CN109886544A - Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model - Google Patents
Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model Download PDFInfo
- Publication number
- CN109886544A CN109886544A CN201910042715.6A CN201910042715A CN109886544A CN 109886544 A CN109886544 A CN 109886544A CN 201910042715 A CN201910042715 A CN 201910042715A CN 109886544 A CN109886544 A CN 109886544A
- Authority
- CN
- China
- Prior art keywords
- operation data
- efficiency
- curve model
- efficiency curve
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention discloses a kind of method, apparatus, readable medium and electronic equipments for constructing energy efficiency of equipment curve model, this method comprises: the operation data of periodically acquisition target device；Wherein operation data includes the first output power and its corresponding first efficiency conversion ratio；Operation data time series is added in operation data；When the number of the operation data in operation data time series reaches setting numerical value, the operation data to rank the first in operation data time series is deleted；Response external triggering, the training of Gaussian process is carried out using each operation data in operation data time series as training data, obtains hyper parameter corresponding with target device；Efficiency curve model corresponding with target device is determined using hyper parameter and each operation data.Technical solution provided by the invention can be adapted for various types of equipment, and the actual motion state for the reflection target device that obtained efficiency curve model can be more accurate.
Description
Technical field
The present invention relates to energy fields, more particularly to method, apparatus, medium and the electronics of building energy efficiency of equipment curve model
Equipment.
Background technique
The efficiency curve of equipment is able to reflect the relationship between the output power of equipment and energy efficiency conversion rate, and operation maintenance personnel can
The operating condition of relevant device is adjusted and is optimized according to the completion of efficiency curve.
Currently, for the large number of equipment in energy source station, it usually needs according to the device type of each equipment, using different
Parameterized model models the efficiency curve model of distinct device to obtain the corresponding efficiency curvilinear mold of each equipment
Type；Meanwhile the efficiency curvilinear mold that practical efficiency when equipment operation is influenced by runing time and environmental factor, therefore constructed
Type may not be able to accurately reflect the actual motion state of target device.
Summary of the invention
The present invention provides a kind of method, apparatus, readable medium and electronic equipments for constructing energy efficiency of equipment curve model, can
To be suitable for various types of equipment, and the reality for the reflection target device that obtained efficiency curve model can be more accurate
Border operating status.
In a first aspect, the present invention provides a kind of methods for constructing energy efficiency of equipment curve model, comprising:
The periodically operation data of acquisition target device；Wherein, the operation data includes the first output power and its right
The the first efficiency conversion ratio answered；
Operation data time series is added in the operation data；
When the number of the operation data in the operation data time series reaches setting numerical value, the fortune is deleted
The operation data to rank the first in row data time series；
Response external triggering carries out high using each operation data in the operation data time series as training data
The training of this process obtains hyper parameter corresponding with the target device；
Efficiency curvilinear mold corresponding with the target device is determined using the hyper parameter and each operation data
Type.
Preferably,
The external trigger includes: the second output power；
Efficiency corresponding with the target device is determined using the hyper parameter and each operation data described
After curve model, further comprise:
Determine that the first prediction efficiency corresponding with second output power converts using the efficiency curve model
Rate.
Preferably,
It is described to determine efficiency curve model corresponding with the target device, further comprise: it is bent to record the efficiency
The building time point of line model；
Further include:
When often collecting a current operating data, detection acquires current point in time and the institute of the current operating data
State whether the time difference between building time point is the default integer multiple for fixing duration, if it is, utilizing the described of acquisition
Whether current operating data detects the efficiency curve model accurate；
When detecting the efficiency curve model inaccuracy, the external trigger is sent.
Preferably,
Whether the current operating data using acquisition detects the efficiency curve model accurate, comprising:
It is corresponding using the third output power that the efficiency curve model is determined with the current operating data carries
Second prediction efficiency conversion ratio；
Detect the phase of third efficiency conversion ratio and the second prediction efficiency conversion ratio that the current operating data carries
Whether default relative error threshold value is less than to error, if it is, determining that the efficiency curve model is accurate, if it is not, then really
The fixed efficiency curve model inaccuracy.
Second aspect, the present invention provides a kind of devices for constructing energy efficiency of equipment curve model, comprising:
Data acquisition module, for periodically acquiring the operation data of target device；Wherein, the operation data includes the
One output power and its corresponding first efficiency conversion ratio；
Sequence forms module, for operation data time series to be added in the operation data；
Data removing module, for reaching setting when the number of the operation data in the operation data time series
When numerical value, the operation data to rank the first in the operation data time series is deleted；
Data training module is triggered for response external, with the operation number each in the operation data time series
According to the training for carrying out Gaussian process for training data, hyper parameter corresponding with the target device is obtained；
Model construction module, for being determined and the target device using the hyper parameter and each operation data
Corresponding efficiency curve model.
Preferably,
The external trigger includes: the second output power；
The model construction module, comprising: the first determination unit；Wherein,
First determination unit, it is opposite with second output power for being determined using the efficiency curve model
The the first prediction efficiency conversion ratio answered.
