CN113776675B - Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium - Google Patents
Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium Download PDFInfo
- Publication number
- CN113776675B CN113776675B CN202111086421.7A CN202111086421A CN113776675B CN 113776675 B CN113776675 B CN 113776675B CN 202111086421 A CN202111086421 A CN 202111086421A CN 113776675 B CN113776675 B CN 113776675B
- Authority
- CN
- China
- Prior art keywords
- emissivity
- temperature
- generalized inverse
- neural network
- voltage value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000005855 radiation Effects 0.000 title claims abstract description 41
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 33
- 238000009529 body temperature measurement Methods 0.000 title claims abstract description 26
- 238000004364 calculation method Methods 0.000 title claims abstract description 25
- 239000011159 matrix material Substances 0.000 claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000001228 spectrum Methods 0.000 claims abstract description 15
- 230000008859 change Effects 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 7
- 238000012795 verification Methods 0.000 claims abstract description 7
- 238000002922 simulated annealing Methods 0.000 claims abstract description 4
- 230000003595 spectral effect Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 12
- 230000001419 dependent effect Effects 0.000 claims description 7
- 238000003556 assay Methods 0.000 claims description 3
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 2
- 238000003672 processing method Methods 0.000 abstract description 9
- 239000000463 material Substances 0.000 abstract description 6
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000000137 annealing Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004861 thermometry Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
- G01J5/10—Radiation pyrometry, e.g. infrared or optical thermometry using electric radiation detectors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
- G06F17/12—Simultaneous equations, e.g. systems of linear equations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The application provides a multispectral radiation temperature measurement inversion calculation method, a computer and a storage medium based on a generalized inverse-neural network, and belongs to the technical field of multispectral radiation temperature measurement inversion calculation. Firstly, simulating a multispectral radiation thermometer to calculate a voltage value of each spectrum channel; secondly, constructing an equation for each spectrum channel to form an equation set, and obtaining temperature and emissivity data with a rule similar to that in an emissivity model through generalized inverse matrix calculation; secondly, comparing the data with the emissivity change rule in the emissivity model, and classifying; secondly, defining a group of super parameters, training the neural network, and modifying the parameters by using a simulated annealing algorithm in the training process to train optimal parameters; and finally, inputting the test set verification sample set into a neural network, and outputting multispectral radiation temperature measurement data. The method solves the problem that the original data processing method cannot be universally applied to materials with different emissivity and cannot be used for rapidly inverting in the prior art.
Description
Technical Field
The application relates to a multispectral radiation temperature measurement inversion calculation method, in particular to a multispectral radiation temperature measurement inversion calculation method, a computer and a storage medium based on a generalized inverse-neural network, and belongs to the technical field of multispectral radiation temperature measurement inversion calculation.
Background
In the aspect of a data processing inversion algorithm in the multispectral temperature measurement field, the difficulty of the multispectral radiation temperature measurement data processing method is that data obtained through n spectrum channels are used for establishing n radiation equations according to a Planck formula, but n+1 unknowns (n unknown spectrum emissivity and 1 unknown real temperature) exist.
Currently, the data processing methods of the multispectral pyrometer are roughly classified into two types, namely a data processing method based on a fixed emissivity model and a data processing method based on a variable emissivity model. The data processing method based on the fixed emissivity model needs to pre-assume the relation between the spectral emissivity and the wavelength and then perform data processing. The method has better results only when the emissivity assumption model is consistent with the actual situation, so the method is only suitable for processing the actual temperature and emissivity data of a certain material
The data processing method based on the variable emissivity model is characterized in that an emissivity assumption model can be changed within a certain range according to different measured objects, and the method can solve the problems of measuring the actual temperature and emissivity of a certain type or a plurality of types of measured objects, but can not be suitable for all materials.
At present, the multi-wavelength radiation temperature measurement technology is still limited to single-point temperature measurement in instrument development, and multi-point measurement needs to completely replicate single-point optics and circuit systems, so that the device is complex, the cost is high, and only measurement of limited temperature points can be realized. The reason is that the existing multi-wavelength pyrometer is difficult to meet the problems of accurate collection of a large amount of radiation information and rapid inversion of the real temperature based on a large amount of radiation information in the line temperature measurement process from the inversion theory. Therefore, to realize online measurement of the multi-wavelength high temperature Ji Xianwen, the problem that the real temperature inversion accuracy is affected by unknown spectral emissivity in the data processing process and the multi-point real temperature quick inversion is difficult to realize under the condition of the multi-point multi-wavelength data volume obtained by online temperature measurement is solved.
Disclosure of Invention
The following presents a simplified summary of the application in order to provide a basic understanding of some aspects of the application. It should be understood that this summary is not an exhaustive overview of the application. It is not intended to identify key or critical elements of the application or to delineate the scope of the application. Its purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.
In view of the above, in order to solve the technical problem that the original data processing method in the prior art cannot be universally applied to materials with different emissivity and cannot quickly perform inversion calculation, the application provides a multispectral radiation temperature measurement inversion calculation method based on a generalized inverse-neural network, which comprises the following steps:
step one, simulating a multispectral radiation thermometer to calculate a voltage value of each spectrum channel; the method comprises the following steps: giving a group of reference temperature and emissivity model parameters, calculating the voltage value of each spectrum channel through a Planck formula, forming a data set by taking the voltage value as an independent variable and the temperature value as a dependent variable, selecting 80% of the voltage value of the data set as a training sample set, and selecting 20% of the voltage value as a verification sample set for training a neural network;
step two, based on a multi-wavelength radiation temperature measurement theory, constructing an equation for each spectrum channel to form an equation set, wherein the equation set is an underdetermined equation set consisting of n equations containing n+1 unknowns, and temperature and emissivity data with a rule similar to that in the emissivity model in the step one are obtained through generalized inverse matrix calculation;
step three, comparing the temperature and emissivity data with similar rules in the emissivity model obtained by generalized inverse matrix calculation with the emissivity change rules in the emissivity model in the step one, classifying, wherein each group of temperature and emissivity data corresponds to a voltage value, and classifying the temperature and emissivity data, namely classifying the voltage value and the temperature value;
step four, defining a group of super parameters, inputting the classified voltage value and temperature value in the step three into a neural network to train the neural network, and modifying parameters by using a simulated annealing algorithm in the training process to train optimal parameters;
and fifthly, inputting the test set verification sample set in the first step into the neural network in the fourth step, and outputting multispectral radiation temperature measurement data.
Preferably, the given set of reference temperatures in step one ranges from 1300K to 2100K, and one reference temperature is set every 100K.
Preferably, in the specific method of the step one, the given set of parameters of the reference emissivity model is that four emissivity models are given, each emissivity model sets twenty emissivity change rules, and the value range of the emissivity is 0-1.
Preferably, the planck formula of step one is specifically,
wherein V is i Representing the voltage value of the ith channel of the multi-channel radiation thermometer for measuring the temperature of an object, V i ' represents the voltage value of the ith channel of the reference temperature multichannel radiation thermometer, ∈ (λ) i T represents ε (λ) i T) is the target spectral emissivity at temperature T, lambda i Represents wavelength, T represents temperature, C 2 Representing a second radiation constant.
Preferably, in the second step, an equation is constructed for each spectrum channel based on the multi-wavelength radiation temperature measurement theory to form an equation set, the equation set is a underdetermined equation set composed of n equations including n+1 unknowns, a specific method for obtaining a set of spectral emissivity and temperature values through generalized inverse matrix calculation is that,
output signal V of ith channel of multi-wavelength thermometer i Is that
Wherein A is λi Is an assay constant; epsilon (lambda) i T) is the target spectral emissivity at temperature T;
at the fixed point blackbody reference temperature T', the output signal V of the ith channel is due to the blackbody emissivity of 1 i ' as
Is obtained by the formulas (1) and (2)
Sorting and unifying known amounts with Y i The unknown amount of the spectral emissivity is represented by X i The unknown quantity containing temperature is represented by X:
thus, formula (4) is represented as Y i =X i +a i X, converted into matrix form, i.e
I.e.
Y=AX (6)
Matrix in which A is n× (n+1), no A -1 If an n m matrix G exists, the generalized inverse matrix definition is used for an m n matrix A
AGA=A (7)
GAG=G (8)
(AG) H =AG (9)
(GA) H =GA (10)
Then G is named as a Moore-Penrose generalized inverse of A, and the four equations from the formula (7) to the formula (10) are named as Moore-Penrose equations, namely M-P equations; let A epsilon Cm x n, if there is some G epsilon Cn x M, satisfy all or some of M-P equations (7) to (10), then G is called A generalized inverse matrix;
the class of generalized inverse matrix satisfying all 4M-P equations of equations (7) through (10) is denoted as A {1,2,3,4}, the generalized inverse matrix is the only generalized inverse of A, denoted as A + Called the plus sign inverse; thus there is
X=A + Y (11)。
A computer comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of a multi-spectral radiation thermometry inversion calculation method based on a generalized inverse-neural network when executing the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements a method of a generalized inverse-neural network based multi-spectral radiation thermometry inversion calculation method.
The beneficial effects of the application are as follows: the application solves the problem of materials with different emissivity and realizes the accurate processing of multispectral radiation temperature measurement data. By introducing a generalized inverse matrix, a method for solving an equation set in an inversion way, the operation efficiency is improved, and by introducing a neural network, the cancellation of the emissivity and wavelength hypothesis model is realized, so that the measurement problem of the real temperature and the spectral emissivity is solved. The neural network is utilized to process the data, so that the influence of emissivity is avoided, and the accuracy is higher. The method solves the technical problems that the original data processing method in the prior art cannot be universally applied to materials with different emissivity and cannot quickly perform inversion calculation.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a schematic flow chart of a method according to an embodiment of the application;
FIG. 2 is a schematic diagram of an emissivity model in accordance with an embodiment of the application.
Detailed Description
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following detailed description of exemplary embodiments of the present application is provided in conjunction with the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application and not exhaustive of all embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
Example 1, referring to fig. 1-2, the method for calculating the inversion of the multispectral radiation temperature measurement based on the generalized inverse-neural network in this example includes the following steps:
step one, simulating a multispectral radiation thermometer to calculate a voltage value of each spectrum channel; the method comprises the following steps: giving a group of reference temperature and emissivity model parameters, calculating the voltage value of each spectrum channel through a Planck formula, forming a data set by taking the voltage value as an independent variable and the temperature value as a dependent variable, selecting 80% of the voltage value of the data set as a training sample set, and selecting 20% of the voltage value as a verification sample set for training a neural network;
specifically, the method for setting the temperature is that the reference temperature ranges from 1300K to 2100K, and one reference temperature is set every 100K.
Specifically, the specific method for giving the emissivity is to give four emissivity models (refer to fig. 2), wherein twenty emissivity change rules are set for each emissivity model, and the value range of the emissivity is 0-1.
Specifically, the four kinds of emissivity models are classified into an a-class emissivity model, a b-class emissivity model, a c-class emissivity model, and a d-class emissivity model (refer to fig. 2).
Specifically, the Planck formula is,
wherein V is i Representing the voltage value of the ith channel of the multi-channel radiation thermometer for measuring the temperature of an object, V i ' represents the voltage value of the ith channel of the reference temperature multichannel radiation thermometer, ∈ (λ) i T represents ε (λ) i T) is the target spectral emissivity at temperature T,the corresponding relation between the obtained voltage value and the temperature and the emissivity are obtained according to the Planck formula.
Specifically, the emissivity model refers to the law of emissivity changing along with wavelength;
specifically, twenty-one kinds of data are emissivity change rules with the same trend based on an emissivity model, and referring to the table, 21 kinds of emissivity data with change of wavelength in an a-type emissivity model in fig. 2
Table I, 21 wavelength dependent emissivity data for the class a emissivity model of FIG. 2
L1=0.5 | L2=0.6 | L3=0.7 | L4=0.8 | L5=0.9 | L6=1.0 |
1 | 0.8 | 0.6 | 0.4 | 0.2 | 0 |
1 | 0.9 | 0.8 | 0.7 | 0.6 | 0.5 |
1 | 0.95 | 0.9 | 0.85 | 0.8 | 0.75 |
1 | 0.98 | 0.96 | 0.94 | 0.92 | 0.9 |
1 | 0.85 | 0.7 | 0.55 | 0.4 | 0.25 |
1 | 0.96 | 0.92 | 0.88 | 0.84 | 0.8 |
0.95 | 0.9 | 0.85 | 0.8 | 0.75 | 0.7 |
0.95 | 0.94 | 0.93 | 0.92 | 0.91 | 0.9 |
0.95 | 0.93 | 0.91 | 0.89 | 0.87 | 0.85 |
0.95 | 0.92 | 0.89 | 0.86 | 0.83 | 0.8 |
0.95 | 0.94 | 0.93 | 0.92 | 0.91 | 0.9 |
0.9 | 0.85 | 0.8 | 0.75 | 0.7 | 0.65 |
0.9 | 0.88 | 0.86 | 0.84 | 0.82 | 0.8 |
0.9 | 0.89 | 0.88 | 0.87 | 0.86 | 0.85 |
0.9 | 0.87 | 0.84 | 0.81 | 0.78 | 0.75 |
0.9 | 0.86 | 0.82 | 0.78 | 0.74 | 0.7 |
0.85 | 0.84 | 0.83 | 0.82 | 0.81 | 0.8 |
0.85 | 0.83 | 0.81 | 0.79 | 0.77 | 0.75 |
0.85 | 0.82 | 0.79 | 0.76 | 0.73 | 0.7 |
0.88 | 0.86 | 0.84 | 0.82 | 0.8 | 0.78 |
0.88 | 0.87 | 0.86 | 0.85 | 0.84 | 0.83 |
Step two, based on a multi-wavelength radiation temperature measurement theory, constructing an equation for each spectrum channel to form an equation set, wherein the equation set is an underdetermined equation set consisting of n equations containing n+1 unknowns, and temperature and emissivity data with a rule similar to that in the emissivity model in the step one are obtained through generalized inverse matrix calculation;
because the equation is a system of underdetermined equations, a specific value cannot be calculated, but a group of data can be obtained, the group of data has similarity with the change rule of the emissivity of the data, so when the voltage value of the temperature measuring instrument is input, a group of data similar to the change rule of the emissivity can be obtained through calculation, and the change rule of the emissivity of the measured object can be known, thereby achieving the aim of classification. The data of the same emissivity model is used for network training, so that a better training effect can be achieved, and the accuracy of the network training can be improved.
Preferably, in the second step, an equation is constructed for each spectrum channel based on the multi-wavelength radiation temperature measurement theory to form an equation set, the equation set is a underdetermined equation set composed of n equations including n+1 unknowns, a specific method for obtaining a set of spectral emissivity and temperature values through generalized inverse matrix calculation is that,
output signal V of ith channel of multi-wavelength thermometer i Is that
Wherein A is λi Is a wavelength-dependent and temperature-independent assay constant that relates to the spectral response, optical transmission, geometry, and first radiation constant of the detector at that wavelength; epsilon (lambda) i T) is the target spectral emissivity at temperature T, lambda i Represents wavelength, T represents temperature, C 2 A second radiation constant, 0.01438769.
At the fixed point blackbody reference temperature T', the output signal V of the ith channel is due to the blackbody emissivity of 1 i ' as
Is obtained by the formulas (1) and (2)
Sorting and unifying known amounts with Y i The unknown amount of the spectral emissivity is represented by X i The unknown quantity containing temperature is represented by X:
thus, formula (4) can be used as Y i =X i +a i X denotes, converted into matrix form, i.e
I.e.
Y=AX (6)
Matrix in which A is n× (n+1), no A -1 If an n m matrix G exists, the generalized inverse matrix definition is used for an m n matrix A
AGA=A (7)
GAG=G (8)
(AG) H =AG (9)
(GA) H =GA (10)
Then G is named as a Moore-Penrose generalized inverse of A, and the four equations are named as Moore-Penrose equations, and are called M-P equations for short; let A epsilon Cm x n, if there is some G epsilon Cn x M, satisfy all or some of M-P equations (7) to (10), then G is called A generalized inverse matrix;
the generalized inverse matrix class satisfying all 4M-P equations of equations (7) through (10) is denoted as A {1,2,3,4}, and such generalized inverse is only one generalized inverse unique to a given A, denoted as A + Called the plus sign inverse; thus there is
X=A + Y (11)。
Step three, comparing the temperature and emissivity data with similar rules in the emissivity model obtained by generalized inverse matrix calculation with the emissivity change rules in the emissivity model in the step one, and classifying;
specifically, temperature and emissivity data with similar rules are grouped in the same emissivity model, each group of temperature and emissivity data corresponds to a voltage value, and the temperature and emissivity data are classified, namely, the voltage value and the temperature value are classified;
step four, defining a group of super parameters, inputting the classified voltage value and temperature value in the step three into a neural network to train the neural network, and modifying parameters by using a simulated annealing algorithm in the training process to train optimal parameters;
the optimal parameters are set through the annealing algorithm, the annealing algorithm refers to random factors, a certain worse result can be accepted, namely, the parameters are changed firstly to enable the loss value of the neural network to be the lowest, the parameters are still continuously modified after the local minimum value is reached, the loss function can be high, but the local optimal solution can be possibly jumped out, better parameters are obtained, the loss function is further reduced, and the network training result is more accurate.
The input of the neural network is a voltage value ratio, the ratio of the voltage of the thermometer and the temperature to be measured have a nonlinear relation, the neural network is used for obtaining the nonlinear relation therein, the spectral emissivity is a part of a formula, when the neural network is used for training the network, a part of the voltage value is calculated, the voltage value is calculated by the neural network and used for training the network, when the neural network is used for practical application after the network is successfully trained, the practical measured voltage value is input, and the corresponding temperature value can be obtained through the trained network.
Specifically, the topology structure of the nonlinear mapping relation network for solving the voltage measurement values under a plurality of wavelengths and the target real temperature by using the neural network adopts a structure of 8-4-4-1, and theoretically, the rising process of the object temperature is known to have a relation, so the temperature data can be regarded as a ordered sequence, and the RNN network can be used for acquiring the inherent relation of the data among different temperatures, so the GRU network structure belonging to the RNN is added in the first layer of the neural network.
t is the input of the current moment, ht-1 is the hidden state of the last moment, and ht is the hidden state calculated at the last moment.
When calculating the hidden state at the current time, it will first calculate a candidate state h't and when calculating the candidate state, the value of the reset gate will be considered.
Reset gate rt=σ (Wr [ ht-1, xt ])
If the reset gate is close to 0, the current candidate h't ignores the previous hidden state ht-1 and is calculated using the current input xt. This effectively allows the hidden state to discard any irrelevant information that is found in the future. The calculation formula of the candidate value:
h’t=tanh(Wh[rt*ht-1,xt])
after the candidate value is calculated, the gate is updated to control how much information from the previous hidden state can be transferred to the current hidden state. This is similar to the memory cells of LSTM, allowing the GRU to memorize long-term information.
Updating a calculation formula of the door:
zt=σ(Wz[ht-1,xt])
calculating the hidden state at the current moment:
ht=(1-zt)*ht-1+zt*h’t
because each hidden unit is a separate reset gate and update gate, each hidden unit will learn to capture the dependence of different time ranges. These learning capture short-term dependent units tend to have reset gates activated frequently, but learning long-term dependent units almost always activates update gates.
And fifthly, inputting the test set verification sample set in the first step into the neural network in the fourth step, and outputting multispectral radiation temperature measurement data.
The computer device of the present application may be a device including a processor and a memory, such as a single chip microcomputer including a central processing unit. And the processor is used for realizing the steps of the recommendation method based on the CREO software and capable of modifying the recommendation data driven by the relation when executing the computer program stored in the memory.
The processor may be a central processing unit (Central Processing Unit, CPU), other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data (such as audio data, phonebook, etc.) created according to the use of the handset, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
Computer-readable storage medium embodiments
The computer readable storage medium of the present application may be any form of storage medium readable by a processor of a computer apparatus, including but not limited to, nonvolatile memory, volatile memory, ferroelectric memory, etc., having a computer program stored thereon, which when read and executed by the processor of the computer apparatus, can implement the steps of the above-described modeling method based on the CREO software, which can modify the modeling data driven by the relationship.
The computer program comprises computer program code which may be in source code form, object code form, executable file or in some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
While the application has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of the above description, will appreciate that other embodiments are contemplated within the scope of the application as described herein. Furthermore, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The disclosure of the present application is intended to be illustrative, but not limiting, of the scope of the application, which is defined by the appended claims.
Claims (5)
1. A multispectral radiation temperature measurement inversion calculation method based on a generalized inverse-neural network is characterized by comprising the following steps:
step one, simulating a multispectral radiation thermometer to calculate a voltage value of each spectrum channel; the method comprises the following steps: giving a group of reference temperature and emissivity model parameters, calculating the voltage value of each spectrum channel through a Planck formula, forming a data set by taking the voltage value as an independent variable and the temperature value as a dependent variable, selecting 80% of the voltage value of the data set as a training sample set, and selecting 20% of the voltage value as a verification sample set for training a neural network;
the planck formula is specifically a planck formula,
wherein V is i Representing the voltage value of the ith channel of the multi-channel radiation thermometer for measuring the temperature of an object, V i ' represents the voltage value of the ith channel of the reference temperature multichannel radiation thermometer, ∈ (λ) i T represents ε (λ) i T) is the target spectral emissivity at temperature T, lambda i Represents wavelength, T represents temperature, C 2 Representing a second radiation constant;
step two, based on a multi-wavelength radiation temperature measurement theory, constructing an equation for each spectrum channel to form an equation set, wherein the equation set is an underdetermined equation set consisting of n equations containing n+1 unknowns, and temperature and emissivity data with a rule similar to that in the emissivity model in the step one are obtained through generalized inverse matrix calculation, and the method comprises the following steps:
output signal V of ith channel of multi-wavelength thermometer i Is that
Wherein A is λi Is an assay constant; epsilon (lambda) i T) is the target spectral emissivity at temperature T;
at the fixed point blackbody reference temperature T', the output signal V of the ith channel is due to the blackbody emissivity of 1 i ' as
Is obtained by the formulas (1) and (2)
Sorting and unifying known amounts with Y i The unknown amount of the spectral emissivity is represented by X i The unknown quantity containing temperature is represented by X:
thus, formula (4) is represented as Y i =X i +a i X, converted into matrix form, i.e
I.e.
Y=AX (6)
Matrix in which A is n× (n+1), no A -1 If an n m matrix G exists, the generalized inverse matrix definition is used for an m n matrix A
AGA=A (7)
GAG=G (8)
(AG) H =AG (9)
(GA) H =GA (10)
Then G is named as a Moore-Penrose generalized inverse of A, and the four equations from the formula (7) to the formula (10) are named as Moore-Penrose equations, namely M-P equations; let A epsilon Cm x n, if there is some G epsilon Cn x M, satisfy all or some of M-P equations (7) to (10), then G is called A generalized inverse matrix;
the class of generalized inverse matrix satisfying all 4M-P equations of equations (7) through (10) is denoted as A {1,2,3,4}, the generalized inverse matrix is the only generalized inverse of A, denoted as A + Called the plus sign inverse, thus
X=A + Y (11);
Step three, comparing the temperature and emissivity data with similar rules in the emissivity model obtained by generalized inverse matrix calculation with the emissivity change rules in the emissivity model in the step one, classifying, wherein each group of temperature and emissivity data corresponds to a voltage value, and classifying the temperature and emissivity data, namely classifying the voltage value and the temperature value;
step four, defining a group of super parameters, inputting the classified voltage value and temperature value in the step three into a neural network to train the neural network, and modifying parameters by using a simulated annealing algorithm in the training process to train optimal parameters;
and fifthly, inputting the verification sample set in the first step into the neural network in the fourth step, and outputting multispectral radiation temperature measurement data.
2. The method of claim 1, wherein the given set of reference temperatures in step one ranges from 1300K to 2100K, with one reference temperature set every 100K.
3. The method according to claim 2, wherein the specific method of the step of giving the set of parameter emissivity model parameters is to give four emissivity models, each emissivity model sets twenty emissivity change rules, and the emissivity is in the range of 0-1.
4. A computer comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of the method of any one of claims 1 to 3 when the computer program is executed.
5. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any one of claims 1 to 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111086421.7A CN113776675B (en) | 2021-09-16 | 2021-09-16 | Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111086421.7A CN113776675B (en) | 2021-09-16 | 2021-09-16 | Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113776675A CN113776675A (en) | 2021-12-10 |
CN113776675B true CN113776675B (en) | 2023-09-29 |
Family
ID=78851505
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111086421.7A Active CN113776675B (en) | 2021-09-16 | 2021-09-16 | Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113776675B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117313553B (en) * | 2023-11-28 | 2024-02-06 | 四川物科光学精密机械有限公司 | Multispectral radiation temperature measurement inversion calculation method based on neural network |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7235773B1 (en) * | 2005-04-12 | 2007-06-26 | Itt Manufacturing Enterprises, Inc. | Method and apparatus for image signal compensation of dark current, focal plane temperature, and electronics temperature |
CN105043555A (en) * | 2015-09-18 | 2015-11-11 | 上海工程技术大学 | Method for calculating spectral emissivity and true temperature |
CN111044565A (en) * | 2019-12-30 | 2020-04-21 | 西安近代化学研究所 | True temperature multispectral measurement method in process of converting combustion into detonation |
CN111707373A (en) * | 2020-06-01 | 2020-09-25 | 中国科学院空天信息创新研究院 | Temperature and emissivity separation method and device |
CN112113666A (en) * | 2020-08-31 | 2020-12-22 | 哈尔滨工程大学 | Multispectral temperature measuring device based on self-adaptive emissivity model and temperature measuring method thereof |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150276479A1 (en) * | 2014-03-26 | 2015-10-01 | Intellectual Property Transfer, LLC | Method accounting for thermal effects of lighting and radiation sources for spectroscopic applications |
-
2021
- 2021-09-16 CN CN202111086421.7A patent/CN113776675B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7235773B1 (en) * | 2005-04-12 | 2007-06-26 | Itt Manufacturing Enterprises, Inc. | Method and apparatus for image signal compensation of dark current, focal plane temperature, and electronics temperature |
CN105043555A (en) * | 2015-09-18 | 2015-11-11 | 上海工程技术大学 | Method for calculating spectral emissivity and true temperature |
CN111044565A (en) * | 2019-12-30 | 2020-04-21 | 西安近代化学研究所 | True temperature multispectral measurement method in process of converting combustion into detonation |
CN111707373A (en) * | 2020-06-01 | 2020-09-25 | 中国科学院空天信息创新研究院 | Temperature and emissivity separation method and device |
CN112113666A (en) * | 2020-08-31 | 2020-12-22 | 哈尔滨工程大学 | Multispectral temperature measuring device based on self-adaptive emissivity model and temperature measuring method thereof |
Non-Patent Citations (2)
Title |
---|
基于参考温度的多目标极小值优化原理的多光谱真温反演;张福才;刘允刚;孙晓刚;;光谱学与光谱分析(第07期);全文 * |
基于神经网络模型的多波长辐射测温的研究;孙晓刚, 何瑾, 戴景民, 褚载祥;哈尔滨工业大学学报(第06期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113776675A (en) | 2021-12-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bajorski | Statistics for imaging, optics, and photonics | |
CN111079780B (en) | Training method for space diagram convolution network, electronic equipment and storage medium | |
CN113776675B (en) | Multispectral radiation temperature measurement inversion calculation method based on generalized inverse-neural network, computer and storage medium | |
Karchev et al. | SICRET: Supernova Ia cosmology with truncated marginal neural Ratio EsTimation | |
Narisetti et al. | Fully-automated root image analysis (faRIA) | |
CN116994117A (en) | Training method, device, equipment and storage medium of target spectrum analysis model | |
CN107742029A (en) | Increasing knowledge and magnanimity based on SVMs are super to return load modeling multi-cure-fitting model | |
CN114974421A (en) | Single-cell transcriptome sequencing data interpolation method and system based on diffusion-noise reduction | |
Su et al. | A precise method of color space conversion in the digital printing process based on PSO-DBN | |
Yoon et al. | Fast computation of latent correlations | |
CN111814804B (en) | Human body three-dimensional size information prediction method and device based on GA-BP-MC neural network | |
US9864834B2 (en) | High-resolution melt curve classification using neural networks | |
Vargas et al. | Relu-based activations: Analysis and experimental study for deep learning | |
CN111478742A (en) | SM4 algorithm analysis method, system and equipment | |
Haraksim et al. | Validation of likelihood ratio methods for forensic evidence evaluation handling multimodal score distributions | |
Bini et al. | A defect-correction algorithm for quadratic matrix equations, with applications to quasi-Toeplitz matrices | |
CN107292280A (en) | A kind of seal automatic font identification method and identifying device | |
Wang et al. | Classification and identification of crop disease based on depthwise separable group convolution and feature fusion | |
Jakubiec et al. | Error model application in neural reconstruction of nonlinear sensor input signal | |
Thilagavathi et al. | A novel hyperspectral image classification technique using deep multi-dimensional recurrent neural network | |
Ağraz et al. | Extended lasso-type MARS (LMARS) model in the description of biological network | |
Li et al. | Kernel-based spectral color image segmentation | |
CN117392564B (en) | River water quality inversion method based on deep learning, electronic equipment and storage medium | |
CN112417188B (en) | Hyperspectral image classification method based on graph model | |
Khachumov et al. | Pattern-Recognition Tools and Their Applications |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |