CN109783873A - The prediction technique and device of axis temperature abnormality state - Google Patents
The prediction technique and device of axis temperature abnormality state Download PDFInfo
- Publication number
- CN109783873A CN109783873A CN201811549136.2A CN201811549136A CN109783873A CN 109783873 A CN109783873 A CN 109783873A CN 201811549136 A CN201811549136 A CN 201811549136A CN 109783873 A CN109783873 A CN 109783873A
- Authority
- CN
- China
- Prior art keywords
- abnormality
- axis temperature
- prediction
- probability
- moment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present invention provides the prediction technique and device of a kind of axis temperature abnormality state.Wherein, the prediction technique of axis temperature abnormality state includes: to judge whether the current axis temperature of generating set is in abnormality；If judging result is to obtain the prediction result of the time interval of abnormality according to the first probability-distribution function and preset first probability in normal condition；Obtain the last abnormality of axis temperature enters the moment, according to it is last enter abnormality at the time of and abnormality time interval prediction result, the prediction result into the moment of acquisition axis temperature abnormality next time；If judging result is, according to the second probability-distribution function and preset second probability, to obtain the prediction result of the prediction result of the duration of this abnormality and the generated energy of this loss in abnormality.The prediction technique and device of axis temperature abnormality state provided in an embodiment of the present invention, the accuracy rate of prediction is higher, and can be improved the operational efficiency of unit.
Description
Technical field
The present embodiments relate to technical field of electric power more particularly to a kind of prediction techniques and dress of axis temperature abnormality state
It sets.
Background technique
The normal operation of generating set be unable to do without the cooperation of major component, and the failure of any one component all can shadow
Ring the generating efficiency of entire unit.Gear-box and generator are large-scale rotary part as component important in generating set,
Bearing is relied primarily in its operational process to complete whole operating, the operating status of bearing directly influences gear-box and hair
The operational efficiency of motor.During unit capacity operation, the temperature of the bearing in sub-health state is easy to increase, temperature
Once being more than that limit value will lead to blower limit and Power operation or directly result in shutdown, to influence power generation efficiency and whole field
Economic benefit.
The factors such as unit bearing operating status and the power of the assembling unit, cabin temperature are related, the variation of state by temperature come
It embodies.Under normal circumstances, after 1.5~2h of motor operation, bearing temperature (abbreviation axis temperature), which stablizes heating≤35K, can think just
Often, but bearing operation rises to 65 DEG C or operation 1.5h or more more than 65 DEG C still without downward trend in short-term (such as 15min), and
Less than 1.5h but bearing equilibrium temperature is more than 75 DEG C, all can be considered bearing temperature it is high (wherein 65 DEG C, 75 DEG C be for summer and
Speech is advisable in cold winter with 55 DEG C, 65 DEG C).
During generating set operation and maintenance, the maintenance of bearing is very important a part.Therefore, axis is grasped in advance
Operating condition is held, the exception of bearing is found in time, and safeguard to bearing, for optimizing bearing maintenance policy, reduce blower
Failure rate is of great significance.The existing thinking analyzed bearing state is divided into two classes, and one kind is based on bear vibration
The spectrum analysis of state, another kind of is the analysis based on bearing state variation.Frequency spectrum analysis method based on bear vibration state
It is very high for data acquisition and data precision requirement, need biggish database to store the data of acquisition, collection process
Need to install specific tool and software, the analysis personnel that need to be equipped with profession analyze etc., to increase maintenance cost and expense
With；Analysis method based on bearing state variation is carried out only in accordance with the variation of bearing itself, and not considering to run bearing has
The factors such as cooling system, the power of the assembling unit of larger impact, therefore it is only limitted to the mutation analysis of bearing state, fail according to analysis
Conclusion provides the bearing maintenance rationalized, and perhaps renewal reward theorem fails to lose the promotion of unit maintenance business efficiency or unit
Electricity carry out quantitatively evaluating.
Summary of the invention
In view of the problems of the existing technology, the embodiment of the present invention provides one kind and overcomes the above problem or at least partly
The prediction technique and device of the axis temperature abnormality state to solve the above problems.
In a first aspect, the embodiment of the present invention provides a kind of prediction technique of axis temperature abnormality state, comprising:
Judge whether the current axis temperature of generating set is in abnormality；
If judging result is, according to the first probability-distribution function and preset first probability, to obtain in normal condition
The prediction result of the time interval of abnormality；
Obtain the last abnormality of axis temperature enters the moment, according to it is described it is last into abnormality at the time of and institute
It states the prediction result of the time interval of abnormality, obtains the prediction result into the moment of axis temperature abnormality next time；
Wherein, first probability-distribution function is obtained according to history axis temperature data；The time interval of abnormality,
It is axis temperature adjacent abnormality twice into the time interval between the moment.
If judging result is, according to the second probability-distribution function and preset second probability, to obtain in abnormality
The prediction result of the duration of this abnormality；
Wherein, second probability-distribution function is obtained according to the history axis temperature data；Abnormality continues
Time is each abnormality into the time interval between moment and the moment of exiting.
According to the output power reduction amount of the prediction result of the duration of this abnormality and the generating set, obtain
Take the prediction result of the generated energy of this loss of the generating set.