Preferably,
The model construction module further comprises: time recording unit；Wherein,
The time recording unit, for recording the building time point of the efficiency curve model；
Further include:
Accuracy detection module, for detecting when the sequence forms model and often collects an operation data
Acquire whether the time difference between the current point in time of the operation data and the building time point is default fixed duration
Integer multiple, if it is, whether detect the efficiency curve model using the operation data of acquisition accurate；When detecting
When the efficiency curve model inaccuracy, the external trigger is sent, triggers the data training module.
Preferably,
The accuracy detection module, comprising:
Second determination unit, for determining carry with the current operating data the using the efficiency curve model
The corresponding second prediction efficiency conversion ratio of three output powers；
Accuracy detection unit, the third efficiency conversion ratio and described second carried for detecting the current operating data
Whether the relative error of prediction efficiency conversion ratio is less than default relative error threshold value, if it is, determining the efficiency curvilinear mold
Type is accurate, if it is not, then determining the efficiency curve model inaccuracy.
The third aspect, the present invention provides a kind of readable mediums, including execute instruction, when the processor of electronic equipment executes
Described when executing instruction, the electronic equipment executes the method as described in any in first aspect.
Fourth aspect, the present invention provides a kind of electronic equipment, including processor and are stored with the storage executed instruction
Device, when executing instruction described in the processor executes memory storage, the processor is executed as in first aspect
Any method.
The present invention provides a kind of method, apparatus, readable medium and electronic equipments for constructing energy efficiency of equipment curve model, should
Method is by periodically acquiring the operation data of target device, wherein operation data includes the first output power and its corresponding
First efficiency conversion ratio；Collected operation data is added in operation data time series, when being charged first to operation data
Between the operation data of sequence come the front end of operation data time series；Set of operation data in operation data time series
Several limit values deletes operation data when the number of the operation data in operation data time series reaches setting numerical value
The operation data to rank the first in time series (deletes the fortune that operation data time series is added in each operation data earliest
Row data), in this way, when needing the efficiency curve model using target device at any one moment (such as at any one
When carving the efficiency conversion ratio for the efficiency curve realization prediction target device for needing to pass through building target device), the operation data time
Time corresponding to each operation data in sequence remains shorter time interval with current time always, i.e., each operation
Data can be more accurate reaction target device current time actual motion state；It is being needed correspondingly, receiving user
Using target device efficiency curve model when the external trigger that inputs, and with each operation number in operation data time series
According to the training for carrying out Gaussian process for training data, obtain hyper parameter corresponding with target device, then using hyper parameter and
Each operation data determines the corresponding efficiency curve model of target device, and realization responds received external trigger.It is high
This process is nonparametric model, and the type of equipment, i.e. this hair are then no longer dependent on when constructing the efficiency curve model of target device
The technical solution of bright offer can be suitable for various types of equipment, meanwhile, construct the efficiency curve model of target device
When used reaction target device that can be more accurate to inscribe when corresponding actual motion state operation data, building
Energy efficiency of equipment curve model then can be more accurate reflection equipment actual motion state.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without creative efforts, can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is a kind of flow diagram of the method for the building energy efficiency of equipment curve model provided in the embodiment of the present invention；
Fig. 2 is the process signal of the method for another building energy efficiency of equipment curve model provided in the embodiment of the present invention
Figure；
Fig. 3 is a kind of structural schematic diagram of the device of the building energy efficiency of equipment curve model provided in the embodiment of the present invention；
Fig. 4 is the structural representation of the device of another the building energy efficiency of equipment curve model provided in the embodiment of the present invention
Figure；
Fig. 5 is the structural representation of the device of another building energy efficiency of equipment curve model provided in the embodiment of the present invention
Figure；
Fig. 6 is the structural representation of the device of another the building energy efficiency of equipment curve model provided in the embodiment of the present invention
Figure；
Fig. 7 is the structural schematic diagram of a kind of electronic equipment provided in the embodiment of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment and accordingly
Technical solution of the present invention is clearly and completely described in attached drawing.Obviously, described embodiment is only a part of the invention
Embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making wound
Every other embodiment obtained under the premise of the property made labour, shall fall within the protection scope of the present invention.
As shown in Figure 1, the embodiment of the present invention provides a kind of method for constructing energy efficiency of equipment curve model, this method comprises:
Step 101, the operation data of target device is periodically acquired；Wherein, the operation data includes the first output work
Rate and its corresponding first efficiency conversion ratio；
Step 102, operation data time series is added in the operation data；
Step 103, it when the number of the operation data in the operation data time series reaches setting numerical value, deletes
Except the operation data to rank the first in the operation data time series；
Step 104, response external triggers, and is training number with each operation data in the operation data time series
According to the training for carrying out Gaussian process, hyper parameter corresponding with the target device is obtained；
Step 105, energy corresponding with the target device is determined using the hyper parameter and each operation data
Imitate curve model.
Embodiment as shown in Figure 1, the operation data that this method passes through periodically acquisition target device, wherein operation number
According to including the first output power and its corresponding first efficiency conversion ratio；When collected operation data is added to operation data
Between in sequence, the operation data for being charged first to operation data time series comes the front end of operation data time series；Setting fortune
The limit value of the number of operation data in row data time series, when the number of the operation data in operation data time series
When reaching setting numerical value, deletes the operation data to rank the first in operation data time series and (delete in each operation data
The operation data of operation data time series is added earliest), in this way, needing the efficiency using target device at any one moment
When curve model (for example need to realize prediction target device by the efficiency curve of building target device at any one moment
When efficiency conversion ratio), the time corresponding to each operation data in operation data time series keeps with current time always
Shorter time interval, i.e., the actual motion of each operation data can be more accurate reaction target device at current time
State；Correspondingly, receiving the external trigger that user inputs when needing the efficiency curve model using target device, and with operation
Each operation data in data time series is the training that training data carries out Gaussian process, is obtained corresponding with target device
Hyper parameter, then determine the corresponding efficiency curve model of target device using hyper parameter and each operation data, realize pair
Received external trigger is responded.Gaussian process is nonparametric model, when constructing the efficiency curve model of target device then not
Depending on the type of equipment again, i.e., technical solution provided in an embodiment of the present invention can be suitable for various types of equipment,
Meanwhile reaction target device that can be more accurate being used to inscribe when constructing the efficiency curve model of target device when corresponding
Actual motion state operation data, the reality for the reflection equipment that the energy efficiency of equipment curve model of building then can be more accurate
Operating status.
For example, the period of acquisition operation data is set as 1 hour, sets the operation number of operation data time series
It is 48 according to number, after collecting the 1st operation data, the 1st for adding it to operation data time series is i.e. the first,
After collecting the 2nd operation data, the 2nd of operation data time series is added it to, and so on, when collecting the 49th
When a operation data, then the operation data to rank the first can be deleted, that is, delete the 1st operation data of acquisition, do not delete the 1st
The 2nd the 2nd operation data is come when operation data can become operation data time series after deleting the 1st operation data
First place, and the 49th operation data is added to the 48th of operation data time series, when acquiring the 50th operation data
When, equally it is to delete the operation data to rank the first, that is, deletes the 2nd operation data of acquisition, and so on, with operation number
According to continuous acquisition, can constantly delete the operation data to rank the first, each operation data in operation data time series
It can constantly update, need to construct the efficiency curve model of the target device, operation data time sequence at any one current time
Each operation data in column then can be more accurate reaction target device actual motion state, correspondingly, when current
Carve building energy efficiency of equipment curve model, also can be more accurate reflection target device current time actual motion shape
State.
It should be noted that is, user is needing the efficiency curvilinear mold using target device when only there is external trigger
When type, method provided in an embodiment of the present invention can just be carried out by training data of each operation data in operation data time series
The efficiency curve model of Gaussian process trained and finally obtain target device is responded as to the external trigger received；
When external trigger is not present, only each operation data in operation data time series can be with newest collected operation
Data are updated, thus avoid when not needing using efficiency curve model still using each operation data as training data into
The training of row Gaussian process avoids the premature training that Gaussian process is carried out using each operation data as training data, so that
Finally obtained efficiency curve model can not accurately react practical fortune of the target device within corresponding moment or corresponding period
Row state.
When carrying out the training of Gaussian process as training data using each operation data in operation data time series, specifically may be used
Using by under each operation data in operation data time series the first output power and the first efficiency conversion ratio as
The training dataset D (x, y) of independent variable and dependent variable composition target device, x is corresponding with the first output power, y and the first efficiency
Conversion ratio is corresponding, establishes the functional relation between x and y according to Gaussian process, as follows:
Wherein, D characterizes operation data collection, the x of target deviceiCharacterize i-th first outputs in n the first output powers
Power, yiCharacterize i-th of first efficiency conversion ratios in n the first efficiency conversion ratios, εiCharacterize Gaussian noise, K characterizes covariance
Function, σ2Characterize variance.The determination of hyper parameter is related with covariance function K, can use Maximum Likelihood Estimation Method and conjugation ladder
Degree method determines optimal hyper parameter value, and is chosen to be the corresponding hyper parameter of target device, and hyper parameter can embody difference
The different characteristics of equipment, because Gaussian process is non-parametric model, then the efficiency curve model constructed using Gaussian process can
To be suitable for various types of equipment.
It should be appreciated by those skilled in the art building energy efficiency of equipment curve model provided in an embodiment of the present invention
Method can be adapted for separate unit target device, be readily applicable to more target devices.It in an embodiment of the invention, can be same
When periodical acquisition is carried out to the operation data of more target devices, form each operation data corresponding to each target device
Time series, and the limit value of the number for each operation data time series setting operation data, the numerical value can unite
One is set as identical numerical value, can also be set as different numerical value according to different target devices；For each target device
Operation data time series, when the number of the operation data in operation data time series reach setting numerical value when, delete fortune
The operation data to rank the first in row data time series；Response external triggering, detects the current goal that the external trigger is related to
Equipment carries out Gauss mistake by training data of each operation data in operation data time series corresponding to current target device
The training of journey obtains hyper parameter corresponding with target device, is determined and current mesh using hyper parameter and each operation data
The corresponding efficiency curve model of marking device.
In one embodiment of the invention, the external trigger includes: the second output power；The hyper parameter is utilized described
And after each operation data determines efficiency curve model corresponding with the target device, further comprise: utilizing
The efficiency curve model determines the first prediction efficiency conversion ratio corresponding with second output power.
In the above-described embodiments, external trigger can be the second output power, which can be any one
A control value corresponding with the first output power needs the efficiency curve model using target device at any one moment
When, then second output power can be obtained as external trigger, thus with each operation data in operation data time series
Efficiency curve model is trained and finally obtains for training data progress Gaussian process, realization rings received external trigger
It answers, after obtaining efficiency curve model, can use efficiency curve model and determine and second output power corresponding first
Predict efficiency conversion ratio；After having had been built up an efficiency curve model, the efficiency curve model of building can use to energy
Effect conversion ratio predicted, at this time obtain second output power, then can using building efficiency curve model determine with
The corresponding first prediction efficiency conversion ratio of second output power.
Need to illustrate when, external trigger may be control instruction, which can periodically exist, such as be arranged
Fixed duration will receive control instruction when time interval reaches the fixation duration, thus to operation data time series
In each operation data carry out the efficiency curve model trained and finally obtain target device of Gaussian process, realize to received
External trigger is responded.
In an embodiment of the invention, described to determine efficiency curve model corresponding with the target device, into one
Step includes: to record the building time point of the efficiency curve model；
Further include:
When often collecting a current operating data, detection acquires current point in time and the institute of the current operating data
State whether the time difference between building time point is the default integer multiple for fixing duration, if it is, utilizing the described of acquisition
Whether current operating data detects the efficiency curve model accurate；
When detecting the efficiency curve model inaccuracy, the external trigger is sent.
In the above-described embodiments, practical efficiency when running because of equipment is influenced by runing time and environmental factor, because
This needs the periodic detection efficiency bent after constructing an efficiency curve model, and when the efficiency curve model is used for a long time
Whether line model is accurate.When testing result shows that efficiency curve model is accurate, then the efficiency curve model is continued to use；Work as inspection
It surveys the result shows that then sending external trigger when the efficiency curve model inaccuracy, then with each in current operating data time series
A operation data is the training that training data carries out Gaussian process, and finally obtains new efficiency curve model, then records newly
The building time of efficiency curve model, realization respond received external trigger.It should be understood by those skilled in the art that
It is that preset fixed duration can be empirical value.
In an embodiment of the invention, the current operating data using acquisition detects the efficiency curvilinear mold
Whether type is accurate, comprising:
It is corresponding using the third output power that the efficiency curve model is determined with the current operating data carries
Second prediction efficiency conversion ratio；
Detect the phase of third efficiency conversion ratio and the second prediction efficiency conversion ratio that the current operating data carries
Whether default relative error threshold value is less than to error, if it is, determining that the efficiency curve model is accurate, if it is not, then really
The fixed efficiency curve model inaccuracy.
In the above-described embodiments, the time difference between the current point in time of current operating data and building time point is default
When the integer multiple of fixed duration, the third output power and its conversion of corresponding third efficiency that current operating data carries are determined
Rate, represents the actual motion value of target device, determines corresponding with third output power second using efficiency curve model
It predicts efficiency conversion ratio, represents predicted value, therefore miss using third efficiency conversion ratio and the second the opposite of prediction efficiency conversion ratio
Difference can determine whether efficiency curve model is accurate, it will be apparent to a skilled person that relative error threshold value can be
Empirical value.
It should be noted that whether the embodiment of the present invention also to can use multiple operation datas detection efficiency curve models quasi-
Really.For example, the time difference between the time point and building time point of the 120th operation data of acquisition is default fixed duration
Integer multiple, it is determined that the 100th operation data third that each operation data carries respectively into the 120th operation data
Output power and its corresponding third efficiency conversion ratio；For each third output power, determined using efficiency curve model
The second prediction efficiency conversion ratio corresponding with the third output power out, and detect the corresponding third energy of the third output power
It is default opposite whether the relative error of effect conversion ratio the second prediction efficiency conversion ratio corresponding with the third output power is less than
Error threshold, if it is, determining that efficiency curve model is accurate, if it is not, then determining efficiency curve model inaccuracy；Work as determination
The number of the accurate operation data of efficiency curve model and the number for the operation data for determining efficiency curve model inaccuracy are greater than
When preset ratio value (95%), then efficiency curve model is accurate, can continue to use the efficiency curve model.User can
Determines according to actual conditions for detecting the number of the whether accurate operation data of efficiency curve model.
Embodiment as shown in Figure 2 can specifically include following each step, comprising:
Step 201a periodically acquires the operation data of target device；Wherein, operation data include the first output power and
Its corresponding first efficiency conversion ratio.
It should be noted that step 201a, step 202, step 203, step 204 and step 205 are suitable for constructing for the first time
The efficiency curve model of target device.In the case where having been built up out the efficiency curve model of target device, periodically acquire
While the operation data of target device, it should also detect the current point in time of one operation data of acquisition and construct efficiency song
Whether the time difference of line model is the default integer multiple for fixing duration, that is, corresponds to the combination of step 201a and step 201b.
Step 202, operation data time series is added in operation data.
Step 203, when the number of the operation data in operation data time series reaches setting numerical value, operation number is deleted
According to the operation data to rank the first in time series.
The acquisition of operation data is periodically, when the number of the operation data in operation data time series reaches setting
After numerical value, that is, starts with newest collected data and eliminate the operation data that operation data time series is added earliest, that is, delete
Each operation data except the operation data to rank the first in operation data time series, therefore in operation data time series can
With the actual motion state of more accurate reaction target device.
Step 204, response external triggers, and carries out using each operation data in operation data time series as training data high
The training of this process obtains hyper parameter corresponding with target device.
Step 205, efficiency curve model corresponding with target device is determined using hyper parameter and each operation data, remember
Record the building time point of efficiency curve model.
When needing the efficiency curve model using target device at any one moment, then may exist external trigger, then
External trigger is received, is training and finally obtaining efficiency curvilinear mold for training data progress Gaussian process using each operation data
Type, realization respond received external trigger, because Gaussian process is nonparametric model, so that the structure of the efficiency curve model
Construction method can be adapted for different types of equipment；After an efficiency curve model has been determined, and to utilize the energy for a long time
When imitating curve model, then needs to record the building time of efficiency curve model, facilitate the subsequent efficiency curve model to building
Accuracy is detected.
Step 206, determine that the first prediction efficiency corresponding with the second output power converts using efficiency curve model
Rate.
Efficiency conversion ratio can be predicted using efficiency curve model, and according to obtained prediction result, operation maintenance personnel can root
It is predicted that result completion is adjusted and optimizes to the operating condition of target device.
Step 201b, when often collecting an operation data, the current point in time of detection acquisition operation data and building
Whether the time difference between time point is the default integer multiple for fixing duration；If so, thening follow the steps 207.
Step 207, corresponding using the third output power that efficiency curve model is determined with current operating data carries
Second prediction efficiency conversion ratio.
The current point in time of current operating data corresponding to third output power and the time difference between building time point
For the integer multiple of default fixed duration.
Step 208, the phase of the third efficiency conversion ratio that detection current operating data carries and the second prediction efficiency conversion ratio
Whether default relative error threshold value is less than to error, if it is, determining that the efficiency curve model is accurate, if it is not, then really
The fixed efficiency curve model inaccuracy, and execute step 209.
Third efficiency conversion ratio is the actual motion value of the periodically acquisition collected target device of operation data, and second is pre-
Surveying efficiency conversion ratio is the efficiency conversion ratio predicted using efficiency curve model, therefore can be to efficiency using the two data
The accuracy of curve model is detected.It will be apparent to a skilled person that relative error threshold value can be empirical value.
Step 209, external trigger is sent, and executes step 204.
When detecting efficiency curve model inaccuracy, external trigger can be sent, then proceed to execute step 204, i.e., with
Each operation data is the training that training data carries out Gaussian process in operation data time series, and finally obtains new efficiency
Curve model, and the building time of new efficiency curve model is recorded, it realizes and responds the external trigger.
Based on inventive concept same as mentioned above, as shown in figure 3, the embodiment of the invention provides a kind of building equipment
The device of efficiency curve model, comprising:
Data acquisition module 301, for periodically acquiring the operation data of target device；Wherein, the operation data packet
Include the first output power and its corresponding first efficiency conversion ratio；
Sequence forms model 302, for operation data time series to be added in the operation data；
Data removing module 303, for reaching when the number of the operation data in the operation data time series
When setting numerical value, the operation data to rank the first in the operation data time series is deleted；
Data training module 304 is triggered for response external, with each operation in the operation data time series
Data are the training that training data carries out Gaussian process, obtain hyper parameter corresponding with the target device；
Model construction module 305, for being determined and the target using the hyper parameter and each operation data
The corresponding efficiency curve model of equipment.
As shown in figure 4, in an embodiment of the invention, the external trigger includes: the second output power；
The model construction module 305, comprising: the first determination unit 3051；Wherein,
First determination unit 3051, for being determined and second output power using the efficiency curve model
Corresponding first prediction efficiency conversion ratio.
As shown in figure 5, in an embodiment of the invention, the model construction module 305 further comprises: time note
Record unit 3052；Wherein, the time recording unit 3052, for recording the building time point of the efficiency curve model；
Further include:
Accuracy detection module 306, for examining when the sequence forms model and often collects an operation data
Whether the time difference surveyed between the current point in time for acquiring the operation data and the building time point is default fixed duration
Integer multiple, if it is, whether detect the efficiency curve model using the operation data of acquisition accurate；Work as detection
When to the efficiency curve model inaccuracy, the external trigger is sent, the data training module 304 is triggered.
As shown in fig. 6, in an embodiment of the invention, the accuracy detection module 306, comprising:
Second determination unit 3061, for determining to carry with the current operating data using the efficiency curve model
Third output power it is corresponding second prediction efficiency conversion ratio；
Accuracy detection unit 3062, for detect third efficiency conversion ratio that the current operating data carries with it is described
Whether the relative error of the second prediction efficiency conversion ratio is less than default relative error threshold value, if it is, determining that the efficiency is bent
Line model is accurate, if it is not, then determining the efficiency curve model inaccuracy.
For convenience of description, it describes to be divided into various units when apparatus above embodiment with function or module describes respectively,
The function of each unit or module can be realized in the same or multiple software and or hardware in carrying out the present invention.
Fig. 7 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.In hardware view, the electronic equipment
Including processor, optionally further comprising internal bus, network interface, memory.Wherein, memory may include memory, such as
High-speed random access memory (Random-Access Memory, RAM), it is also possible to further include nonvolatile memory (non-
Volatile memory), for example, at least 1 magnetic disk storage etc..Certainly, which is also possible that other business institutes
The hardware needed.
Processor, network interface and memory can be connected with each other by internal bus, which can be ISA
(Industry StandardArchitecture, industry standard architecture) bus, PCI (Peripheral Component
Interconnect, Peripheral Component Interconnect standard) bus or EISA (Extended Industry
StandardArchitecture, expanding the industrial standard structure) bus etc..It is total that the bus can be divided into address bus, data
Line, control bus etc..Only to be indicated with a four-headed arrow in Fig. 7, it is not intended that an only bus or one convenient for indicating
The bus of seed type.
Memory is executed instruction for storing.Specifically, the computer program that can be performed is executed instruction.Memory
It may include memory and nonvolatile memory, and execute instruction to processor offer and data.
In a kind of mode in the cards, processor reads corresponding execute instruction to interior from nonvolatile memory
It is then run in depositing, can also obtain from other equipment and execute instruction accordingly, to form building equipment energy on logic level
Imitate the device of curve model.What processor execution memory was stored executes instruction, to execute instruction realization originally by what is executed
A kind of method of the building energy efficiency of equipment curve model provided in invention any embodiment.
The device of the above-mentioned building energy efficiency of equipment curve model provided such as Fig. 3, Fig. 4 of the present invention, Fig. 5, embodiment illustrated in fig. 6
It can be applied in processor, or realized by processor.Processor may be a kind of IC chip, the place with signal
Reason ability.During realization, each step of the above method can by the integrated logic circuit of the hardware in processor or
The instruction of software form is completed.Above-mentioned processor can be general processor, including central processing unit (Central
Processing Unit, CPU), network processing unit (Network Processor, NP) etc.；It can also be Digital Signal Processing
Device (Digital Signal Processor, DSP), specific integrated circuit (Application Specific Integrated
Circuit, ASIC), field programmable gate array (Field-Programmable GateArray, FPGA) or other can
Programmed logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute present invention implementation
Disclosed each method, step and logic diagram in example.General processor can be microprocessor or the processor can also be with
It is any conventional processor etc..
The step of method in conjunction with disclosed in the embodiment of the present invention, can be embodied directly in hardware decoding processor and execute
At, or in decoding processor hardware and software module combination execute completion.Software module can be located at random access memory,
This fields such as flash memory, read-only memory, programmable read only memory or electrically erasable programmable memory, register maturation
In storage medium.The storage medium is located at memory, and processor reads the information in memory, completes above-mentioned side in conjunction with its hardware
The step of method.
The embodiment of the present invention also proposed a kind of readable medium, which, which is stored with, executes instruction, storage
It executes instruction when being executed by the processor of electronic equipment, the electronic equipment can be made to execute and provided in any embodiment of the present invention
The method for constructing energy efficiency of equipment curve model, and be specifically used for executing method as shown in Figure 1 and Figure 2.
Electronic equipment described in foregoing individual embodiments can be computer.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method or computer program product.
Therefore, the form that complete hardware embodiment, complete software embodiment or software and hardware combine can be used in the present invention.
Various embodiments are described in a progressive manner in the present invention, same and similar part between each embodiment
It may refer to each other, each embodiment focuses on the differences from other embodiments.Implement especially for device
For example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part illustrates.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
The above description is only an embodiment of the present invention, is not intended to restrict the invention.For those skilled in the art
For, the invention may be variously modified and varied.All any modifications made within the spirit and principles of the present invention are equal
Replacement, improvement etc., should be included within scope of the presently claimed invention.
Claims (10)
1. a kind of method for constructing energy efficiency of equipment curve model characterized by comprising
The periodically operation data of acquisition target device；Wherein, the operation data includes the first output power and its corresponding
First efficiency conversion ratio；
Operation data time series is added in the operation data；
When the number of the operation data in the operation data time series reaches setting numerical value, the operation number is deleted
According to the operation data to rank the first in time series；
Response external triggering carries out Gauss mistake by training data of each operation data in the operation data time series
The training of journey obtains hyper parameter corresponding with the target device；
Efficiency curve model corresponding with the target device is determined using the hyper parameter and each operation data.
2. the method according to claim 1, wherein
The external trigger includes: the second output power；
Efficiency curve corresponding with the target device is determined using the hyper parameter and each operation data described
After model, further comprise:
The first prediction efficiency conversion ratio corresponding with second output power is determined using the efficiency curve model.
3. method according to claim 1 or 2, which is characterized in that
It is described to determine efficiency curve model corresponding with the target device, further comprise: recording the efficiency curvilinear mold
The building time point of type；
Further include:
When often collecting a current operating data, detection acquires the current point in time and the structure of the current operating data
Whether the time difference built between time point is the integer multiple of default fixed duration, if it is, utilizing the described current of acquisition
Whether operation data detects the efficiency curve model accurate；
When detecting the efficiency curve model inaccuracy, the external trigger is sent.
4. according to the method described in claim 3, it is characterized in that,
Whether the current operating data using acquisition detects the efficiency curve model accurate, comprising:
Utilize the third output power corresponding that the efficiency curve model is determined with the current operating data carries
Two prediction efficiency conversion ratios；
It detects the third efficiency conversion ratio and described second that the current operating data carries and predicts that the opposite of efficiency conversion ratio is missed
Whether difference is less than default relative error threshold value, if it is, determining that the efficiency curve model is accurate, if it is not, then determining institute
State efficiency curve model inaccuracy.
5. a kind of device for constructing energy efficiency of equipment curve model characterized by comprising
Data acquisition module, for periodically acquiring the operation data of target device；Wherein, the operation data includes first defeated
Power and its corresponding first efficiency conversion ratio out；
Sequence forms module, for operation data time series to be added in the operation data；
Data removing module, for reaching setting numerical value when the number of the operation data in the operation data time series
When, delete the operation data to rank the first in the operation data time series；
Data training module triggers for response external, is with each operation data in the operation data time series
Training data carries out the training of Gaussian process, obtains hyper parameter corresponding with the target device；
Model construction module, it is corresponding with the target device for being determined using the hyper parameter and each operation data
Efficiency curve model.
6. device according to claim 5, which is characterized in that
The external trigger includes: the second output power；
The model construction module, comprising: the first determination unit；Wherein,
First determination unit, it is corresponding with second output power for being determined using the efficiency curve model
First prediction efficiency conversion ratio.
7. device according to claim 5 or 6, which is characterized in that
The model construction module further comprises: time recording unit；Wherein,
The time recording unit, for recording the building time point of the efficiency curve model；
Further include:
Accuracy detection module, for when the sequence forms model and often collects an operation data, detection to be acquired
Whether the time difference between the current point in time of the operation data and the building time point is the default integer for fixing duration
Multiple, if it is, whether detect the efficiency curve model using the operation data of acquisition accurate；It is described when detecting
When efficiency curve model inaccuracy, the external trigger is sent, triggers the data training module.
8. the method according to the description of claim 7 is characterized in that
The accuracy detection module, comprising:
Second determination unit, for determining that the third carried with the current operating data is defeated using the efficiency curve model
The corresponding second prediction efficiency conversion ratio of power out；
Accuracy detection unit, for detecting the third efficiency conversion ratio and second prediction that the current operating data carries
Whether the relative error of efficiency conversion ratio is less than default relative error threshold value, if it is, determining that the efficiency curve model is quasi-
Really, if it is not, then determining the efficiency curve model inaccuracy.
9. a kind of readable medium, including execute instruction, when executing instruction described in the processor of electronic equipment executes, the electronics
Equipment executes the method as described in any in Claims 1-4.
10. a kind of electronic equipment including processor and is stored with the memory executed instruction, described in processor execution
When executing instruction described in memory storage, the processor executes the method as described in any in Claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910042715.6A CN109886544A (en) | 2019-01-17 | 2019-01-17 | Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910042715.6A CN109886544A (en) | 2019-01-17 | 2019-01-17 | Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109886544A true CN109886544A (en) | 2019-06-14 |
Family
ID=66926133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910042715.6A Pending CN109886544A (en) | 2019-01-17 | 2019-01-17 | Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109886544A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111625754A (en) * | 2020-05-12 | 2020-09-04 | 新智数字科技有限公司 | Method and device for calculating boiler energy efficiency, terminal equipment and storage medium |
| CN114756525A (en) * | 2022-03-14 | 2022-07-15 | 广州城建职业学院 | Sequence data construction method, device, equipment and computer readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102056182A (en) * | 2010-12-13 | 2011-05-11 | 哈尔滨工业大学 | Method for predicting mobile traffic based on LS-SVM |
| CN103011356A (en) * | 2012-08-15 | 2013-04-03 | 重庆水务集团股份有限公司 | Method for controlling automatic chemical dosing of high-turbidity water system |
| CN105890109A (en) * | 2016-03-31 | 2016-08-24 | 华南理工大学 | Online long-acting performance detection and optimized running method for room air conditioner |
| CN106056127A (en) * | 2016-04-07 | 2016-10-26 | 江南大学 | GPR (gaussian process regression) online soft measurement method with model updating |
| CN106765932A (en) * | 2016-12-14 | 2017-05-31 | 深圳达实智能股份有限公司 | The Energy Efficiency Ratio Forecasting Methodology and device of central air conditioner system refrigeration host computer |
| CN106951611A (en) * | 2017-03-07 | 2017-07-14 | 哈尔滨工业大学 | A kind of severe cold area energy-saving design in construction optimization method based on user's behavior |
| CN107632995A (en) * | 2017-03-13 | 2018-01-26 | 平安科技（深圳）有限公司 | The method and model training control system of Random Forest model training |
-
2019
- 2019-01-17 CN CN201910042715.6A patent/CN109886544A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102056182A (en) * | 2010-12-13 | 2011-05-11 | 哈尔滨工业大学 | Method for predicting mobile traffic based on LS-SVM |
| CN103011356A (en) * | 2012-08-15 | 2013-04-03 | 重庆水务集团股份有限公司 | Method for controlling automatic chemical dosing of high-turbidity water system |
| CN105890109A (en) * | 2016-03-31 | 2016-08-24 | 华南理工大学 | Online long-acting performance detection and optimized running method for room air conditioner |
| CN106056127A (en) * | 2016-04-07 | 2016-10-26 | 江南大学 | GPR (gaussian process regression) online soft measurement method with model updating |
| CN106765932A (en) * | 2016-12-14 | 2017-05-31 | 深圳达实智能股份有限公司 | The Energy Efficiency Ratio Forecasting Methodology and device of central air conditioner system refrigeration host computer |
| CN106951611A (en) * | 2017-03-07 | 2017-07-14 | 哈尔滨工业大学 | A kind of severe cold area energy-saving design in construction optimization method based on user's behavior |
| CN107632995A (en) * | 2017-03-13 | 2018-01-26 | 平安科技（深圳）有限公司 | The method and model training control system of Random Forest model training |
Non-Patent Citations (2)
| Title |
|---|
| 郭鹏，姜漫利，李航涛: "基于运行数据和高斯过程回归的风电机组发电性能分析与监测", 《电力自动化设备》 * |
| 魏芬芬: "灰色预测模型在血糖预测中的研究", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111625754A (en) * | 2020-05-12 | 2020-09-04 | 新智数字科技有限公司 | Method and device for calculating boiler energy efficiency, terminal equipment and storage medium |
| CN111625754B (en) * | 2020-05-12 | 2023-05-02 | 新智数字科技有限公司 | Method, device, terminal equipment and storage medium for calculating energy efficiency of boiler |
| CN114756525A (en) * | 2022-03-14 | 2022-07-15 | 广州城建职业学院 | Sequence data construction method, device, equipment and computer readable storage medium |
| CN114756525B (en) * | 2022-03-14 | 2023-11-28 | 广州城建职业学院 | Sequence data construction method, device, equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871483A (en) | A kind of determination method and device of recommendation information | |
| CN104090912A (en) | Information pushing method and device | |
| CN109886544A (en) | Construct method, apparatus, medium and the electronic equipment of energy efficiency of equipment curve model | |
| CN106227844A (en) | The method of a kind of application recommendation and terminal | |
| CN111461168A (en) | Training sample expansion method and device, electronic equipment and storage medium | |
| CN110009359A (en) | Training method, update method and the device of unsupervised risk prevention system model | |
| US20230349721A1 (en) | Step counting method, step counting device, and computer readable storage medium | |
| CN110263824A (en) | The training method of model, calculates equipment and computer readable storage medium at device | |
| CN115424053B (en) | Small sample image recognition method, device, equipment and storage medium | |
| CN110490595A (en) | A kind of risk control method and device | |
| CN110010224A (en) | User movement data processing method, device, wearable device and storage medium | |
| CN111639470A (en) | Simulation test method and system for processor cooperative chip and related components | |
| CN104641370A (en) | Time sequential data accumulation device | |
| CN108562867A (en) | A kind of fingerprint positioning method and device based on cluster | |
| CN108108558A (en) | A kind of method and system based on degree of covering statistical appraisal accidental validation quality | |
| CN112348188A (en) | Model generation method and device, electronic device and storage medium | |
| CN114693011A (en) | Policy matching method, device, equipment and medium | |
| CN106821336A (en) | Sleep monitor method and system | |
| CN105760646A (en) | Method and device for activity classification | |
| CN110309947A (en) | Complete vehicle logistics order forecast method and device, logistics system and computer-readable medium | |
| CN112882907B (en) | User state determination method and device based on log data | |
| CN111182354B (en) | Video scoring recommendation method, device and equipment and computer readable storage medium | |
| CN109829575A (en) | Processing method, device, readable medium and the electronic equipment of multi-energy data prediction | |
| CN110070858A (en) | A kind of civilization term based reminding method, device and mobile device | |
| CN109684700A (en) | Efficiency curve-fitting method, device, medium and the electronic equipment of gas fired-boiler |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |
|
| RJ01 | Rejection of invention patent application after publication |