to the fifth embodiment of the invention model relevant with wind power generating set is not true
Qualitatively equipment 400 includes input parameter acquiring unit 410, interval judgement unit 420, counting unit 430, the second estimation unit
440。
The equipment 400 the model every time by using when operated to estimate the uncertainty.In other words, exist
The model is every time by use, input parameter acquiring unit 410, interval judgement unit 420, counting unit 430, second are estimated
Unit 440 is operated, to estimate the uncertainty.
Input parameter acquiring unit 410 is used in use, obtaining the model when previous every time in the model
When at least one of the input that receives parameter (that is, one or more parameters).At least one parameter can be received
All parameters or partial parameters in input.At least one parameter is determined in advance, so as to perform the model every time
Same type of parameter is all obtained during probabilistic estimation.
Interval judgement unit 420 determines that at least one parameter that input parameter acquiring unit 410 obtains each is fallen
The pre-set interval entered.
In the present invention, for every kind of parameter at least one parameter, division is on the more of every kind of parameter in advance
Individual pre-set interval.When getting at least one parameter, it is multiple pre- corresponding to this kind of parameter to determine that every kind of parameter is fallen into
If which of section pre-set interval.
For example, in the case where at least one parameter includes the first parameter and the second parameter, drawn for the first parameter
Divide the pre-set interval of the first quantity, the pre-set interval of the second quantity is divided for the second parameter.Getting the first parameter and
After two parameters, it is determined that the first parameter obtained falls into which of the pre-set interval of the first quantity section, it is determined that obtain second
Parameter falls into which of the pre-set interval of the second quantity section.It should be understood that the invention is not restricted to two kinds of parameters, retouching above
It is only exemplary for stating.
Counting unit 430 is directed to every kind of parameter at least one parameter respectively, counts to so far described
Each time of model by using when every kind of parameter respectively fall in determination the pre-set interval each fallen into total degree.
In other words, each pre-set interval determined for interval judgement unit 420, count including this all times
The total degree of the corresponding parameter of each pre-set interval of determination is fallen into uncertainty estimation.That is, only it is to be directed to section
The section that judging unit 420 determines determines to have in history the parameter of corresponding types fall into total degree therein.
For example, the section fallen into for the first parameter determined in interval judgement unit 420, it is determined that including this
In the uncertainty estimation of all times, the first parameter drops into the total degree in the section.If for example, the section only this with
And thering is the first parameter to fall into last uncertainty estimation, then total degree is 2.
Second estimation unit 440 determines the second uncertainty of the output of the model based on the total degree.
Specifically, it is determined that the pre-set interval pair of the type, the total degree and determination with least one parameter
Probabilistic value of the output for the instruction model answered.For example, can pre-establish on it is uncertain with the type of parameter,
The function of pre-set interval, total degree, or pre-establish the mapping of uncertain type, pre-set interval, total degree with parameter
Table.So, can the type based at least one parameter, determine pre-set interval (e.g., can by pre-set interval identifier,
Label etc. represent), statistics total degree, uncertainty is determined by function or mapping table.
It is more seed ginsengs at least one parameter although input parameter acquiring unit 410 obtains at least one parameter
In the case of number, for some models, not all situation will will now consider wherein all using many kinds of parameters
The section fallen into that is determined by interval judgement unit 420 of predefined parameter determine using which of described many kinds of parameters ginseng
Count to be used for probabilistic estimation.That is, the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is come from institute
The parameter that predetermined quantity is determined in many kinds of parameters is stated, the total degree corresponding to the parameter based on the predetermined quantity is (that is, described pre-
Being counted by counting unit 430 corresponding to the section fallen into determined as interval judgement unit 420 of the parameter of fixed number amount
Total degree) determine the model output it is second uncertain.In the case, the equipment 400 also includes selecting unit,
The pre-set interval that predefined parameter of the selecting unit in many kinds of parameters is fallen into is pre- to be determined from many kinds of parameters
The parameter of fixed number amount.Total degree corresponding to second parameter of the estimation unit 440 based on the predetermined quantity determines the model
Output it is second uncertain.
In an example embodiment, second probabilistic place of the output of the model is determined based on the total degree
Reason includes：When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is the first predetermined pre-set interval, second estimates
Meter unit is based only upon the second uncertainty that total degree corresponding with the predefined parameter determines the output of the model；When described
The pre-set interval that predefined parameter in many kinds of parameters is fallen into is the second predetermined preset areas different from the first predetermined pre-set interval
Between when, the second estimation unit determines the output of the model based on total degree corresponding with all parameters of many kinds of parameters
Second is uncertain.It should be understood that the first predetermined pre-set interval or the quantity of the second predetermined pre-set interval here can be one
It is individual or multiple.
It is based only upon second probabilistic place that total degree corresponding with the predefined parameter determines the output of the model
Reason includes：It is determined that type, total degree corresponding with the predefined parameter and predefined parameter institute with the predefined parameter
Probabilistic value of the output of the instruction model corresponding to the pre-set interval fallen into.For example, can by the type of predefined parameter,
Corresponding to pre-set interval that predefined parameter is fallen into (for example, can be represented as the identifier of pre-set interval, label etc.), predefined parameter
Total degree as input, based on pre-establish on it is uncertain with the type of parameter, the function of pre-set interval, total degree,
Or the uncertain and mapping table of the type of parameter, pre-set interval, total degree pre-established indicates the model to obtain
Output probabilistic value.
The second not true of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Qualitatively processing includes：It is determined that type, total degree corresponding with various parameters with the various parameters in many kinds of parameters with
And probabilistic value of the output of the instruction model corresponding to the pre-set interval that is fallen into of various parameters.For example, can by institute
Have pre-set interval that the type of parameter, all parameters fallen into (for example, can be represented by the identifier of pre-set interval, label etc.),
Total degree corresponding to all parameters as input, based on pre-establish on it is uncertain with the type of parameter, pre-set interval,
The function of total degree, or the mapping table of the uncertainty pre-established and the type of parameter, pre-set interval, total degree obtain
Indicate probabilistic value of the output of the model.
In addition, the second of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters
Probabilistic processing also includes：When total degree corresponding with any one parameter in many kinds of parameters is appointed no more than described
During frequency threshold value corresponding to a kind of parameter of anticipating, it is determined that type and various parameters pair with the various parameters in many kinds of parameters
Probabilistic value of the output of the instruction model corresponding to the pre-set interval that the total degree and various parameters answered are fallen into.
When total degree corresponding with every kind of parameter is both greater than or during equal to frequency threshold value corresponding to every kind of parameter, by the output of the model
Second probabilistic value be defined as second predetermined value.Second predetermined value can represent that the output of the model is completely believable.
For example, according to probabilistic occupation mode, the second predetermined value can be 1.
In other words, frequency threshold value is set for all or part of pre-set intervals of every kind of parameter, according to uncertain at this
Property estimation in total degree corresponding to the pre-set interval that determines it is whether uncertain to further determine that beyond corresponding frequency threshold value
Property.
In one embodiment, at least one parameter includes wind speed and/or turbulence intensity.In at least one ginseng
In the case that number includes wind speed and turbulence intensity, predefined parameter above-mentioned is wind speed, the lower limit of the first predetermined pre-set interval
More than the upper limit of the second predetermined pre-set interval.It should be understood that this is only exemplary, depending on the difference of model, it is described at least
A kind of parameter is also different.
According to one embodiment of present invention, the present invention also provides estimation and wind-force according to the sixth embodiment of the invention
Probabilistic equipment of the relevant model of generating set.The equipment includes setting shown in fourth embodiment and the 5th embodiment
Standby 300 and 400, and the equipment also includes the 3rd estimation unit, and the model is every time by use, the 3rd estimation unit meter
Calculate threeth uncertainty of the second uncertain and first probabilistic product as the output of the model.
According to one embodiment of present invention, the present invention also provides a kind of correction model relevant with wind power generating set
The equipment of output.The equipment includes the estimation and wind-power electricity generation shown in fourth embodiment, the 5th embodiment or sixth embodiment
Probabilistic equipment of the relevant model of unit.In addition, the equipment also includes estimation unit, estimation unit is not true using first
Qualitative or the second uncertain or the 3rd uncertain uncertainty as estimation, correction have with wind power generating set
The output of the model of pass.Specifically, estimation unit can calculate the uncertain and product of the output of estimation to correct original
The output of beginning, original output is updated using the result of the product.
According to one embodiment of present invention, the present invention also provides a kind of correction model relevant with wind power generating set
The system of output.The system includes：Processor and memory.Memory storage has computer-readable code, when the calculating
When machine readable code is executed by processor, the method shown in first embodiment, second embodiment or 3rd embodiment is performed.
Moreover, it should be understood that the unit in equipment according to an exemplary embodiment of the present invention can be implemented hardware group
Part and/or component software.Processing of the those skilled in the art according to performed by the unit of restriction, can be for example using scene
Programmable gate array (FPGA) or application specific integrated circuit (ASIC) realize unit.
In addition, method according to an exemplary embodiment of the present invention may be implemented as the meter in computer readable recording medium storing program for performing
Calculation machine code.Those skilled in the art can realize the computer code according to the description to the above method.When the meter
The above method of the present invention is realized when calculation machine code is performed in a computer.
According to probabilistic method and apparatus of the estimation of the present invention model relevant with wind power generating set, Ke Yizhun
Really evaluate the uncertainty of the model relevant with wind power generating set.In addition, by using the uncertainty according to the present invention
Evaluation method and equipment, the output accuracy or accuracy of model with the output of calibration model, can be improved.
Although the present invention, those skilled in the art are particularly shown and described with reference to its exemplary embodiment
It should be understood that in the case where not departing from the spirit and scope of the present invention that claim is limited, form can be carried out to it
With the various changes in details.
Claims (26)
- A kind of 1. probabilistic method for estimating the model relevant with wind power generating set, it is characterised in that methods described bag Include and following estimating step is performed when the model is used every time：Obtain output of model when when previous used；The parameter of distribution based on the error exported described in the output estimation；Parameter based on the distribution obtains the first uncertainty of the output of the model.
- 2. according to the method for claim 1, it is characterised in that first is uncertain for first predetermined value and the distribution The sum of parameters.
- 3. according to the method for claim 2, it is characterised in that the first predetermined value represents predetermined uncertainty.
- 4. according to the method for claim 1, it is characterised in that include the step of the parameter for estimating the distribution of the error： Based on it is described output and the model it is upper once used when output error distribution parameter, estimate that the model exists The parameter of the distribution of the error of output when previous used.
- 5. according to the method for claim 4, it is characterised in that estimate that the model is made when previous by recursive algorithm The parameter of the distribution of the error of the output of used time.
- 6. according to the method for claim 1, it is characterised in that the estimating step also includes：Obtain at least one of the input that the model receives when when previous used parameter；It is determined that the pre-set interval that at least one parameter obtained is each fallen into；The every kind of parameter being directed to respectively at least one parameter, count to the model so far used for each time when Every kind of parameter respectively falls in the total degree of the pre-set interval each fallen into of determination；The second uncertainty of the output of the model is determined based on the total degree.
- 7. according to the method for claim 6, it is characterised in that the of the output of the model is determined based on the total degree Two uncertain steps include：It is determined that the preset areas of the type, the total degree and determination with least one parameter Between the corresponding instruction model output probabilistic value.
- 8. the method according to claim 6 or 7, it is characterised in that at least one parameter is many kinds of parameters, described to estimate Meter step also includes：The pre-set interval that predefined parameter in many kinds of parameters is fallen into is come true from many kinds of parameters Determine the parameter of predetermined quantity,The second uncertain step of the output of the model is determined based on the total degree to be included：Based on the predetermined quantity Parameter corresponding to total degree determine the model output it is second uncertain.
- 9. according to the method for claim 8, it is characterised in that the of the output of the model is determined based on the total degree Two uncertain steps include：When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is the first predetermined pre-set interval, it is based only upon and institute State the second uncertainty that total degree corresponding to predefined parameter determines the output of the model；When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is different from the first predetermined pre-set interval second During predetermined pre-set interval, the of the output of the model is determined based on total degree corresponding with all parameters of many kinds of parameters Two is uncertain.
- 10. according to the method for claim 6, the estimating step also includes：Calculating second is uncertain and first is true Qualitatively threeth uncertainty of the product as the output of the model.
- A kind of 11. method for the output for correcting the model relevant with wind power generating set, it is characterised in that methods described includes：The first uncertainty of the output of the model relevant with wind power generating set is estimated by the method described in claim 1, Or the second uncertainty of the output of the model relevant with wind power generating set is estimated by the method described in claim 6, Or the 3rd uncertainty of the output of the model relevant with wind power generating set is estimated by the method described in claim 10；Use the output of the Uncertainty correction of the estimation model relevant with wind power generating set.
- 12. according to the method for claim 11, it is characterised in that use the Uncertainty correction and wind-driven generator of estimation The step of output of the relevant model of group, includes：Calculate the uncertain and product of the output of estimation.
- 13. a kind of probabilistic equipment for estimating the model relevant with wind power generating set, it is characterised in that the equipment exists The model every time by using when operated to estimate the uncertainty, the equipment includes：Acquiring unit is exported, obtains output of model when when previous used；Estimation of distribution parameters unit, the parameter of the distribution based on the error exported described in the output estimation；First estimation unit, the parameter based on the distribution obtain the first uncertainty of the output of the model.
- 14. equipment according to claim 13, it is characterised in that the first uncertainty is first predetermined value and the distribution Parameters sum.
- 15. equipment according to claim 14, it is characterised in that the first predetermined value represents predetermined uncertainty.
- 16. equipment according to claim 13, it is characterised in that estimation of distribution parameters unit be based on it is described output and it is described Model it is upper once used when output error distribution parameter, estimate that the model is defeated when when previous used The parameter of the distribution of the error gone out.
- 17. equipment according to claim 16, it is characterised in that estimation of distribution parameters unit estimates institute by recursive algorithm State the parameter of the distribution of the error of output of model when when previous used.
- 18. equipment according to claim 13, it is characterised in that the equipment also includes：Input parameter acquiring unit, obtain at least one of the input that the model receives when when previous used parameter；Interval judgement unit, it is determined that the pre-set interval that at least one parameter obtained is each fallen into；Counting unit, respectively for every kind of parameter at least one parameter, count each to the model so far It is secondary by using when every kind of parameter respectively fall in determination the pre-set interval each fallen into total degree；Second estimation unit, determine that the second of the output of the model is uncertain based on the total degree.
- 19. equipment according to claim 18, it is characterised in that the second estimation unit determines and at least one parameter Type, the instruction model corresponding to the total degree and the pre-set interval of determination output probabilistic value.
- 20. the equipment according to claim 18 or 19, it is characterised in that at least one parameter is many kinds of parameters, institute Stating equipment also includes：Selecting unit, the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is come from described more The parameter of predetermined quantity is determined in kind parameter,Total degree corresponding to parameter of second estimation unit based on the predetermined quantity determines the second of the output of the model It is uncertain.
- 21. equipment according to claim 20, it is characterised in thatWhen the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is the first predetermined pre-set interval, the second estimation is single Member is based only upon the second uncertainty that total degree corresponding with the predefined parameter determines the output of the model；When the pre-set interval that the predefined parameter in many kinds of parameters is fallen into is different from the first predetermined pre-set interval second During predetermined pre-set interval, the second estimation unit determines the mould based on total degree corresponding with all parameters of many kinds of parameters The second of the output of type is uncertain.
- 22. equipment according to claim 18, the equipment also includes：3rd estimation unit, it is uncertain to calculate second With threeth uncertainty of the first probabilistic product as the output of the model.
- 23. a kind of equipment for the output for correcting the model relevant with wind power generating set, it is characterised in that the equipment includes：Equipment as claimed in claim 13, either equipment as claimed in claim 18 or as claimed in claim 22 Equipment；Estimation unit, the of the output of the model relevant with wind power generating set of the equipment estimation described in usage right requirement 13 The of the output of the model relevant with wind power generating set of one equipment estimation uncertain or as claimed in claim 18 The 3rd uncertain uncertainty as estimation of two equipment estimations uncertain or as claimed in claim 22, school The output of model just relevant with wind power generating set.
- 24. equipment according to claim 23, it is characterised in that estimation unit calculate estimation it is uncertain with it is described defeated The product gone out.
- 25. a kind of system for the output for correcting the model relevant with wind power generating set, it is characterised in that the system includes：Processor；Memory, computer-readable code is stored with, when the computer-readable code is executed by processor, perform claim will Seek the method described in any one in 1 to 12.
- A kind of 26. computer-readable recording medium for being wherein stored with computer-readable code, when the computer-readable code The method described in any one when being performed in perform claim requirement 1 to 12.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710841534.0A CN107622452A (en) | 2017-09-18 | 2017-09-18 | Method and apparatus for estimating uncertainty of model related to wind turbine generator set |
| PCT/CN2017/118975 WO2019052083A1 (en) | 2017-09-18 | 2017-12-27 | Method and apparatus for estimating uncertainty of model related to wind turbine generator set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710841534.0A CN107622452A (en) | 2017-09-18 | 2017-09-18 | Method and apparatus for estimating uncertainty of model related to wind turbine generator set |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107622452A true CN107622452A (en) | 2018-01-23 |
Family
ID=61090346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710841534.0A Pending CN107622452A (en) | 2017-09-18 | 2017-09-18 | Method and apparatus for estimating uncertainty of model related to wind turbine generator set |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107622452A (en) |
| WO (1) | WO2019052083A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291514A (en) * | 2020-02-03 | 2020-06-16 | 华北电力大学 | Method for rapidly predicting fatigue life of wind turbine generator |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968529B (en) * | 2012-11-14 | 2015-06-03 | 上海城市水资源开发利用国家工程中心有限公司 | Method for quantifying computed result non-determinacy interval of water supply pipe network model |
| CN103676646B (en) * | 2013-12-29 | 2016-01-20 | 哈尔滨理工大学 | A kind of method for estimating state with the random uncertainty of generation and the network control system of distributed sensor time lag |
| CN105183997B (en) * | 2015-09-14 | 2018-03-16 | 哈尔滨工业大学 | A kind of heat conduction model calibration method based on double-layer nested uncertainty propagation |
-
2017
- 2017-09-18 CN CN201710841534.0A patent/CN107622452A/en active Pending
- 2017-12-27 WO PCT/CN2017/118975 patent/WO2019052083A1/en active Application Filing
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291514A (en) * | 2020-02-03 | 2020-06-16 | 华北电力大学 | Method for rapidly predicting fatigue life of wind turbine generator |
| CN111291514B (en) * | 2020-02-03 | 2024-04-05 | 华北电力大学 | Method for rapidly predicting fatigue life of wind turbine generator |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019052083A1 (en) | 2019-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6404832B2 (en) | Method of determining control technical observer for SoC | |
| CN107480028B (en) | Method and device for acquiring usable residual time of disk | |
| CN106933650A (en) | load management method and system of cloud application system | |
| JP2019507442A (en) | Training method and training system for machine learning system | |
| JPWO2013111560A1 (en) | Operation management apparatus, operation management method, and program | |
| WO2018105104A1 (en) | Failure risk index estimation device and failure risk index estimation method | |
| CN114978956B (en) | Method and device for detecting abnormal mutation points of performance of intelligent city network equipment | |
| CN110837432A (en) | Method and device for determining abnormal node in service cluster and monitoring server | |
| JP6200076B2 (en) | Method and system for evaluating measurements obtained from a system | |
| US20080288213A1 (en) | Machine condition monitoring using discontinuity detection | |
| CN107809737B (en) | Method and device for determining network flow of base station | |
| CN107622452A (en) | Method and apparatus for estimating uncertainty of model related to wind turbine generator set | |
| CN108459991B (en) | Method for obtaining equipment reliability data | |
| US20180121794A1 (en) | Method and system for machine failure prediction | |
| JP2020047846A (en) | Data processing method, data processor, data processing system and data processing program | |
| US11636001B2 (en) | Method and system for determining an error threshold value for machine failure prediction | |
| CN107609333A (en) | Method and apparatus for estimating uncertainty of model related to wind turbine generator set | |
| CN111258863B (en) | Data anomaly detection method, device, server and computer readable storage medium | |
| US11762562B2 (en) | Performance analysis apparatus and performance analysis method | |
| Guo et al. | New discount and average optimality conditions for continuous-time Markov decision processes | |
| CN111274687B (en) | Component failure rate prediction method and device, computer equipment and storage medium | |
| Danilov et al. | Dynamic software testing models | |
| CN114760190A (en) | Service-oriented converged network performance anomaly detection method | |
| CN111461342A (en) | System integration method, medium, device and system based on machine learning | |
| Akhavan-Tabatabaei et al. | A method for cycle time estimation of semiconductor manufacturing toolsets with correlations |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180123 |