Second aspect, the embodiment of the present invention provide a kind of prediction meanss of axis temperature abnormality state, comprising:
Condition judgment module, for judging whether the current axis temperature of generating set is in abnormality；
First prediction module, if being in normal condition, according to the first probability-distribution function and in advance for judging result
If the first probability, obtain the prediction result of the time interval of abnormality；
Second prediction module, for obtaining the entrance moment of the last abnormality of axis temperature, according to the last entrance
At the time of abnormality and the prediction result of the time interval of the abnormality, the entrance of axis temperature abnormality next time is obtained
The prediction result at moment；
Wherein, first probability-distribution function is obtained according to history axis temperature data；The time interval of abnormality,
It is axis temperature adjacent abnormality twice into the time interval between the moment.
Third prediction module, if being in abnormality, according to the second probability-distribution function and in advance for judging result
If the second probability, obtain the duration of this abnormality prediction result and this loss generated energy prediction knot
Fruit；
Wherein, second probability-distribution function is obtained according to the history axis temperature data；Abnormality continues
Time is each abnormality into the time interval between moment and the moment of exiting.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, comprising:
At least one processor；And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
Order is able to carry out axis temperature abnormality provided by any possible implementation in the various possible implementations of first aspect
The prediction technique of state.
Fourth aspect, the embodiment of the present invention provide a kind of non-transient computer readable storage medium, the non-transient calculating
Machine readable storage medium storing program for executing stores computer instruction, and the computer instruction makes the various possibility of the computer execution first aspect
Implementation in axis temperature abnormality state provided by any possible implementation prediction technique.
The prediction technique and device of axis temperature abnormality state provided in an embodiment of the present invention, according to the time interval of abnormality
Time Distribution, obtain the prediction result into the moment of axis temperature abnormality next time, the accuracy rate of prediction is higher, from
And can make staff before the arrival of abnormality next time, prepare bearing maintenance or renewal reward theorem, early to reduce
Economic loss caused by abnormal or shutdown, improves the operational efficiency of unit.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is the flow diagram according to the prediction technique of axis temperature abnormality state provided in an embodiment of the present invention；
Fig. 2 is the functional block diagram according to the prediction meanss of axis temperature abnormality state provided in an embodiment of the present invention；
Fig. 3 is the structural block diagram according to electronic equipment provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Embodiment in the present invention, those of ordinary skill in the art are not having
Every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
In order to overcome the above problem of the prior art, the embodiment of the present invention provides a kind of prediction technique of axis temperature abnormality state
And device, inventive concept are, respectively according to the time interval of axis temperature abnormality state and the probability of abnormality duration point
Cloth function predicts the duration at the arrival moment and current abnormality of next abnormality, so as to prepare bearing early
Maintenance or renewal reward theorem predict power generation situation, estimate the loss of generated energy.
Fig. 1 is the flow diagram according to the prediction technique of axis temperature abnormality state provided in an embodiment of the present invention.Such as Fig. 1 institute
Show, a kind of prediction technique of axis temperature abnormality state includes:
Step S101, judge whether the current axis temperature of generating set is in abnormality.
It should be noted that can be divided according to bearing temperature to the state of bearing temperature.Axis temperature is in abnormal shape
State, spindle temperature are at or above preset temperature threshold；Axis temperature is in normal condition, and spindle temperature is lower than preset temperature threshold
Value.Axis temperature abnormality state enters the moment, at the time of spindle temperature in normal condition by being changed into abnormality；Axis Wen Yi
Normal state exits the moment, at the time of spindle temperature in abnormality by being changed into normal condition.Axis temperature is in normal shape
State indicates that bearing is in normal condition；Axis temperature is in abnormality, indicates that bearing is in abnormality.
It is understood that axis temperature can be made pre- from rising to over lower than preset temperature threshold due to the raising of axis temperature
If temperature threshold, rise to over preset temperature threshold from lower than preset temperature threshold, i.e., axis temperature is by being in normal shape
State is changed into abnormality；Due to the reduction of axis temperature, axis temperature can be made pre- from decreasing below more than preset temperature threshold
If temperature threshold, from being more than that preset temperature threshold decreases below preset temperature threshold, i.e., axis temperature is by abnormal shape
State is changed into normal condition.Axis temperature constantly changes in the use process of bearing, and axis temperature can show alternately in abnormal
State and normal condition.
After the current axis temperature for obtaining generating set, and judge whether current axis temperature is greater than preset temperature threshold.Currently
Axis temperature refers to the axis temperature at current time.
If being equal to or more than, judging result is that current axis temperature is in abnormality；If being less than, judging result is current
Axis temperature is in normal condition.
Preset temperature threshold can determine that the embodiment of the present invention is to preset according to the running technology parameter of generating set
The specific value of temperature threshold is with no restriction.
If step S102, judging result is in normal condition, according to the first probability-distribution function and preset first
Probability obtains the prediction result of the time interval of abnormality；Wherein, the first probability-distribution function is according to history axis temperature data
It obtains；The time interval of abnormality is the adjacent abnormality twice of axis temperature into the time interval between the moment.
When judging result is in normal condition, concern when axis temperature in normal condition from being changed into
In abnormality, that is, need to predict at the time of entering abnormality next time to axis temperature.
Enter prediction result at the time of abnormality next time in order to obtain, first this is in front of normal condition most
It is close primary into abnormality and this be in after normal condition the last time interval between abnormality into
Row prediction.
The time interval of abnormality, refer in particular to abnormality twice into the time interval between the moment.
It is understood that before the time interval of predicted anomaly state, according to history axis temperature the first probability of data acquisition
Distribution function.First probability-distribution function F1, it is the probability-distribution function of the time interval of abnormality.First probability distribution letter
Number, can reflect the regularity of distribution of the time interval of abnormality.
History axis temperature data refer to the actual axle temperature interior for the previous period at current time.
According to the first probability-distribution function F1With preset first probability uplimN, obtain so that probability is greater than the first probability
uplimNTime interval, the prediction result of the time interval as abnormality.
Wherein, uplimN∈ (0,1].Suitable first probability uplim can be selected rule of thumb with axis temperature modelN.It is right
In the first probability uplimNSpecific value, the embodiment of the present invention is not specifically limited.
Step S103, at the time of the acquisition axis temperature last time enters abnormality, when according to the entrance of last abnormality
It carves and the prediction result of the time interval of abnormality, obtains the prediction result into the moment of axis temperature abnormality next time.
It should be noted that axis temperature is obtained according to the time interval of prediction, it is therefore, the last before current time
What axis temperature was in abnormality is available into the moment.After determining at the time of the axis temperature last time enters abnormality,
By axis temperature is last enter abnormality at the time of plus abnormality time interval prediction result, can obtain under axis temperature
The prediction result into the moment of abnormality, it can prediction axis temperature next time abnormality enter the moment.
The prediction result into the moment of axis temperature abnormality next time meets following condition:
F1(t > Tnext) > uplimN
Wherein, TnextIndicate that axis temperature enters the time interval prediction result of abnormality next time.
Above-mentioned condition explanation, it is assumed that last time enters the abnormality moment for T0, in T0+TnextMoment, axis temperature have uplimN's
Probability is more than preset temperature threshold.
Acquisition axis temperature after the prediction result into the moment of abnormality, can obtain abnormality next time next time
Waiting time prediction result.The waiting time of abnormality next time refers to that current time, (current time was in normal shape
State) and next time abnormality into the time between the moment.
Time Distribution of the embodiment of the present invention according to the time interval of abnormality, this time regularity of distribution random groups
Dynamic operational behaviour and change, obtain the prediction result into the moment of axis temperature abnormality next time, the accuracy rate of prediction is more
Height prepares bearing maintenance or renewal reward theorem so as to make staff before the arrival of abnormality next time early, with
Economic loss caused by abnormal or shutdown is reduced, and can be improved the operational efficiency of unit.
Content based on the various embodiments described above judges also to wrap after whether the current axis temperature of generating set is in abnormality
It includes: if judging result is that it is different to obtain this according to the second probability-distribution function and preset second probability in abnormality
The prediction result of the duration of normal state；Wherein, the second probability-distribution function is obtained according to history axis temperature data；It is abnormal
The duration of state is each abnormality into the time interval between moment and the moment of exiting.
When judging result be it is when in an abnormal state, concern that this is in when abnormality terminates, that is, needs
The duration of axis temperature this abnormality is predicted.
The duration of each abnormality, refer to the secondary abnormality between the time between moment and the moment of exiting
Every.
It is understood that before the duration for predicting this abnormality, according to history axis temperature data acquisition second
Probability-distribution function.Second probability-distribution function F2, it is the probability-distribution function of the duration of abnormality.Second probability point
Cloth function can reflect the Time Distribution of the duration of abnormality.
According to the second probability-distribution function F2With preset second probability uplimL, obtain so that probability is greater than the second probability
uplimLTime interval, the prediction result of the duration as this abnormality.
Wherein, uplimL∈(0,1].Suitable second probability uplim can be selected rule of thumb with axis temperature modelL.It is right
In the second probability uplimLSpecific value, the embodiment of the present invention is not specifically limited.
It should be noted that uplimNIn footnote N and uplimLIn footnote L be only used for distinguishing preset first probability
With preset second probability.
Since axis temperature was obtained according to the time, the last axis temperature is in abnormality before current time
It is available into the moment (i.e. the entrance moment of this abnormality).Determine this abnormality of axis temperature enters the moment
Later, by the entrance moment of this abnormality of axis temperature, in addition the prediction result of the duration of this abnormality, can obtain
Obtain this abnormality of axis temperature exits the moment.That is, can be according to the different with this into the moment of this abnormality
The prediction result of the duration of normal state, obtains the prediction result for exiting the moment of this abnormality.
The prediction result for exiting the moment of this abnormality of axis temperature meets following condition:
F2(t > Tlast) > uplimL
Wherein, TlastIndicate the prediction result for exiting the moment of this abnormality of axis temperature.
Above-mentioned condition explanation, it is assumed that this enters the abnormality moment for T0, in T0+TlastMoment, axis temperature have uplimL's
Probability is higher than preset temperature threshold.
It is understood that in the case that current axis temperature is in abnormality, it can also be according to the first probability-distribution function
With preset first probability, the prediction result of the time interval of abnormality is obtained, and obtains axis temperature abnormality next time
Into the prediction result at moment.
The embodiment of the present invention obtains axis temperature abnormal shape next time according to the Time Distribution of the duration of abnormality
The prediction result into the moment of state, the accuracy rate of prediction are higher.
Content based on the various embodiments described above, the prediction result for obtaining the duration of this abnormality are also wrapped later
It includes: according to the output power reduction amount of the prediction result of the duration of this abnormality and generating set, obtaining generator
The prediction result of the generated energy of this loss of group.
It specifically, can be according to this abnormality after the prediction result for obtaining the duration of this abnormality
Duration prediction result and generating set output power reduction amount, prediction generating set this loss power generation
Amount.
The generated energy of this loss of generating set, if referring to, generating set is in normal in the duration of this abnormality
The generated energy of state, and this abnormality generated energy difference.
Output power reduction amount refers to that generating set is in the difference of normal condition and the output power in abnormality.
The prediction result of the generated energy of this loss of generating set, is equal to output power reduction amount Δ P and this normal state
Duration prediction result product.
The embodiment of the present invention is according to the prediction result of the duration of this abnormality and the output power of generating set
Reduction amount obtains the prediction result of this generated energy lost of generating set, quantifies to the generated energy of generating set loss
Evaluation, the accuracy rate of prediction are higher.
Content based on the various embodiments described above, the specific steps for obtaining the first probability-distribution function include: to history axis temperature
Data are fitted, and determine the coefficient in polynary distributed lag model, obtain polynary distributed lag prediction model；According to polynary point
Cloth delay prediction model obtains prediction axis temperature series；According to prediction axis temperature series, the adjacent state change twice of prediction axis temperature is obtained
Occur the moment between time interval, makeup time intervening sequence；According to time interval sequence, acquisition time intervening sequence
Probability density function；According to the probability density function of time interval sequence, the probability-distribution function of acquisition time intervening sequence is made
For the first probability-distribution function.
Specifically, history axis temperature data can use actual axle temperature sequence Y={ y1, y2..., ymIndicate.Wherein, m is indicated
The number of actual axle temperature in history axis temperature data.
It is influenced since current axis temperature has axis temperature later, by the influence factor being had an impact to axis temperature except axis temperature
Referred to as other influences factor.Other influences factor can use vector X={ x1, x2..., xnIndicate.Wherein.N indicates other shadows
The number of the factor of sound.Other influences factor may include cabin temperature, power, radiating state etc..
Relationship between each influence factor of the mild axis temperature of axis can be described using polynary distributed lag model.
For moment t,
Wherein, ytIndicate actual axle temperature；T=1,2 ..., m；I=1,2 ..., n；αi、βi、γ1、γ2It is stagnant for polynary distribution
Coefficient in model afterwards；yt-1、yt-2Respectively indicate ytFirst-order lag item and second-order lag item；xt-1,iIndicate i-th of other shadow
The first-order lag item of the factor of sound；xt-2,iIndicate the second-order lag item of i-th of other influences factor.
It is fitted according to the value of history axis temperature data and the corresponding other influences factor of each history axis temperature data, determination is more
Coefficient in first distributed lag model.Polynary distributed lag model after coefficient therein is determined is pre- as polynary distributed lag
Survey model.
The value of each moment other influences factor is substituted into polynary distributed lag prediction model, each moment is obtained by iteration
The predicted value (i.e. prediction axis temperature) of axis temperature, predicted composition axis temperature sequence { y 't}.The above-mentioned moment refers to obtaining for each history axis temperature data
Take the moment.Predict axis temperature sequence { y 'tIn respectively predict the putting in order as time sequencing of axis temperature.
Prediction axis temperature sequence is time series, is had according to the prediction axis temperature sequence of polynary distributed lag model construction as follows
Advantage: in view of the factor for influencing bearing temperature variation is intricate, model, minor effect factor is added in major influence factors
Model only is added as residual error portion, the numerous reasons for influencing bearing temperature are simplified to a certain extent, by principal element list
It solely separates and is studied, avoid influence of the disturbing factor to result；The predicted value of chosen axis temperature is as state demarcation
Time series can also obtain accurate bearing state conducive to changing greatly in other non-principal factors；The error of model is distributed
It is more stable, it can be realized the purpose divided only in accordance with major influence factors to bearing state.
For predicting axis temperature sequence { y 'tIn each prediction axis temperature, judge whether the prediction axis temperature is greater than the temperature of prediction
Spend threshold valueCan will be more thanIt is denoted as "+", is less than and is denoted as "-", obtains status switch Z={ z1,z2,z3,...,zm}。
If two neighboring value is different in status switch Z, illustrates that axis temperature state is changed, i.e., be changed by normal condition
Abnormality, or normal condition is changed by abnormality；If two neighboring value is identical, illustrate that axis temperature state does not change,
It is continuously abnormality or normal condition.
It obtains adjacent at the time of enter abnormality twice in status switch Z, and then obtains and adjacent enter abnormal shape twice
The time interval of state, makeup time intervening sequence δ={ a+,1, a+,2..., a+,u}。
Wherein, i=1,2,3 ..., u；U indicates the number for entering abnormality in status switch Z；a+,iIndicate that i-th is different
The time interval of normal state；Indicate that i-th enters the generation moment of abnormality.
According to time interval sequence δ, the first probability density function f of available time interval sequence δ1.First probability
Density function f1, it is the probability density function of the time interval of abnormality.First probability density function can reflect abnormal shape
The Time Distribution of the time interval of state.
To the first probability density function f1It is integrated, obtains the first probability-distribution function F1。
The embodiment of the present invention is according to polynary distributed lag model and history axis temperature the first probability-distribution function of data acquisition, energy
The first probability-distribution function more to tally with the actual situation is obtained, so as to obtain accuracy rate more according to the first probability-distribution function
The prediction result into the moment of high axis temperature abnormality next time.
Content based on the various embodiments described above, the specific steps for obtaining the second probability-distribution function include: according to prediction axis
Warm sequence obtains the duration of each abnormality of prediction axis temperature, forms duration time sequence；According to duration time sequence,
Obtain the probability density function of duration time sequence；According to the probability density function of duration time sequence, duration sequence is obtained
The probability-distribution function of column, as the second probability-distribution function.
Specifically, at the time of being changed into abnormality by normal condition every time in acquisition status switch Z, and according to state sequence
At the time of being changed into abnormality by normal condition every time in column Z, the duration of each abnormality is obtained, when composition continues
Between sequence τ={ b+,1, b+,2..., b+,v}。
Wherein, i=1,2,3 ..., v；V indicates the segments of persistent anomaly state in status switch Z；b+,iIndicate label
For the time of i-th persistent anomaly state；Respectively indicate label i-th persistent anomaly state at the time of occur and
At the time of end.
According to duration time sequence τ, the second probability density function f of available duration time sequence τ2.Second probability
Density function f2, it is the probability density function of the duration of abnormality.Second probability density function can reflect abnormal shape
The Time Distribution of the duration of state.
To the second probability density function f2It is integrated, obtains the second probability-distribution function F2。
The embodiment of the present invention is according to polynary distributed lag model and history axis temperature the second probability-distribution function of data acquisition, energy
The second probability-distribution function more to tally with the actual situation is obtained, so as to obtain accuracy rate more according to the second probability-distribution function
The prediction result of the duration of high this abnormality of axis temperature.
Content based on the various embodiments described above judges also to wrap before whether the current axis temperature of generating set is in abnormality
It includes: obtaining current axis temperature.
It is understood that obtaining the current axis temperature of generating set by measurement；Obtain generating set current axis temperature it
Afterwards, judge whether the current axis temperature of generating set is in abnormality.
Current axis temperature is obtained when can be set according to preset time intervals.Preset time interval can be second grade or divide
Clock grade, can also day be unit, the embodiment of the present invention is not specifically limited this.
For the ease of the understanding to various embodiments of the present invention, illustrate below by an example provided in an embodiment of the present invention
The process of the prediction technique of axis temperature abnormality state.
The operation data of certain generating set 2016.03-2016.07 is obtained, the time interval of data is 10min.Run number
According to including: gearbox shaft temperature (y), power (pow), cabin temperature (NacTem).After the exceptional value for rejecting operation data, tooth is chosen
Roller box axis temperature lags 2 rank yt-1、yt-2, 2 rank pow of each index lagt-1、powt-2、NacTemt-1、NacTemt-2Obtain polynary distribution
Delay prediction model are as follows:
yt=c0+γ1yt-1+γ2yt-2+α1powt-1+β1powt-2+α2NacTemt-1+β2NacTemt-2
According to above-mentioned polynary distributed lag prediction model, prediction axis temperature sequence { y ' is obtainedt}。
Choose temperature thresholdIt is 65 DEG C, will predicts axis temperature sequence { y 'tIn more than or equal to 65 DEG C of data markers be
" 1 ", the data markers less than 65 DEG C are " 0 ", obtain status switch Z={ z1,z2,z3,…,zm}。
According to status switch Z, each adjacent time interval for entering state " 1 " from state " 0 " is calculated, time interval is obtained
Sequence δ={ a+, 1, a+, 2..., a+, u}。
According to status switch Z, the duration of each state " 1 " is calculated, obtains duration time sequence τ={ b+,1,
b+, 2..., b+, v}。
According to time interval sequence δ and duration time sequence τ, the corresponding probability density function f of two sequences is obtained respectively1
And f2。
The data unit operation that 2016.08 period time intervals are 10min is chosen, uplim is chosenN=0.8 and uplimL=
0.83。
Current time t is abnormality start time, according to the distribution of time interval sequence δ and duration time sequence τ, meter
Calculating and obtaining the prediction result of the duration of current abnormality is 1.1 days, next time at the beginning of abnormality and currently
The prediction result of time interval between moment t is 12.36 days.It follows that current abnormality has at least 83% probability
Continuing 1.1 days, next abnormality has 80% probability at least to occur once in following 12.36 days, thus in future
In 12.36 days, suitable technological transformation or maintenance policy can be chosen according to operating status, to promote unit operation efficiency.
Fig. 2 is the functional block diagram according to the prediction meanss of axis temperature abnormality state provided in an embodiment of the present invention.Based on above-mentioned
The content of each embodiment, as shown in Fig. 2, the prediction meanss of the axis temperature abnormality state include that condition judgment module 201, first is predicted
Module 202 and the second prediction module 203, in which:
Condition judgment module 201, for judging whether the current axis temperature of generating set is in abnormality；
First prediction module 202, if for judging result be in normal condition, according to the first probability-distribution function and
Preset first probability, obtains the prediction result of the time interval of abnormality；
Second prediction module 203 enters different at the time of entering abnormality for obtaining the axis temperature last time according to the last time
At the time of normal state and the prediction result of the time interval of abnormality, obtain at the time of axis temperature enters abnormality next time
Prediction result；
Wherein, the first probability-distribution function is obtained according to history axis temperature data；The time interval of abnormality is axis
The adjacent abnormality twice of temperature into the time interval between the moment.
Specifically, after the current axis temperature of the acquisition of condition judgment module 201 generating set, and judge whether current axis temperature is big
In preset temperature threshold.If being equal to or more than, judging result is that current axis temperature is in abnormality；If being less than, judge
As a result normal condition is in for current axis temperature.
If judging result is in normal condition, the first prediction module 202 is according to the first probability-distribution function and preset
First probability obtains the prediction result of the time interval of abnormality.
Second prediction module 203, by axis temperature is last enter abnormality at the time of plus abnormality time interval
Prediction result, the prediction result into the moment of axis temperature abnormality next time can be obtained, it can prediction axis temperature is next
Secondary abnormality enters the moment.
The prediction meanss of axis temperature abnormality state provided in an embodiment of the present invention, for executing axis provided in an embodiment of the present invention
The prediction technique of temperature abnormality state, each module that the prediction meanss of the axis temperature abnormality state include realize the specific side of corresponding function
Method and process are detailed in the embodiment of the prediction technique of above-mentioned axis temperature abnormality state, and details are not described herein again.
Prediction technique of the prediction meanss of the axis temperature abnormality state for the axis temperature abnormality state of foregoing embodiments.Cause
This, description and definition in the prediction technique of the axis temperature abnormality state in foregoing embodiments can be used for implementation of the present invention
The understanding of each execution module in example.
The embodiment of the present invention obtains axis temperature abnormal shape next time according to the Time Distribution of the time interval of abnormality
The prediction result into the moment of state, the accuracy rate of prediction is higher, so as to arrive staff in abnormality next time
Before coming, preparing bearing maintenance early, perhaps renewal reward theorem and can be mentioned with reducing abnormal or economic loss caused by shutting down
The operational efficiency of high unit.
Content based on the various embodiments described above, the prediction meanss of axis temperature abnormality state include: third prediction module, if for
Judging result is, then according to the second probability-distribution function and preset second probability, to obtain this abnormal shape in abnormality
The prediction result of the prediction result of the duration of state and the generated energy of this loss；Wherein, the second probability-distribution function is
It is obtained according to history axis temperature data；The duration of abnormality into the moment and exits the moment for each abnormality
Between time interval.
Specifically, if judging result is in abnormality, third prediction module is according to the second probability-distribution function and in advance
If the second probability, obtain the prediction result of the duration of this abnormality.
The embodiment of the present invention obtains axis temperature abnormal shape next time according to the Time Distribution of the duration of abnormality
The prediction result into the moment of state, the accuracy rate of prediction are higher.
Fig. 3 is the structural block diagram according to electronic equipment provided in an embodiment of the present invention.Content based on the above embodiment, such as
Shown in Fig. 3, which may include: processor (processor) 301, memory (memory) 302 and bus 303；Its
In, processor 301 and memory 302 pass through bus 303 and complete mutual communication；Processor 301 is stored in for calling
In reservoir 302 and the computer program instructions that can be run on processor 301, to execute provided by above-mentioned each method embodiment
Method, for example, judge whether the current axis temperature of generating set is in abnormality；If judging result is in normal shape
State obtains the prediction result of the time interval of abnormality then according to the first probability-distribution function and preset first probability；It obtains
Take the last abnormality of axis temperature enters the moment, according to it is last enter abnormality at the time of and abnormality time between
Every prediction result, obtain the prediction result into the moment of axis temperature abnormality next time.
Another embodiment of the present invention discloses a kind of computer program product, and computer program product is non-transient including being stored in
Computer program on computer readable storage medium, computer program include program instruction, when program instruction is held by computer
When row, computer is able to carry out method provided by above-mentioned each method embodiment, for example, judges the current axis of generating set
Whether temperature is in abnormality；If judging result is in normal condition, according to the first probability-distribution function and preset the
One probability obtains the prediction result of the time interval of abnormality；Obtain the last abnormality of axis temperature enters the moment, according to
At the time of last time enters abnormality and the prediction result of the time interval of abnormality, axis temperature abnormality next time is obtained
The prediction result into the moment.
In addition, the logical order in above-mentioned memory 302 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.Based on this understanding, originally
The technical solution of the inventive embodiments substantially part of the part that contributes to existing technology or the technical solution in other words
It can be embodied in the form of software products, which is stored in a storage medium, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes the present invention respectively
The all or part of the steps of a embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory
(ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk
Etc. the various media that can store program code.
Another embodiment of the present invention provides a kind of non-transient computer readable storage medium, non-transient computer readable storages
Medium storing computer instruction, computer instruction makes computer execute method provided by above-mentioned each method embodiment, such as wraps
It includes: judging whether the current axis temperature of generating set is in abnormality；If judging result is in normal condition, according to first
Probability-distribution function and preset first probability, obtain the prediction result of the time interval of abnormality；It is last to obtain axis temperature
Abnormality enters the moment, according to it is last enter abnormality at the time of and abnormality time interval prediction knot
Fruit obtains the prediction result into the moment of axis temperature abnormality next time.
The apparatus embodiments described above are merely exemplary, wherein unit can be as illustrated by the separation member
Or may not be and be physically separated, component shown as a unit may or may not be physical unit, i.e.,
It can be located in one place, or may be distributed over multiple network units.It can select according to the actual needs therein
Some or all of the modules achieves the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creative labor
In the case where dynamic, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Such understanding, above-mentioned skill
Substantially the part that contributes to existing technology can be embodied in the form of software products art scheme in other words, the calculating
Machine software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used
So that a computer equipment (can be personal computer, server or the network equipment etc.) executes above-mentioned each implementation
The method of certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of prediction technique of axis temperature abnormality state characterized by comprising
Judge whether the current axis temperature of generating set is in abnormality；
If judging result is, according to the first probability-distribution function and preset first probability, to obtain abnormal in normal condition
The prediction result of the time interval of state；
Obtain the last abnormality of axis temperature enters the moment, according to it is described it is last into abnormality at the time of and it is described different
The prediction result of the time interval of normal state obtains the prediction result into the moment of axis temperature abnormality next time；
Wherein, first probability-distribution function is obtained according to history axis temperature data；The time interval of abnormality is axis
The adjacent abnormality twice of temperature into the time interval between the moment.
2. the prediction technique of axis temperature abnormality state according to claim 1, which is characterized in that the judgement generating set
Whether current axis temperature is in after abnormality further include:
If judging result is to obtain this according to the second probability-distribution function and preset second probability in abnormality
The prediction result of the duration of abnormality；
Wherein, second probability-distribution function is obtained according to the history axis temperature data；The duration of abnormality,
It is each abnormality into time interval between moment and the moment of exiting.
3. the prediction technique of axis temperature abnormality state according to claim 2, which is characterized in that obtain this abnormality
After the prediction result of duration further include:
According to the output power reduction amount of the prediction result of the duration of this abnormality and the generating set, institute is obtained
State the prediction result of this generated energy lost of generating set.
4. the prediction technique of axis temperature abnormality state according to claim 2 or 3, which is characterized in that it is general to obtain described first
The specific steps of rate distribution function include:
The history axis temperature data are fitted, the coefficient in polynary distributed lag model is determined, obtains polynary distributed lag
Prediction model；
According to polynary distributed lag prediction model, prediction axis temperature sequence is obtained；
According to the prediction axis temperature sequence, between the time occurred between the moment for obtaining the adjacent state change twice of prediction axis temperature
Every makeup time intervening sequence；
According to the time interval sequence, the probability density function of the time interval sequence is obtained；
According to the probability density function of the time interval sequence, the probability-distribution function of the time interval sequence is obtained, is made
For first probability-distribution function.
5. the prediction technique of axis temperature abnormality state according to claim 4, which is characterized in that obtain second probability point
The specific steps of cloth function include:
According to the prediction axis temperature sequence, the duration of each abnormality of prediction axis temperature is obtained, duration time sequence is formed；
According to the duration time sequence, the probability density function of the duration time sequence is obtained；
According to the probability density function of the duration time sequence, the probability-distribution function of the duration time sequence is obtained, is made
For second probability-distribution function.
6. the prediction technique of axis temperature abnormality state according to claim 1, which is characterized in that judge the current of generating set
Whether axis temperature is in front of abnormality further include:
Obtain the current axis temperature.
7. a kind of prediction meanss of axis temperature abnormality state characterized by comprising
Condition judgment module, for judging whether the current axis temperature of generating set is in abnormality；
First prediction module, if being in normal condition, according to the first probability-distribution function and preset for judging result
First probability obtains the prediction result of the time interval of abnormality；
Second prediction module enters the moment for obtain the last abnormality of axis temperature, according to it is described it is last enter it is abnormal
At the time of state and the prediction result of the time interval of the abnormality, obtaining axis temperature, abnormality enters the moment next time
Prediction result；
Wherein, first probability-distribution function is obtained according to history axis temperature data；The time interval of abnormality is axis
The adjacent abnormality twice of temperature into the time interval between the moment.
8. the prediction meanss of axis temperature abnormality state according to claim 7, which is characterized in that further include:
Third prediction module, if being in abnormality, according to the second probability-distribution function and preset for judging result
Second probability obtains the prediction result of the prediction result of the duration of this abnormality and the generated energy of this loss；
Wherein, second probability-distribution function is obtained according to the history axis temperature data；The duration of abnormality,
It is each abnormality into time interval between moment and the moment of exiting.
9. a kind of electronic equipment characterized by comprising
At least one processor；And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to instruct energy
Enough methods executed as described in claim 1 to 6 is any.
10. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Computer instruction is stored up, the computer instruction makes the computer execute the method as described in claim 1 to 6 is any.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811549136.2A CN109783873B (en) | 2018-12-18 | 2018-12-18 | Method and device for predicting abnormal state of shaft temperature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811549136.2A CN109783873B (en) | 2018-12-18 | 2018-12-18 | Method and device for predicting abnormal state of shaft temperature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109783873A true CN109783873A (en) | 2019-05-21 |
| CN109783873B CN109783873B (en) | 2023-04-07 |
Family
ID=66497226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811549136.2A Active CN109783873B (en) | 2018-12-18 | 2018-12-18 | Method and device for predicting abnormal state of shaft temperature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109783873B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717376A (en) * | 2019-08-23 | 2020-01-21 | 深圳绿米联创科技有限公司 | Room state prediction method and device, electronic device and storage medium |
| CN112329249A (en) * | 2020-11-11 | 2021-02-05 | 中国人民解放军陆军工程大学 | Failure prediction method of bearing and terminal equipment |
| CN112697306A (en) * | 2020-12-09 | 2021-04-23 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for detecting temperature abnormality inside battery pack and related product |
| CN117034176A (en) * | 2023-10-08 | 2023-11-10 | 西安石油大学 | Casting abnormality prediction system in directional solidification process |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103063319A (en) * | 2012-12-30 | 2013-04-24 | 兰州理工大学 | Temperature measurement system for bearings at the front and back of high speed machining central motorized spindle |
| US20150227837A1 (en) * | 2012-09-03 | 2015-08-13 | Isis Innovation Limited | System monitoring |
| CN204612929U (en) * | 2015-05-24 | 2015-09-02 | 王伟 | Based on the bearing fault diagnosing apparatus that infrared measurement of temperature and vibration signal detect |
| CN106080655A (en) * | 2016-08-24 | 2016-11-09 | 中车株洲电力机车研究所有限公司 | Detection method, device and the train that a kind of train axle temperature is abnormal |
| US20170236065A1 (en) * | 2014-08-20 | 2017-08-17 | Cassantec Ag | Malfunction prediction for components and units of technical entities |
| CN108051211A (en) * | 2017-12-29 | 2018-05-18 | 湖南优利泰克自动化系统有限公司 | A kind of wind generator set main shaft holds temperature pre-warning diagnostic method |
-
2018
- 2018-12-18 CN CN201811549136.2A patent/CN109783873B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150227837A1 (en) * | 2012-09-03 | 2015-08-13 | Isis Innovation Limited | System monitoring |
| CN103063319A (en) * | 2012-12-30 | 2013-04-24 | 兰州理工大学 | Temperature measurement system for bearings at the front and back of high speed machining central motorized spindle |
| US20170236065A1 (en) * | 2014-08-20 | 2017-08-17 | Cassantec Ag | Malfunction prediction for components and units of technical entities |
| CN204612929U (en) * | 2015-05-24 | 2015-09-02 | 王伟 | Based on the bearing fault diagnosing apparatus that infrared measurement of temperature and vibration signal detect |
| CN106080655A (en) * | 2016-08-24 | 2016-11-09 | 中车株洲电力机车研究所有限公司 | Detection method, device and the train that a kind of train axle temperature is abnormal |
| CN108051211A (en) * | 2017-12-29 | 2018-05-18 | 湖南优利泰克自动化系统有限公司 | A kind of wind generator set main shaft holds temperature pre-warning diagnostic method |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717376A (en) * | 2019-08-23 | 2020-01-21 | 深圳绿米联创科技有限公司 | Room state prediction method and device, electronic device and storage medium |
| CN110717376B (en) * | 2019-08-23 | 2022-09-09 | 深圳绿米联创科技有限公司 | Room state prediction method, room state prediction device, electronic equipment and storage medium |
| CN112329249A (en) * | 2020-11-11 | 2021-02-05 | 中国人民解放军陆军工程大学 | Failure prediction method of bearing and terminal equipment |
| CN112697306A (en) * | 2020-12-09 | 2021-04-23 | 东软睿驰汽车技术（沈阳）有限公司 | Method and device for detecting temperature abnormality inside battery pack and related product |
| CN117034176A (en) * | 2023-10-08 | 2023-11-10 | 西安石油大学 | Casting abnormality prediction system in directional solidification process |
| CN117034176B (en) * | 2023-10-08 | 2024-01-09 | 西安石油大学 | Casting abnormality prediction system in directional solidification process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109783873B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109783873A (en) | The prediction technique and device of axis temperature abnormality state | |
| Correa-Florez et al. | Stochastic operation of home energy management systems including battery cycling | |
| Bracale et al. | Multivariate quantile regression for short-term probabilistic load forecasting | |
| JP7051856B2 (en) | Systems and methods for controlling dynamic energy storage systems | |
| CN105790265B (en) | A kind of uncertain Unit Combination model and method for solving considering AC power flow constraint | |
| CN108334997B (en) | Standby optimization method and device based on support fault event constraint unit combination | |
| Mohammadi et al. | Machine learning assisted stochastic unit commitment during hurricanes with predictable line outages | |
| CN110135064B (en) | Method, system and controller for predicting temperature faults of rear bearing of generator | |
| US20210224755A1 (en) | Decision method of condition-based maintenance to power grid risk | |
| CN102195362A (en) | Dynamic probability dispatching method for calculating system operation reliability | |
| Li et al. | Investment optimization of grid-scale energy storage for supporting different wind power utilization levels | |
| Li et al. | Efficient resource scaling based on load fluctuation in edge-cloud computing environment | |
| CN104316879A (en) | Method for forecasting service life of lead-acid storage battery | |
| CN104598984A (en) | Fuzzy neural network based fault prediction method | |
| CN107145975A (en) | The Forecasting Methodology of optical transmission device spare part quantity | |
| Liu et al. | Online short-term reliability evaluation using a fast sorting technique | |
| JP2019216534A (en) | Electric power system monitoring system, electric power system monitoring method, and program | |
| CN115456304A (en) | Offshore wind farm reliability index calculation method and device considering typhoon influence | |
| Radovanovic et al. | Power modeling for effective datacenter planning and compute management | |
| US20200044446A1 (en) | System and method for management of an electricity distribution grid | |
| Hamedi et al. | Optimization of preventive maintenance scheduling based on Monte Carlo simulation in processing plants | |
| KR102232931B1 (en) | Method, apparatus and system for managing charge and discharge of ess group for voltage stabilization of system connected to distributed generators | |
| Amjady et al. | Optimal reliable operation of hydrothermal power systems with random unit outages | |
| CN112686555A (en) | Comprehensive analysis method, device and system for regional power distribution network | |
| Wang et al. | HARRD: Real-time software rejuvenation decision based on hierarchical analysis under weibull distribution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |