CN109707658A - Method for determination of performance parameter, device and the electronic equipment of blower - Google Patents
Method for determination of performance parameter, device and the electronic equipment of blower Download PDFInfo
- Publication number
- CN109707658A CN109707658A CN201910151323.3A CN201910151323A CN109707658A CN 109707658 A CN109707658 A CN 109707658A CN 201910151323 A CN201910151323 A CN 201910151323A CN 109707658 A CN109707658 A CN 109707658A
- Authority
- CN
- China
- Prior art keywords
- parameter
- design parameter
- blower
- sample data
- performance parameter
- 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
Landscapes
- Structures Of Non-Positive Displacement Pumps (AREA)
Abstract
Present description provides a kind of method for determination of performance parameter of blower, device and electronic equipments.Wherein, this method comprises: obtaining the design parameter of blower;Using preset processing model using the design parameter as mode input, the performance parameter of blower corresponding with the design parameter is determined, wherein preset processing model is preparatory trained neural network model.In this specification embodiment, by advance with include design parameter and corresponding performance parameter sample data, carry out the model training of neural network, to establish preset processing model, the performance parameter for recycling the preset processing model that complicated numerical simulation operation is replaced to determine blower, to solve the low technical problem for the treatment of effeciency present in existing method, reach the performance parameter for efficiently and accurately determining blower corresponding with design parameter, user is facilitated to be adjusted the technical effect of modification to design parameter in time.
Description
Technical field
This specification belongs to the technical field more particularly to a kind of determination side of the performance parameter of blower that machine component designs
Method, device and electronic equipment.
Background technique
When carrying out machine component design, user generally requires first to require that (such as energy consumption requires, power is wanted according to specific
Ask) etc. determine corresponding expected performance objective, and then determine initial design parameter (such as some portion in machine component
The radius etc. of some component in the height of part, machine component);Numerical simulation fortune is carried out by corresponding numerical simulation software again
It calculates, obtains corresponding performance parameter;And then identified design parameter before can be carried out according to obtained performance parameter
It optimizes and revises, finally obtains the design parameter met the requirements.
But numerical simulation calculating process is typically more complicated, can expend a large amount of time and resource.For example, in benefit
When being carried out with CFD (one kind is based on fluid dynamic simulation software) about centrifugal fan-spiral case component Flow Field Calculation, by
It needs to solve the equation group of large amount of complex, solving complexity is relatively according to corresponding principle of hydrodynamics in data simulation
Height, the operand being related to is also relatively large, causes simulation calculating process that can occupy a large amount of computing resource, while emulating fortune
The time of calculation also can be relatively long.For example, it may be possible to which the time of a simulation calculating just needs more than ten hour.Lead to user in this way
Based on existing method when carrying out centrifugal fan-spiral case component design, needing to wait longer time just can be obtained accordingly
Performance parameter affects the working efficiency of user.It can be seen that when it is implemented, often there is the low technology for the treatment of effeciency in existing method
Problem.
In view of the above-mentioned problems, currently no effective solution has been proposed.
Summary of the invention
This specification is designed to provide the method for determination of performance parameter, device and electronic equipment of a kind of blower, with solution
Existing method of the having determined technical problem low there are treatment effeciency.
Method for determination of performance parameter, device and the electronic equipment for a kind of blower that this specification provides are realized in
:
A kind of method for determination of performance parameter of blower, comprising:
Obtain the design parameter of blower;
Using preset processing model using the design parameter as mode input, determine corresponding with the design parameter
Blower performance parameter, wherein the preset processing model is preparatory trained neural network model.
In one embodiment, the design parameter includes at least one of: spiral case maximum height, impeller height, snail
Tongue radius, spiral case depth, volute tongue gap, impeller are apart from wind-guiding ring gap, impeller in spiral case gap, impeller outer diameter, impeller
Diameter, blade inflow angle, blade go out to flow angle, leaf chord length, vane thickness, maximum camber.
In one embodiment, the performance parameter includes at least one of: air quantity, wind pressure, effective power, shaft work
Rate, efficiency.
In one embodiment, the preset processing model is established in the following way:
Obtain sample data, wherein the sample data includes: the sample design parameter of multiple groups blower, and with it is described
The corresponding performance parameter of sample design parameter;
Using the sample data, neural network model is trained, establishes the preset processing model.
In one embodiment, obtaining sample data includes:
Sample design parameter, and performance parameter corresponding with the sample design parameter are obtained by experiment;
And/or sample design parameter, and performance corresponding with sample design parameter ginseng are obtained by numerical simulation
Number.
In one embodiment, after obtaining sample data, the method also includes:
Interpolation fitting is carried out to the sample data, with exptended sample data.
In one embodiment, using the sample data, neural network model is trained, is established described preset
Handle model, comprising:
The sample data is divided into first sample data and the second sample data;
Using the first sample data, neural network model is trained, establishes the first processing model;
According to the network parameter of the first processing model, correlation of the design parameter with performance parameter is determined;
According to the correlation of the design parameter and performance parameter, correlation is less than preset in rejecting the first processing model
The design parameter of relevance threshold obtains second processing model;And reject in second sample data correlation be less than it is default
Relevance threshold design parameter, obtain third sample data;
Using the third sample data, the second processing model is trained, establishes the preset processing mould
Type.
In one embodiment, it, using the design parameter as mode input, is determined using preset processing model
After the performance parameter of blower corresponding with the design parameter, the method also includes:
According to the performance parameter of blower corresponding with the design parameter, the design parameter of the blower is adjusted.
A kind of determining device of the performance parameter of blower, comprising:
Module is obtained, for obtaining the design parameter of blower;
Determining module, for, using the design parameter as mode input, being determined and institute using preset processing model
State the performance parameter of the corresponding blower of design parameter, wherein the preset processing model is preparatory trained neural network
Model.
Including processor and for the memory of storage processor executable instruction, the processor executes described instruction
Shi Shixian: the design parameter of blower is obtained;Using preset processing model using the design parameter as mode input, determine
The performance parameter of blower corresponding with the design parameter, wherein the preset processing model is preparatory trained nerve
Network model.
A kind of computer readable storage medium is stored thereon with computer instruction, and described instruction is performed realization: obtaining
The design parameter of blower;Using preset processing model using the design parameter as mode input, determine and the design
The performance parameter of the corresponding blower of parameter, wherein the preset processing model is preparatory trained neural network model.
Method for determination of performance parameter, device and the electronic equipment for a kind of blower that this specification provides, due to by pre-
First obtain and using include design parameter and performance parameter sample data carry out neural network model training, establish preset
Processing model, recycle the preset processing model that complicated numerical simulation operation is replaced to determine corresponding performance parameter,
To solve the low technical problem for the treatment of effeciency present in existing method, reaching reduces computation complexity and accounts for resource
With efficiently and accurately determining performance parameter corresponding with design parameter;The waiting time for reducing user, facilitate user timely
Technical effect design parameter is adjusted, modified.
Detailed description of the invention
In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
The some embodiments recorded in this specification, for those of ordinary skill in the art, in not making the creative labor property
Under the premise of, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of embodiment of the process of the method for determination of performance parameter for the blower that this specification embodiment provides
Schematic diagram;
Fig. 2 is in a Sample Scenario, using the determination side of the performance parameter of the blower of this specification embodiment offer
A kind of schematic diagram of embodiment of method;
Fig. 3 is in a Sample Scenario, using the determination side of the performance parameter of the blower of this specification embodiment offer
A kind of schematic diagram of embodiment of method;
Fig. 4 is in a Sample Scenario, using the determination side of the performance parameter of the blower of this specification embodiment offer
A kind of schematic diagram of embodiment of method;
Fig. 5 is in a Sample Scenario, using the determination side of the performance parameter of the blower of this specification embodiment offer
A kind of schematic diagram of embodiment of method;
Fig. 6 is a kind of embodiment of the structure of the determining device of the performance parameter for the blower that this specification embodiment provides
Schematic diagram;
Fig. 7 is a kind of schematic diagram of embodiment of the structure for the electronic equipment that this specification embodiment provides.
Specific embodiment
In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation
Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment be only this specification a part of the embodiment, instead of all the embodiments.The embodiment of base in this manual,
Every other embodiment obtained by those of ordinary skill in the art without making creative efforts, all should belong to
The range of this specification protection.
In view of being based on existing method, every time when calculating the performance parameter of blower, require to lead to using corresponding software
Solution is crossed largely based on the equation group of correlation theory, carries out numerical simulation operation.The data being related in above-mentioned treatment process
Treating capacity is relatively large, causes the processing time relatively long, treatment effeciency is relatively low, so that user needs to wait long time
Corresponding performance parameter could be obtained, user's design efficiency is influenced.
For the basic reason for generating the above problem, it includes multiple groups design ginseng that the application consideration, which can be obtained first and be utilized,
The numeric data of several and corresponding performance parameter carries out the training of neural network model as sample data;So as to quasi-
It builds the vertical complicated equation group not depended on out based on correlation theory jointly, but numerical value phase of the design parameter with performance parameter can be symbolized
The model of fit of closing property is as preset processing model;The numerical simulation for recycling the preset processing model replacement originally complicated
Operation determines corresponding performance parameter, so as to be effectively reduced the complexity of calculating, improves treatment effeciency, solves existing side
The low technical problem for the treatment of effeciency present in method.
Based on above-mentioned thinking thinking, the embodiment of the present application provides a kind of method for determination of performance parameter of blower.Specifically
Please refer to a kind of embodiment of the process of the method for determination of performance parameter for the blower that this specification embodiment shown in FIG. 1 provides
Schematic diagram.The method for determination of performance parameter of blower provided by the embodiments of the present application, when it is implemented, may include in following
Hold.
S11: the design parameter of blower is obtained.
In the present embodiment, above-mentioned blower specifically can be understood as user (such as Production Engineer or product designer etc.)
Want the product pair for reaching expected performance objective (such as meet some or several particular characteristic indexs) by design optimization
As.For example, engineer is according to customer requirement, it is desirable to which the air quantity that design optimization goes out reaches the blower of some index parameter, so that it may
It is considered a kind of blower.Certainly when it is implemented, may be incorporated into other machines group according to concrete application scene and customer demand
Part carries out the determination of performance parameter as above-mentioned blower is replaced.In this regard, this specification is not construed as limiting.
In the present embodiment, above-mentioned design parameter specifically can be understood as customer-furnished for limiting the attribute of blower
The parameter of feature.For example, specifically can be the dimensional parameters of blower or the material parameter of blower etc..User can pass through
It is arranged, adjusts above-mentioned design parameter blower is made to progressively reach expected performance objective.
It should be noted that for different types of blower and different expected performance objectives, above-mentioned design parameter
It can specifically include different types of supplemental characteristic.When it is implemented, can as the case may be, in conjunction with the concrete type of blower
With expected performance objective to be achieved, used design parameter is determined.In this regard, this specification is not construed as limiting.
Lower mask body will be explained in detail the determination side for how applying the performance parameter of blower provided by the embodiment of the present application
Method efficiently designs the centrifugal fan-spiral case component for meeting expected performance objective.It, can be with for other kinds of blower
Refering to the associated description execution about fan assembly, this specification is not repeated.
In the present embodiment, in the case where blower is centrifugal fan-spiral case component in fan assembly, refering to Fig. 2
Shown in a Sample Scenario, using this specification embodiment provide blower method for determination of performance parameter one kind
The schematic diagram of embodiment, the design parameter that may influence fan performance can specifically include at least one of: spiral case is maximum high
Degree (can be denoted as H1), impeller height (can be denoted as H2), volute tongue radius (can be denoted as R1), spiral case depth (can be denoted as
L1), volute tongue gap (can be denoted as L2), impeller (can be with apart from spiral case gap apart from wind-guiding ring gap (can be denoted as B1), impeller
Be denoted as B2), impeller outer diameter (D1 can be denoted as), profile ID (D2 can be denoted as), blade inflow angle (α can be denoted as), blade
Stream angle (can be denoted as β), leaf chord length (can be denoted as CL), vane thickness (can be denoted as THc), maximum camber (can be remembered out
For h) etc..When it is implemented, can using any one or any number of combinations in above-mentioned cited parameter as
The design parameter that user provides.Certainly, as the case may be and process demand, may be incorporated into except above-mentioned cited parameter class
Design parameter of the outer other parameters of type as blower.In this regard, this specification is not construed as limiting.
In the present embodiment, when it is implemented, can provide a Data Input Interface for user, user can be by this
Data Input Interface inputs corresponding design parameter, so as to acquire above-mentioned design parameter.
In the present embodiment, when it is implemented, performance parameter in order to preferably determine blower, above-mentioned sets obtaining
While counting parameter, fixed input parameter can also be obtained.
Wherein, above-mentioned fixed input parameter specifically can be understood as one kind values constant during design optimization and not need
Adjustment, but also will affect the supplemental characteristic of fan performance.Specifically, above-mentioned fixed data parameter may include the revolving speed of blower
Peripheral speed (V2 can be denoted as) of (n can be denoted as), impeller periphery etc..Certainly, above-mentioned cited fixed input parameter
It is that one kind schematically illustrates.When it is implemented, other kinds of supplemental characteristic can also be introduced as the case may be as above-mentioned
Fixed input parameter.In this regard, this specification is not construed as limiting.
S13: it using preset processing model using the design parameter as mode input, determines and the design parameter
The performance parameter of corresponding blower, wherein the preset processing model is preparatory trained neural network model.
In the present embodiment, above-mentioned performance parameter specifically can be understood as the parameter number of the performance state for characterizing blower
According to may determine that whether the blower based on corresponding design parameter can reach expected performance indicator by above-mentioned performance parameter.
Wherein, the influence that will receive design parameter of above-mentioned performance parameter, numerically has certain relevance with design parameter.Tool
Body, above-mentioned performance parameter can be the parameter of characterization power, be also possible to characterize the parameter etc. of efficiency.According to blower type
Difference and the above-mentioned performance parameters of the difference supplemental characteristic that is included of expected performance objective can not also be identical.Specifically
, suitable supplemental characteristic can be determined as above-mentioned performance parameter according to mesh object type and expected performance objective.
In the present embodiment, in the case where blower is centrifugal fan-spiral case component in fan assembly, refering to table 1
Shown in performance parameter list, the performance parameter of blower may include at least one of: air quantity (can be denoted as Q), wind pressure (can
To be denoted as H), effective power (N can be denoted asy), shaft power (N can be denoted as), efficiency (η can be denoted as) etc..
1 performance parameter list of table
Parameter of structure design | Meaning | Parameter of structure design | Meaning |
H1 | Spiral case maximum width | D1 | Impeller outer diameter |
H2 | Impeller height | D2 | Profile ID |
R1 | Volute tongue radius | α | Blade inflow angle |
L1 | Spiral case depth | β | Blade goes out to flow angle |
L2 | Volute tongue gap | CL | Leaf chord length |
B1 | Impeller is apart from wind-guiding ring gap | Thc | Vane thickness |
B2 | Impeller is apart from spiral case gap | h | Maximum camber |
When it is implemented, can using any one or any number of combinations in above-mentioned cited parameter as
The performance parameter of blower.Certainly, as the case may be and process demand, it may be incorporated into addition to above-mentioned cited parameter other
Performance parameter of the parameter as blower.In this regard, this specification is not construed as limiting.
In the present embodiment, above-mentioned preset processing model specifically can be understood as a kind of first passing through to sample data in advance
Learning training is carried out, what is be fitted can characterize the neural network model of the numerical relation of design parameter and performance parameter.
It should be noted that above-mentioned preset processing model is different from the numerical model in convenient value emulation.Wherein, often
The equation group that numerical model in rule numerical simulation is often based on correlation theory (such as fluid dynamic theory etc.) is established
Numerical solution model, this class model can be related to a large amount of more complicated solving equations in operation solution procedure, and operation is asked
The data volume being related in solution preocess is relatively large, and the complexity of calculating is relatively high, also opposite to the occupancy of computing resource
It is larger, cause the time-consuming of solution procedure also relatively long.And preset processing model used in the embodiment of the present application is not
It is that the equation group based on correlation theory is established, but from another angle, by existing to design parameter and performance parameter
Correlation numerically carries out learning training, a kind of prediction model established.Therefore, it is determined by preset processing model
When the performance parameter of blower corresponding to design parameter, relative to using the numerical model in conventional numerical simulation, do not need
Solve the equation group of large amount of complex, it is only necessary to according to the phase numerically of trained obtained design parameter and performance parameter
Mutual relation, so that it may the performance parameter of blower corresponding to design parameter is predicted, so as to be effectively reduced model running
Complexity in the process reduces calculation amount, improves the efficiency of model treatment.
In the present embodiment, it is above-mentioned using preset processing model using the design parameter as mode input, determine
The performance parameter of blower corresponding with the design parameter, when it is implemented, may include: that calling is trained preset
Handle model;Using acquired design parameter as mode input, it is input in preset processing model;It runs described preset
Model is handled, obtains model output, and the model is exported into the performance parameter as blower corresponding with the design parameter.
To efficiently and accurately obtain the performance parameter of blower.
In the embodiment of the present application, compared to existing method, due to including design parameter by obtaining and utilizing in advance
The model training that neural network is carried out with the sample data of performance parameter, establishes preset processing model, recycles this preset
Processing model replaces complicated numerical simulation operation to determine corresponding performance parameter, to solve present in existing method
The low technical problem for the treatment of effeciency, reaching reduces computation complexity and the occupancy to resource, efficiently and accurately determines and designs
The corresponding performance parameter of parameter;The waiting time for reducing user, user is facilitated targetedly to be adjusted to design parameter in time
Whole, modification technical effect.
In one embodiment, above-mentioned solving equations but energy based on relative theory complexity are not depended in order to obtain
The preset processing model of enough design parameters and performance parameter numerical lineardependence for accurately characterizing blowing machine, when it is implemented,
By model training, above-mentioned preset processing model can be established in the following way:
S1: obtain sample data, wherein the sample data includes: the sample design parameter of multiple groups blower, and with institute
State the corresponding performance parameter of sample design parameter.
In the present embodiment, above-mentioned sample data can specifically include multi-group data, wherein in above-mentioned multi-group data
Each group of data separately includes the design parameter of blower, and the performance parameter of blower corresponding with the design parameter.It needs
Bright, design parameter and performance parameter included in sample data are thought with design parameter provided by user and user
The performance parameter to be determined respectively corresponds.Above-mentioned sample data is subsequent to can be used for model training.
In the present embodiment, in the case where blower is centrifugal fan-spiral case component in fan assembly, if user
Wanting determining design parameter includes: spiral case maximum height, impeller height, volute tongue radius, spiral case depth, volute tongue gap, impeller
Go out to flow angle, blade apart from spiral case gap, impeller outer diameter, profile ID, blade inflow angle, blade apart from wind-guiding ring gap, impeller
Chord length, vane thickness, maximum camber 14 different types of supplemental characteristics in total, then every group of data in above-mentioned sample data
It needs to include identical 14 supplemental characteristics.Similar, if user is of interest, wants the determining property for being used to characterize blower
The performance parameter of energy state includes: air quantity, wind pressure, effective power, shaft power, efficiency 5 different types of supplemental characteristics in total,
Then every group of data in above-mentioned sample data are also required to include above-mentioned identical 5 supplemental characteristics.
Specifically, (wherein, i-th group of data in multi-group data in above-mentioned sample data can be expressed as following form
Design parameter specifically can be used to be indicated with lower target x, and performance parameter specifically can be used to be identified with lower target y):
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)i
Wherein, it is k parameter data that xk, which can specifically be expressed as number in design parameter, with above-mentioned 14 design parameters pair
It answers, wherein k is the integer more than or equal to 1, and less than or equal to 14;Y1, y2, y3, y4, y5 can specifically respectively indicate above-mentioned 5
Performance parameter;I can specifically be expressed as the number of the data group in sample data, and i is and to be less than or equal to sample more than or equal to 1
The integer of the group number of data group included in data.
In one embodiment, above-mentioned acquisition sample data, when it is implemented, may include the following contents: passing through experiment
Obtain sample design parameter, and performance parameter corresponding with the sample design parameter;And/or it is obtained by numerical simulation
Sample design parameter, and performance parameter corresponding with the sample design parameter.
In the present embodiment, sample design parameter is obtained above by experiment, and corresponding with the sample design parameter
Performance parameter, when it is implemented, corresponding blower can be made according to corresponding design parameter;The blower is tested again
Test obtains result data according to experiment test, determines corresponding performance parameter, and then can be and right by the design parameter
The performance parameter answered is as one group of sample data.
In the present embodiment, above by numerical simulation obtain sample design parameter, and with the sample design parameter
Corresponding performance parameter, when it is implemented, can be set accordingly by using relevant numerical simulation software (such as CFD) basis
It counts parameter and carries out numerical simulation operation, obtain operation result, and according to operation result, determine corresponding performance parameter, in turn
It can be by the design parameter, with corresponding performance parameter as another group of sample data.
In the present embodiment, history cases of design can also be obtained by enquiry of historical data, it will be in history cases of design
The design parameter for being included and corresponding performance parameter are as one group of sample data.Wherein, above-mentioned history cases of design specifically may be used
To be that blower that user filters out from historical data and performance parameter are close with current expected performance objective to be achieved
Like the higher cases of design of degree.Further, it is also possible to obtain design parameter and corresponding performance according to the engineering experience of user
Parameter is as one group of sample data etc..
Certainly, it should be noted that the mode of above-mentioned cited acquisition sample data is that one kind schematically illustrates.Tool
When body is implemented, it can be used alone above-mentioned a certain mode and obtain sample data, above-mentioned various ways can also be combined and be obtained
Sample data can also introduce other suitable modes and obtain sample data according to specific circumstances.In this regard, this specification is not
It limits.
In one embodiment, available more in order to enable the preset processing model of subsequent foundation is more accurate
Sample data carry out subsequent model training.Specific implementation can be obtained more by summary a variety of acquisition modes
Sample data.More sample datas can also be obtained by carrying out data extending to the sample data obtained.
In one embodiment, after obtaining sample data, the method is when it is implemented, can also include in following
Hold: interpolation fitting being carried out to the sample data, with exptended sample data.
In the present embodiment, specifically new sample can be generated by the way of difference fitting according to existing sample data
Notebook data, to realize the expansion of sample data.
Specifically, for example, can first construct the lagrange polynomial of a multidimensional as composed formula;Further according to existing
Sample data in a certain group of sample data design parameter, equably construct new design parameter;Utilize above-mentioned fitting
Formula calculates corresponding new performance parameter according to new design parameter;And then it can be and corresponding by new design parameter
The sample data that newly expands as one group of new performance parameter.For example, being carried out through the above way according to p group sample data
Sample data expands the sample data of available p+q group, can indicate are as follows:
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)1
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)2
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)3
……
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)p
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)p+1
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)p+2
……
(x1/x2/x3/x4/...xk.../x14,y1/y2/y3/y4/y5)p+q
Wherein, in above-mentioned data under be designated as 1 to p be script sample data, under to be designated as p+1 to p+q be to expand
The new sample data arrived.The expansion that can rapidly realize sample data through the above way, obtains sample more abundant
Data.
Certainly, it should be noted that the extended mode of above-mentioned cited sample data is that one kind schematically illustrates.Tool
When body is implemented, as the case may be, the expansion of sample data can also be carried out using other suitable extended modes.In this regard, this
Specification is not construed as limiting.
S2: the sample data is utilized, neural network model is trained, the preset processing model is established.
In the present embodiment, after obtaining above-mentioned sample data, it can use above-mentioned sample data to neural network model
Carry out learning training, with establish the corresponding design parameter that can characterize blower and performance parameter numerical lineardependence it is preset
Handle model.
In the present embodiment, by taking blower is centrifugal fan-spiral case component in fan assembly as an example.It can be refering to Fig. 3
Shown in a Sample Scenario, using this specification embodiment provide blower method for determination of performance parameter one kind
The schematic diagram of embodiment, above-mentioned neural network may include: input layer, output layer and hidden layer.Wherein, include in input layer
14 input parameters, 14 supplemental characteristics that corresponding design parameter is included.It include 5 output parameters in output layer, it is corresponding
5 supplemental characteristics that performance parameter is included.Hidden layer can specifically include multiple hidden layers (such as L-1 layers), also can wrap
Include a hidden layer, wherein can (wherein, each node may be considered a nerve with multiple nodes in each hidden layer
Member).When it is implemented, the number of plies (can be denoted as layers) of hidden layer can be set as the case may be and required precision, with
And the number of nodes (nodes) in each hidden layer.
For any one hidden layer in above-mentioned neural network, for example, for the hidden layer of kth layer, the hidden layer packet
The number of nodes contained is Nk, wherein uk(i) input of i-th of node in the hidden layer, w are indicatedk(i, j) indicates kth -1
Weight of j-th of node to i-th of node in kth layer in layer, ak(i) output of i-th of node in kth layer, θ are indicatedk
(i) biasing of i-th of node in kth layer is indicated.Wherein, input layer can be understood as the 0th layer, the number of nodes N for being included0=
14, output layer can be understood as L layers, the number of nodes N for being includedL=5.
Corresponding, the numerical relation of the node in above-mentioned neural network between each layer output and input can be expressed as
Following form:
a1(i)=f (u1(i)),1≤i≤N1,
a2(i)=f (u2(i)),1≤i≤N2,
......
aL(i)=f (uL(i)),1≤i≤NL
Wherein, above-mentioned f (x) can specifically indicate a kind of activation primitive.In the present embodiment, in order to avoid in model training
Occur the problems such as gradient explosion or gradient hour in the process, while being also the complexity for taking into account model, simplified model training, tool
When body is implemented, Relu function can be used as activation primitive.Wherein, above-mentioned Relu function can be expressed as following form:
Certainly, it should be noted that above-mentioned cited activation primitive be one kind schematically have a talk about it is bright.Specific implementation
When, it as the case may be, can also be using other suitable functions as activation primitive.In this regard, this specification is not construed as limiting.
In the present embodiment, using the sample data, neural network model is trained, it can be understood as by defeated
Enter sample data, determine include each hidden layer interior joint weight and biasing specific value network parameter.Specifically
, it can first determine the weight of each layer of interior joint and initial value (the i.e. initial network parameter, or initially weigh square of biasing
Battle array);Repeatedly input different sample datas again, and according to the sample data of input constantly to the weight of each layer of interior joint and
Biasing is repeatedly corrected, so that the weight of each layer of interior joint and biasing level off to accurate numerical value, finally obtains standard
The higher network parameter of exactness, training have obtained satisfactory preset processing model.
It in the present embodiment, when it is implemented, can be by using back-propagation algorithm (BP, Back Propagation)
Constantly the weight of each layer of interior joint and biasing are repeatedly corrected according to the sample data of input.
When it is implemented, can be modified in the following way to weight and biasing:
Wherein,It can be specifically expressed as error function, p can specifically indicate sample
Sample data group number in data.
Model training is carried out in input sample data each time, is declined according to gradient, corresponding weight iterative formula can be with
It is expressed as following form:
wl (p)(i, j)=wl (p-1)(i,j)+ηδl (p)al (p)(j), l=1 ..., L
Wherein, wl (p)(i, j) can specifically be expressed as in l-1 layers of hidden layer j-th of node to i-th of section in l layers
Weight when the pth time iteration of point.
1≤i≤nl, 1≤j≤nl-1
Wherein, δL (p)(i) it can specifically be expressed as error function and local derviation is asked to weight.
Specifically, δL (p)(i) it can be determined according to following formula:
Wherein, η can specifically be expressed as learning rate, and value range updates between 0 to 1 for controlling algorithm every time
Weight step-length;t(p)(i) when can specifically be expressed as pth time iteration, the desired output of neural network is (i.e. according to sample number
According to the known value provided);aL (p)(i) it is defeated that the reality that neural network generates the input value of training data can be specifically expressed as
Out.
In one embodiment, in order to be further simplified preset processing model, the complexity of model is reduced, is improved subsequent
Model running efficiency, it is above-mentioned to utilize the sample data, neural network model is trained, the preset processing mould is established
Type, refering to shown in 4 in a Sample Scenario, the determination side of the performance parameter of the blower provided using this specification embodiment
A kind of schematic diagram of embodiment of method, when it is implemented, further including the following contents:
S1: the sample data is divided into first sample data and the second sample data;
S2: the first sample data are utilized, neural network model is trained, the first processing model is established;
S3: according to the network parameter of the first processing model, correlation of the design parameter with performance parameter is determined;
S4: it according to the correlation of the design parameter and performance parameter, rejects correlation in the first processing model and is less than in advance
If relevance threshold design parameter, obtain second processing model;And it rejects correlation in second sample data and is less than
The design parameter of preset relevance threshold obtains third sample data;
S5: the third sample data is utilized, the second processing model is trained, the preset processing is established
Model.
In the present embodiment, the first processing model of above-mentioned foundation is not final, the higher model of accuracy, still
By the weight and/or biasing in the network parameter in the first processing model, the design parameter inputted and output can analyze out
The correlation of performance parameter numerically.For example, each layer of section in network parameter in analysis the first processing model can be passed through
The weighted value size of point finds that some possible data are relatively large to the value effect of performance parameter in design parameter, then related
Property is also relatively large;And some data may value effect to performance parameter it is relatively small, then correlation is also relatively small.
For simplified model, the complexity of model is reduced, treatment effeciency is further increased, it can be according to above-mentioned first processing
The network parameter of model goes out to characterize the correlation of each design parameter with performance parameter respectively;Again with preset relevance threshold
It makes comparisons, the design parameter that correlation is less than preset relevance threshold is determined as to influence relatively small set to performance parameter
Count parameter.That is, the change of the type design parameter tends not to generate more apparent influence to performance parameter.Therefore, in mould
When type training, the influence of the type design parameter, simplified model training process can be ignored.Specifically, can be from the first processing
The design parameter of the above-mentioned type is eliminated in model, i.e. correlation obtains letter less than the design parameter of preset relevance threshold
The second processing model changed;Meanwhile the design parameter of the above-mentioned type can be eliminated from the second sample data, obtain data
Measure relatively less third sample data;And then it can be only using the third sample data after above-mentioned rejecting, to the after rejecting
Two processing models are trained, so as to more be efficiently obtained satisfactory preset processing model.Wherein, above-mentioned pre-
If the value of relevance threshold can flexibly be set according to required precision.In this regard, this specification is not construed as limiting.
In the present embodiment, when it is implemented, the relatively small number of sample data of by a relatively simple, data volume can be drawn
It is divided into first sample data;The relatively large number of sample data of relatively complicated, data volume is divided into the second sample data.This
Outside, when it is implemented, can not also be divided to sample data, for example, it is also possible to first with sample data to neural network
Model is trained, and establishes the first processing model;According to the network parameter of the first processing model, design parameter and property are determined
The correlation of energy parameter;According to the correlation of the design parameter and performance parameter, rejects correlation in the first processing model and be less than
The design parameter of preset relevance threshold obtains second processing model;And it rejects correlation in the sample data and is less than in advance
If relevance threshold design parameter, the sample data after being rejected;Using the sample data after the rejecting, to described
Second processing model is trained, and establishes the preset processing model.
In one embodiment, as shown in fig.4, using preset processing model using the design parameter as model
It inputs, after the performance parameter for determining blower corresponding with the design parameter, the method is when it is implemented, can also include
The following contents: according to the performance parameter of blower corresponding with the design parameter, the design parameter of the blower is adjusted.
In the present embodiment, the design parameter of user's offer is being acquired, and is determining corresponding performance parameter, also
It can be compared according to the expected performance indicator of above-mentioned performance parameter and user, determine and currently available join with the design
Whether the performance parameter of the corresponding blower of number meets the expected performance indicator of user;Determining currently available performance parameter not
The case where meeting expected performance indicator, targetedly above-mentioned design parameter can be adjusted according to above-mentioned performance parameter
It is whole, so that expected performance indicator can be more nearly by obtaining performance parameter based on design parameter adjusted.Tune is obtained again
Design parameter after whole, and determine the performance parameter adjusted of blower corresponding with design parameter adjusted.It will adjust again
Performance parameter and expected performance indicator after whole are compared, and are determined based on after the obtained adjustment of design parameter adjusted
Performance parameter whether meet expected performance objective.If performance parameter adjusted meets expected performance indicator, can
The design parameter that design parameter adjusted determines final blower will be corresponded to, the optimization design to blower is completed, is obtained
Relatively preferable design scheme.If performance parameter adjusted is still unsatisfactory for expected performance indicator, can exchange again
Design parameter after whole is targetedly adjusted again, to obtain better design parameter.
Specifically, by taking the centrifugal fan in fan assembly-spiral case component as an example, it can be refering to shown in fig. 5 at one
In Sample Scenario, using a kind of signal of embodiment of the method for determination of performance parameter of the blower of this specification embodiment offer
Figure after repeatedly being adjusted through the above way, can be obtained quickly accurately, the optimization for meeting estimated performance index is set
Count parameter.The optimization design flow field figure obtained by comparing the design parameter based on optimization design, and set based on the original being not optimised
The design parameter of meter obtains known to the long figure of intrinsic Liu: based on the optimal design parameter, the velocity field of impeller outlet is obviously more
Add uniformly, dissipate and reduce, advantageously reduce shaft power, promotes air quantity, and then improve fan efficiency, meet customer need.
Further, list can also be compared refering to performance parameter shown in table 2.The optimization design based on determined by this method
Parameter achieves in particular that air quantity promotes 21.1%, shaft power declines 4.7%, and uses wind relative to former design parameter in performance
Amount characterizes ratio divided by the efficiency that shaft power obtains characterization fan efficiency, which has also been obviously improved 27.2%.It can be seen that this method
Improving treatment effeciency simultaneously, it may have higher accuracy rate.
2 performance parameter of table compares list
In one embodiment, user is utilizing above-mentioned preset processing model, has obtained wind corresponding with design parameter
After the performance parameter of machine, when it is implemented, sample data that can also be new as one group using this group of design parameter and performance parameter,
Preset processing model is trained, so that preset processing model can carry out model according to the data handled in real time
Training is updated in real time, so as to automatically real-time be modified to the network parameter in model, is further increased preset
Handle the model accuracy of model.
In one embodiment, in order to further increase user experience, treatment effeciency, the design of above-mentioned acquisition blower are improved
Parameter, when it is implemented, can also include: the scan data that acquisition includes the design parameter of multiple blowers;Using preset
Model is handled using the scan data as mode input, is determined and the performance parameter of multiple corresponding blowers.
In the present embodiment, the scanning range of above-mentioned scan data can be arranged as the case may be and swept with demand by user
Retouch step-length, so it is subsequent multiple design parameters can be determined according to scanning range and scanning step, and multiple set according to above-mentioned
Parameter is counted, corresponding performance parameter is calculated by preset processing model.
For example, user may be interested in the parameter of some in design parameter A, it is desirable to further determine that A pairs of lower parameter
The influence size of performance parameter.At this moment, user can keep the other parameters in design parameter constant, and the first of parameter A is only arranged
Initial value and stop value are as scanning range, while the spacing accuracy that parameter A is arranged obtains above-mentioned scanning number as scanning step
According to;Again using above-mentioned scan data as mode input, it is input in preset processing model, obtains multiple and different parameter A numerical value
Corresponding performance parameter.And then user can determine parameter A to property by comparing the performance parameter of corresponding different parameters A
The influence size of energy parameter.
In one embodiment, the blower can specifically include: fan assembly etc..Certainly, above-mentioned cited blower
Component is that one kind schematically illustrates.When it is implemented, may be incorporated into as the case may be with process demand other kinds of
Machine component is as above-mentioned blower.In this regard, this specification is not construed as limiting.
In one embodiment, the design parameter can specifically include at least one of: spiral case maximum height, impeller
Highly, volute tongue radius, spiral case depth, volute tongue gap, impeller apart from wind-guiding ring gap, impeller apart from spiral case gap, impeller outer diameter,
Profile ID, blade inflow angle, blade go out to flow angle, leaf chord length, vane thickness, maximum camber etc..Certainly, above-mentioned cited
Design parameter is that one kind schematically illustrates.When it is implemented, as the case may be and processing requirement, may be incorporated into other classes
The parameter of type is as design parameter.In this regard, this specification is not construed as limiting.
In one embodiment, the performance parameter can specifically include at least one of: air quantity, wind pressure, Effective power
Rate, shaft power, efficiency etc..Certainly, above-mentioned cited performance parameter is that one kind schematically illustrates.When it is implemented, according to
Concrete condition and processing requirement may be incorporated into other kinds of parameter as performance parameter.In this regard, this specification does not limit
It is fixed.
It can be seen from the above description that the method for determination of performance parameter of blower provided by the embodiments of the present application, leads to
It crosses and obtains in advance and utilize the model training for the sample data progress neural network for including design parameter and performance parameter, foundation
Preset processing model recycles the preset processing model to replace complicated numerical simulation operation to determine corresponding performance ginseng
Number, to solve the low technical problem for the treatment of effeciency present in existing method, reaching reduces computation complexity and to resource
Occupancy, efficiently and accurately determine performance parameter corresponding with design parameter;The waiting time for reducing user, facilitate user
Technical effect design parameter is targetedly adjusted in time, modified;Also by when establishing preset processing model,
First according to the correlation of design parameter and performance parameter, weed out the first processing model established based on first sample data and
The relatively poor design parameter of correlation in second sample data recycles the second sample data after rejecting to the after rejecting
One processing model is trained, and is reduced design parameter in need of consideration, is reduced the complexity of processing model, further increase
Treatment effeciency.
Based on the same inventive concept, a kind of determining device of the performance parameter of blower is additionally provided in the embodiment of the present application,
As described in the following examples.The performance parameter of the principle and blower that are solved the problems, such as due to the determining device of the performance parameter of blower
Determination method it is similar, therefore the implementation of the determining device of the performance parameter of blower may refer to the determination of the performance parameter of blower
The implementation of method, overlaps will not be repeated.Used below, predetermined function may be implemented in term " unit " or " module "
Software and/or hardware combination.Although device described in following embodiment is preferably realized with software, hardware,
Or the realization of the combination of software and hardware is also that may and be contemplated.This specification embodiment shown in fig. 6 is please referred to mention
A kind of schematic diagram of embodiment of the structure of the determining device of the performance parameter of the blower of confession is wind provided by the embodiments of the present application
A kind of composite structural diagram of the determining device of the performance parameter of machine, the device can specifically include: obtaining module 61 and determine mould
Block 62 is below specifically described the structure.
Module 61 is obtained, specifically can be used for obtaining the design parameter of blower;
Determining module 62 specifically can be used for using preset processing model using the design parameter as mode input,
Determine the performance parameter of blower corresponding with the design parameter, wherein the preset processing model is to train in advance
Neural network model.
In one embodiment, the blower can specifically include: fan assembly etc..Certainly, above-mentioned cited blower
Component is that one kind schematically illustrates.When it is implemented, may be incorporated into as the case may be with process demand other kinds of
Machine component is as above-mentioned blower.In this regard, this specification is not construed as limiting.
In one embodiment, the design parameter can specifically include at least one of: spiral case maximum height, impeller
Highly, volute tongue radius, spiral case depth, volute tongue gap, impeller apart from wind-guiding ring gap, impeller apart from spiral case gap, impeller outer diameter,
Profile ID, blade inflow angle, blade go out to flow angle, leaf chord length, vane thickness, maximum camber etc..Certainly, above-mentioned cited
Design parameter is that one kind schematically illustrates.When it is implemented, as the case may be and processing requirement, may be incorporated into other classes
The parameter of type is as design parameter.In this regard, this specification is not construed as limiting.
In one embodiment, the performance parameter can specifically include at least one of: air quantity, wind pressure, Effective power
Rate, shaft power, efficiency etc..Certainly, above-mentioned cited performance parameter is that one kind schematically illustrates.When it is implemented, according to
Concrete condition and processing requirement may be incorporated into other kinds of parameter as performance parameter.In this regard, this specification does not limit
It is fixed.
In one embodiment, above-mentioned apparatus specifically can also include training module, pre- for being established by model training
If processing model.Wherein, above-mentioned training module can specifically include following structural unit:
Acquiring unit specifically can be used for obtaining sample data, wherein the sample data includes: the sample of multiple groups blower
The design parameter, and performance parameter corresponding with the sample design parameter;
Training unit specifically can be used for being trained neural network model, described in foundation using the sample data
Preset processing model.
In one embodiment, above-mentioned acquiring unit specifically can obtain sample data according to following procedure: pass through experiment
Obtain sample design parameter, and performance parameter corresponding with the sample design parameter;And/or it is obtained by numerical simulation
Sample design parameter, and performance parameter corresponding with the sample design parameter.
In one embodiment, above-mentioned training module can also include expansion unit, specifically can be used in acquiring unit
After obtaining sample data, interpolation fitting is carried out to the sample data, with exptended sample data.
In one embodiment, in order to utilize the sample data, neural network model is trained, institute is established
Preset processing model is stated, above-mentioned training unit can specifically include following structural sub-units:
Subelement is divided, specifically can be used for the sample data being divided into first sample data and the second sample number
According to;
First training subelement specifically can be used for instructing neural network model using the first sample data
Practice, establishes the first processing model;
Determine subelement, specifically can be used for according to it is described first processing model network parameter, determine design parameter with
The correlation of performance parameter;
Subelement is rejected, specifically can be used for the correlation according to the design parameter and performance parameter, is rejected at first
The design parameter that correlation in model is less than preset relevance threshold is managed, second processing model is obtained;And reject described second
Correlation is less than the design parameter of preset relevance threshold in sample data, obtains third sample data;
Second training subelement, specifically can be used for using the third sample data, to the second processing model into
Row training, establishes the preset processing model.
In one embodiment, described device can also include specifically adjustment module, specifically can be used in determining module
62 using preset processing model using the design parameter as mode input, determine blower corresponding with the design parameter
Performance parameter after, according to the performance parameter of blower corresponding with the design parameter, adjust the design parameter of the blower.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying 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 explanation.
It should be noted that system, device, module or unit that above embodiment illustrates, it specifically can be by computer
Chip or entity are realized, or are realized by the product with certain function.For convenience of description, in the present specification, it retouches
It is divided into various units when stating apparatus above with function to describe respectively.It certainly, when implementing the application can be the function of each unit
It realizes in the same or multiple software and or hardware.
In addition, in the present specification, such as adjective as first and second can be only used for an element or move
Make to distinguish with another element or movement, without requiring or implying any actual this relationship or sequence.Permit in environment
Perhaps in the case where, it should not be interpreted as limited to one in only element, component or step referring to element or component or step (s)
It is a, and can be the one or more etc. in element, component or step.
It can be seen from the above description that the determining device of the performance parameter of blower provided by the embodiments of the present application, leads to
Cross advance with training module obtain and utilize include design parameter and performance parameter sample data carry out neural network
Model training establishes preset processing model, then replaces complicated numerical value using the preset processing model by determining module
Simulation calculating determines corresponding performance parameter, thus solve the low technical problem for the treatment of effeciency present in existing method,
Reaching reduces computation complexity and the occupancy to resource, efficiently and accurately determines performance parameter corresponding with design parameter;Subtract
The waiting time of few user, the technical effect for facilitating user targetedly to be adjusted, modified to design parameter in time;It is also logical
Training module is crossed when establishing preset processing model, first according to the correlation of design parameter and performance parameter, weeds out and is based on
The relatively poor design parameter of correlation in the first processing model that first sample data are established and the second sample data, then benefit
The first processing model after rejecting is trained with the second sample data after rejecting, reduces design ginseng in need of consideration
Number reduces the complexity of processing model, further improves treatment effeciency.
The embodiment of the present application also provides a kind of electronic equipment, can specifically be implemented refering to shown in Fig. 7 based on the application
The electronic equipment composed structure schematic diagram of the method for determination of performance parameter for the blower that example provides, the electronic equipment specifically can be with
Including input equipment 71, processor 72, memory 73.Wherein, the input equipment 71 specifically can be used for the design ginseng of blower
Number.The processor 72 specifically can be used for using preset processing model determining using the design parameter as mode input
The performance parameter of blower corresponding with the design parameter out, wherein the preset processing model is preparatory trained mind
Through network model.The memory 73 specifically can be used for the instruction repertorie that storage processor 72 is based on.
In the present embodiment, the input equipment, which specifically can be, carries out information exchange between user and computer system
One of main device.The input equipment may include keyboard, mouse, camera, scanner, light pen, writing input board, language
Sound input unit etc.;Input equipment is used to initial data be input in computer with the programs for handling these numbers.The input
Equipment, which can also obtain, receives the data that other modules, unit, equipment transmit.The processor can be by any appropriate
Mode is realized.For example, processor can take such as microprocessor or processor and storage that can be executed by (micro-) processor
Computer readable program code (such as software or firmware) computer-readable medium, logic gate, switch, specific integrated circuit
(Application Specific Integrated Circuit, ASIC), programmable logic controller (PLC) and insertion microcontroller
Form etc..The storage implement body can be in modern information technologies for protecting stored memory device.The storage
Device may include many levels, in digital display circuit, as long as can save binary data can be memory;In integrated electricity
The circuit with store function of Lu Zhong, a not no physical form are also memory, such as RAM, FIFO;In systems, have
There is the storage equipment of physical form to be also memory, such as memory bar, TF card.
In the present embodiment, the function and effect of electronic equipment specific implementation, can compare with other embodiment
It explains, details are not described herein.
The embodiment of the present application also provides a kind of computer storage medium of method for determination of performance parameter based on blower,
The computer storage medium is stored with computer program instructions, is performed realization in the computer program instructions: obtaining
The design parameter of blower;Using preset processing model using the design parameter as mode input, determine and the design
The performance parameter of the corresponding blower of parameter, wherein the preset processing model is preparatory trained neural network model.
In the present embodiment, above-mentioned storage medium includes but is not limited to random access memory (Random Access
Memory, RAM), read-only memory (Read-Only Memory, ROM), caching (Cache), hard disk (Hard Disk
Drive, HDD) or storage card (Memory Card).The memory can be used for storing computer program instructions.Network is logical
Letter unit can be according to standard setting as defined in communication protocol, for carrying out the interface of network connection communication.
In the present embodiment, the function and effect of the program instruction specific implementation of computer storage medium storage, can
To compare explanation with other embodiment, details are not described herein.
Although mentioning different specific embodiments in teachings herein, the application is not limited to be industry
Situation described in standard or embodiment etc., certain professional standards or the implementation base described using customized mode or embodiment
On plinth embodiment modified slightly also may be implemented above-described embodiment it is identical, it is equivalent or it is close or deformation after it is anticipated that
Implementation result.It, still can be with using these modifications or the embodiment of deformed data acquisition, processing, output, judgment mode etc.
Belong within the scope of the optional embodiment of the application.
Although this application provides the method operating procedure as described in embodiment or flow chart, based on conventional or noninvasive
The means for the property made may include more or less operating procedure.The step of enumerating in embodiment sequence is only numerous steps
One of execution sequence mode, does not represent and unique executes sequence.It, can when device or client production in practice executes
To execute or parallel execute (such as at parallel processor or multithreading according to embodiment or method shown in the drawings sequence
The environment of reason, even distributed data processing environment).The terms "include", "comprise" or its any other variant are intended to contain
Lid non-exclusive inclusion, so that process, method, product or equipment including a series of elements are not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, product or equipment
Intrinsic element.In the absence of more restrictions, be not precluded include the process, method of the element, product or
There is also other identical or equivalent elements in person's equipment.
Device that above-described embodiment illustrates or module etc. can specifically realize by computer chip or entity, or by having
There is the product of certain function to realize.For convenience of description, it is divided into various modules when description apparatus above with function to retouch respectively
It states.Certainly, the function of each module can be realized in the same or multiple software and or hardware when implementing the application,
The module for realizing same function can be realized by the combination of multiple submodule etc..Installation practice described above is only
Schematically, for example, the division of the module, only a kind of logical function partition, can there is other draw in actual implementation
The mode of dividing, such as multiple module or components can be combined or can be integrated into another system, or some features can be ignored,
Or it does not execute.
It is also known in the art that other than realizing controller in a manner of pure computer readable program code, it is complete
Entirely can by by method and step carry out programming in logic come so that controller with logic gate, switch, specific integrated circuit, programmable
Logic controller realizes identical function with the form for being embedded in microcontroller etc..Therefore this controller is considered one kind
Hardware component, and the structure that the device for realizing various functions that its inside includes can also be considered as in hardware component.Or
Person even, can will be considered as realizing the device of various functions either the software module of implementation method can be hardware again
Structure in component.
The application can describe in the general context of computer-executable instructions executed by a computer, such as program
Module.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, group
Part, data structure, class etc..The application can also be practiced in a distributed computing environment, in these distributed computing environments,
By executing task by the connected remote processing devices of communication network.In a distributed computing environment, program module can
To be located in the local and remote computer storage media including storage equipment.
As seen through the above description of the embodiments, those skilled in the art can be understood that the application can
It realizes by means of software and necessary general hardware platform.Based on this understanding, the technical solution essence of the application
On in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software product
It can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment
(can be personal computer, mobile terminal, server or the network equipment etc.) executes each embodiment of the application or implementation
Method described in certain parts of example.
Each embodiment in this specification is described in a progressive manner, the same or similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.The application can be used for crowd
In mostly general or special purpose computing system environments or configuration.Such as: personal computer, server computer, handheld device or
Portable device, laptop device, multicomputer system, microprocessor-based system, set top box, programmable electronics set
Standby, network PC, minicomputer, mainframe computer, distributed computing environment including any of the above system or equipment etc..
Although depicting the application by embodiment, it will be appreciated by the skilled addressee that the application there are many deformation and
Variation is without departing from spirit herein, it is desirable to which appended embodiment includes these deformations and changes without departing from the application.
Claims (10)
1. a kind of method for determination of performance parameter of blower characterized by comprising
Obtain the design parameter of blower;
Using preset processing model using the design parameter as mode input, wind corresponding with the design parameter is determined
The performance parameter of machine, wherein the preset processing model is preparatory trained neural network model.
2. the method according to claim 1, wherein the design parameter includes at least one of: spiral case is most
Big height, impeller height, volute tongue radius, spiral case depth, volute tongue gap, impeller are apart from wind-guiding ring gap, impeller between spiral case
Gap, impeller outer diameter, profile ID, blade inflow angle, blade go out to flow angle, leaf chord length, vane thickness, maximum camber.
3. the method according to claim 1, wherein the performance parameter includes at least one of: air quantity, wind
Pressure, effective power, shaft power, efficiency.
4. the method according to claim 1, wherein the preset processing model is established in the following way:
Obtain sample data, wherein the sample data includes: the sample design parameter of multiple groups blower, and with the sample
The corresponding performance parameter of design parameter;
Using the sample data, neural network model is trained, establishes the preset processing model.
5. according to the method described in claim 4, it is characterized in that, obtaining sample data, comprising:
Sample design parameter, and performance parameter corresponding with the sample design parameter are obtained by experiment;
And/or sample design parameter, and performance parameter corresponding with the sample design parameter are obtained by numerical simulation.
6. according to the method described in claim 5, it is characterized in that, after obtaining sample data, the method also includes:
Interpolation fitting, exptended sample data are carried out to the sample data.
7. according to the method described in claim 6, it is characterized in that, being carried out using the sample data to neural network model
Training, establishes the preset processing model, comprising:
The sample data is divided into first sample data and the second sample data;
Using the first sample data, neural network model is trained, establishes the first processing model;
According to the network parameter of the first processing model, correlation of the design parameter with performance parameter is determined;
According to the correlation of the design parameter and performance parameter, rejects correlation in the first processing model and be less than preset correlation
The design parameter of property threshold value, obtains second processing model;And it rejects correlation in second sample data and is less than preset phase
The design parameter of closing property threshold value, obtains third sample data;
Using the third sample data, the second processing model is trained, establishes the preset processing model.
8. the method according to claim 1, wherein being made using preset processing model with the design parameter
For mode input, after the performance parameter for determining blower corresponding with the design parameter, the method also includes:
According to the performance parameter of blower corresponding with the design parameter, the design parameter of the blower is adjusted.
9. a kind of determining device of the performance parameter of blower characterized by comprising
Module is obtained, for obtaining the design parameter of blower;
Determining module, for, using the design parameter as mode input, determining to set with described using preset processing model
Count the performance parameter of the corresponding blower of parameter, wherein the preset processing model is preparatory trained neural network model.
10. a kind of electronic equipment, including processor and for the memory of storage processor executable instruction, feature exists
In the step of processor realizes any one of claims 1 to 8 the method when executing described instruction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910151323.3A CN109707658A (en) | 2019-02-28 | 2019-02-28 | Method for determination of performance parameter, device and the electronic equipment of blower |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910151323.3A CN109707658A (en) | 2019-02-28 | 2019-02-28 | Method for determination of performance parameter, device and the electronic equipment of blower |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109707658A true CN109707658A (en) | 2019-05-03 |
Family
ID=66265945
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910151323.3A Pending CN109707658A (en) | 2019-02-28 | 2019-02-28 | Method for determination of performance parameter, device and the electronic equipment of blower |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109707658A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021051356A1 (en) * | 2019-09-19 | 2021-03-25 | 西门子股份公司 | Design parameter value generation method and apparatus, and computer-readable medium |
CN113325700A (en) * | 2021-05-31 | 2021-08-31 | 西安热工研究院有限公司 | Fan opening and efficiency online calculation method based on fan performance curve |
CN115114785A (en) * | 2022-06-28 | 2022-09-27 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103646297A (en) * | 2013-12-02 | 2014-03-19 | 江苏大学 | Double-channel pump optimization method based on multi-objective genetic algorithm |
CN103852727A (en) * | 2014-02-14 | 2014-06-11 | 清华大学深圳研究生院 | Method and device for estimating power battery charge state on line |
CN106777527A (en) * | 2016-11-24 | 2017-05-31 | 上海市特种设备监督检验技术研究院 | Monkey operation energy consumption analysis method based on neural network model |
CN107330480A (en) * | 2017-07-03 | 2017-11-07 | 贵州大学 | Hand-written character Computer Identification |
CN108629441A (en) * | 2018-03-13 | 2018-10-09 | 中南林业科技大学 | Prediction technique and device based on clustering and the improved fan noise of small echo |
-
2019
- 2019-02-28 CN CN201910151323.3A patent/CN109707658A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103646297A (en) * | 2013-12-02 | 2014-03-19 | 江苏大学 | Double-channel pump optimization method based on multi-objective genetic algorithm |
CN103852727A (en) * | 2014-02-14 | 2014-06-11 | 清华大学深圳研究生院 | Method and device for estimating power battery charge state on line |
CN106777527A (en) * | 2016-11-24 | 2017-05-31 | 上海市特种设备监督检验技术研究院 | Monkey operation energy consumption analysis method based on neural network model |
CN107330480A (en) * | 2017-07-03 | 2017-11-07 | 贵州大学 | Hand-written character Computer Identification |
CN108629441A (en) * | 2018-03-13 | 2018-10-09 | 中南林业科技大学 | Prediction technique and device based on clustering and the improved fan noise of small echo |
Non-Patent Citations (1)
Title |
---|
张鹏: "轴流风机结构参数优化设计", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021051356A1 (en) * | 2019-09-19 | 2021-03-25 | 西门子股份公司 | Design parameter value generation method and apparatus, and computer-readable medium |
CN113325700A (en) * | 2021-05-31 | 2021-08-31 | 西安热工研究院有限公司 | Fan opening and efficiency online calculation method based on fan performance curve |
CN113325700B (en) * | 2021-05-31 | 2022-06-28 | 西安热工研究院有限公司 | Fan opening and efficiency online calculation method based on fan performance curve |
CN115114785A (en) * | 2022-06-28 | 2022-09-27 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
CN115114785B (en) * | 2022-06-28 | 2024-01-23 | 北京理工大学 | Engine performance index parameterization rapid prediction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109707658A (en) | Method for determination of performance parameter, device and the electronic equipment of blower | |
Deb et al. | An integrated approach to automated innovization for discovering useful design principles: Case studies from engineering | |
CN109948149B (en) | Text classification method and device | |
CN108133223B (en) | Device and method for determining convolutional neural network CNN model | |
CN105373703B (en) | A kind of adaptive ability test system based on forgetting curve | |
CN110378016B (en) | Multi-objective optimization design method for pump impeller adjustable hydraulic torque converter | |
CN108898216A (en) | Activation processing unit applied to neural network | |
CN114372318A (en) | Design parameter determination method and device, electronic equipment and readable storage medium | |
CN110443428A (en) | A kind of air compressor group load forecasting method and its control equipment | |
CN108921288A (en) | Neural network activates processing unit and the neural network processor based on the device | |
CN110308658A (en) | A kind of pid parameter setting method, device, system and readable storage medium storing program for executing | |
CN114676522B (en) | Pneumatic shape optimization design method, system and equipment integrating GAN and migration learning | |
CN116050031A (en) | Axial flow impeller blade design method and device, storage medium and electronic equipment | |
CN109829248A (en) | Method for determination of performance parameter, device and the electronic equipment of water pump | |
CN113220815B (en) | Regional information processing method and device | |
CN111310341B (en) | Fan operation parameter determining method, device, equipment and readable storage medium | |
CN116579252A (en) | Particle swarm optimization-based vortex mechanical exhaust hole optimization method and system | |
CN114880890B (en) | Far-field boundary condition multi-window processing method, device and medium | |
CN113993343A (en) | Energy-saving control method and device for refrigeration equipment, terminal and storage medium | |
Rozenberg et al. | Fluid structure interaction problems in turbomachinery using rbf interpolation and greedy algorithm | |
CN114492163A (en) | Blast furnace fan operation condition prediction method, device, equipment and storage medium | |
CN114548539A (en) | Method, device, equipment and medium for predicting turbine energy consumption of compressor of circulating water system | |
CN106294490A (en) | The feature Enhancement Method of data sample and device and classifier training method and apparatus | |
CN113361575A (en) | Model training method and device and electronic equipment | |
CN101866426A (en) | Weighting contraction method based on K near neighbor method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |