CN112648923A - Virtual calculation method for detecting size of product part - Google Patents
Virtual calculation method for detecting size of product part Download PDFInfo
- Publication number
- CN112648923A CN112648923A CN202011468463.2A CN202011468463A CN112648923A CN 112648923 A CN112648923 A CN 112648923A CN 202011468463 A CN202011468463 A CN 202011468463A CN 112648923 A CN112648923 A CN 112648923A
- Authority
- CN
- China
- Prior art keywords
- module
- signal
- data
- output
- size
- 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
- 238000001514 detection method Methods 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims description 20
- 230000005284 excitation Effects 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 13
- 238000013139 quantization Methods 0.000 claims description 13
- 230000005540 biological transmission Effects 0.000 claims description 10
- 239000004065 semiconductor Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 8
- 210000002569 neuron Anatomy 0.000 abstract description 3
- 230000008034 disappearance Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 15
- 230000008859 change Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000005070 sampling Methods 0.000 description 8
- 238000005259 measurement Methods 0.000 description 6
- 230000003750 conditioning effect Effects 0.000 description 4
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001143 conditioned effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- 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
- G06N3/045—Combinations of 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
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of size detection, in particular to a method for detecting the size of a product part through virtual calculation. The device comprises a part size detection platform, wherein the part size detection platform comprises a size detection unit, a data processing unit, a recording unit and a network unit. In addition, when the neuron is in an inactive state, a non-0 gradient is allowed to exist, so that the situations of gradient disappearance and high convergence speed are avoided, the forward calculation amount is small, the back propagation calculation is fast, the derivative calculation is simple, indexes and starting calculation are not needed, and overfitting is reduced.
Description
Technical Field
The invention relates to the technical field of size detection, in particular to a method for detecting the size of a product part through virtual calculation.
Background
At present, along with the progress of industrialization in China, the size requirement on product parts is higher and higher, errors existing in the conventional tool measurement and manual measurement are larger, at present, a plurality of factories adopt laser ranging sensors to measure the distance of the product parts, and display is carried out through a network connection device end, the detection on the product parts is realized, but output layers and input layers of a plurality of networks are connected by a plurality of linear vectors, so that the expression capacity is insufficient, the network learning speed is very low, the detection speed is influenced, the detection efficiency is greatly reduced, and in addition, the overfitting condition is easy to occur when a neuron is in an inactive state.
Disclosure of Invention
The present invention is directed to a method for detecting dimensions of product components through virtual computing, so as to solve the problems mentioned in the background art.
In order to achieve the above object, the present invention provides a method for detecting the dimensions of product components by virtual calculation, which comprises the following steps:
s1.1, size detection: the size detection unit detects the size of the part in real time and outputs a detection signal to the data processing unit;
s1.2, data processing: the data processing unit collects the detection signals and converts the collected detection signals into digital signals;
s1.3, data recording: the recording unit records and stores the data detected by the parts;
s1.4, establishing a network: the network unit establishes a network for data transmission among the units;
the size detection unit comprises a size detection module, a signal acquisition module, a signal processing module, a signal output module and a signal display module; the size detection module is used for controlling the sensor to measure the size of the part in real time; the signal acquisition module is used for acquiring the signal data measured by the size detection module; the signal processing module is used for processing the signal data acquired by the signal acquisition module; the signal processing module detects, converts, filters, amplifies and the like weak signals output by the sensor during detection so as to facilitate subsequent processing or display of the detection system, for example, the output signals of the laser ranging sensor are array changes during measurement of the size of a part, and a four-arm bridge is generally required to be designed so as to facilitate subsequent processing and convert the changes along with the array to be detected into voltage signals; because a noise voltage is often mixed in a signal, a signal conditioning circuit of the signal conditioning circuit usually comprises links of filtering, amplifying, linearizing and the like.
If the remote transmission is needed, the voltage signal obtained by a D/A or V/I circuit is converted into a standard 4-20 m A current signal and then transmitted remotely.
The signal output module is used for outputting the processed signal data to an equipment end; the signal display module is used for displaying the output signal data through the equipment end, and the equipment end comprises:
the value of the measured parameter is represented by the relative position of an optical indicator or a pointer on a scale. It is convenient and intuitive for a tangible pointer displacement to be used to simulate an intangible measurand. The indicating instrument has various moving coil and moving magnet forms, but has simple structure, low cost and visual display, and is applied to single parameter measurement and display occasions with low detection precision requirement;
the digital display equipment end directly displays the value of the measured parameter in a digital form; under normal conditions, the digital display thoroughly eliminates display driving errors, can effectively overcome subjective errors of reading, improves display and reading precision, and can be conveniently connected with a computer and carry out data transmission;
the screen display device has the advantages of image and easy reading, and can simultaneously display one or more measured change curves on the same screen, thereby facilitating the comparison and analysis of the measured change curves, and the display of the change curves is usually controlled by a computer.
The network unit comprises a neural network module and an excitation module; the neural network module is used for establishing data transmission among all units or modules; the excitation module is used for converting a plurality of linear inputs in the neural network module into nonlinear inputs.
As a further improvement of the technical scheme, the data processing unit comprises a data acquisition module and a data conversion module; the data acquisition module is used for acquiring the signal data output by the signal output module and transmitting the acquired signal data to the data conversion module; the data conversion module is used for receiving the signal data of the data acquisition module and converting the analog signal of the signal data into a digital signal through the A/D converter.
As a further improvement of the present technical solution, the conversion steps of the a/D converter are as follows:
s2.1, extracting an instantaneous amplitude value of the detection signal at regular time intervals, wherein a series of sampling values which are discrete in time are obtained after sampling and are called sample value sequences, and the sample value sequences after sampling are discrete in time and can be subjected to time division multiplexing or each sampling value can be converted into a binary digital signal through quantization and coding;
s2.2, performing quantization rounding on the extracted instantaneous amplitude value, namely outputting 0V for all input voltages between 0 and 1V, outputting 1V for all input voltages between 1 and 2V and the like, wherein by adopting the quantization mode, the input voltage is always greater than the output voltage, so that the generated quantization error is always positive, and the maximum quantization error is equal to the interval between two adjacent quantization levels.
And S2.3, coding the quantized signal.
As a further improvement of the present technical solution, the recording unit includes a recording module and a storage module; the recording module is used for recording the detected data and storing the data through the storage module.
As a further improvement of the technical solution, the sensor is a laser ranging sensor, and the measuring method thereof is as follows:
s3.1, opening the semiconductor laser, and focusing the laser generated by the semiconductor laser on the tested part through a focusing lens;
s3.2, emitting laser focused on the part to be detected, and collecting reflected light by a reflecting mirror;
s3.3, projecting the laser collected by the reflecting mirror onto the CMOS array;
and S3.4, the signal processor calculates the position of the light spot on the array through a trigonometric function to obtain the distance from the light spot to the object.
As a further improvement of the technical solution, the neural network module includes an input layer, an output layer, and a hidden layer, the nodes of the input layer and the output layer transmit data signals of each unit or module by adjusting the mutual connection relationship, and the input and output of the nodes of the hidden layer and the output layer have a relationship established by the excitation module.
As a further improvement of the technical solution, the neural network module adopts an RBF neural network, and the working process thereof is as follows:
s4.1, firstly, using RBF as a base of a hidden unit to form a hidden layer space;
s4.2, when the central point of the RBF is determined, the output of the network becomes a linear weighted sum of the output of the hidden unit;
s4.3, the hidden layer maps the linear weighted sum function from the low-dimensional one to the high-dimensional one.
As a further improvement of the present technical solution, a functional formula of the linear weighted sum function in S3.3 is as follows:
wherein the content of the first and second substances,
is an approximation function;
are approximation coefficients;
a different center is assigned to the RBF.
As a further improvement of the present technical solution, the excitation module (142) adopts a leak ReLu function, and the formula thereof is as follows:
as a further improvement of the technical solution, the signal processing module includes a pulse width modulation module, and the working steps thereof are as follows:
s5.1, electrifying the switching device of the inverter circuit to enable the output end to obtain a series of pulses with equal amplitude;
s5.2, replacing a sine wave or a required waveform with a pulse output by an inverter circuit switching device, namely generating a plurality of pulses in a half cycle of an output waveform to enable equivalent voltage of each pulse to be a sine waveform, and obtaining smooth output and less low-order harmonic waves;
and S5.3, modulating the width of each pulse, so that the output voltage of the inverter circuit is changed, and the change of the output frequency is realized.
Compared with the prior art, the invention has the beneficial effects that: in the virtual calculation product part size detection method, the RBF neural network can be used for linearly dividing the low-dimensional linear inseparable condition into the high-dimensional condition, then the mapping of the network from input to output is nonlinear, the network output is linear for adjustable parameters, the weight of the network can be directly solved by a linear equation set, so that the learning speed is greatly accelerated, and the local minimum problem is avoided.
Drawings
FIG. 1 is a block diagram of a component dimension inspection platform according to embodiment 1;
FIG. 2 is a block diagram of a size detection unit of embodiment 1;
FIG. 3 is a block diagram of a data processing unit of embodiment 1;
FIG. 4 is a block diagram of a recording unit of embodiment 1;
fig. 5 is a block diagram of a network unit of embodiment 1;
FIG. 6 is a block flow diagram of the excitation module of embodiment 1;
FIG. 7 is a block diagram showing a flow of the part size inspecting platform according to embodiment 1;
fig. 8 is a waveform diagram of the deployment and binary signals of example 1.
The various reference numbers in the figures mean:
100. a part size detection platform;
110. a size detection unit; 111. a size detection module; 112. a signal acquisition module; 113. a signal processing module; 114. a signal output module; 115. a signal display module;
120. a data processing unit; 121. a data acquisition module; 122. a data conversion module;
130. a recording unit; 131. a recording module; 132. a storage module;
140. a network unit; 141. a neural network module; 142. and an excitation module.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
The invention provides a method for detecting the size of a product part through virtual calculation, which is shown in fig. 1-8 and comprises a part size detection platform 100, wherein the method comprises the following steps:
s1.1, size detection: the size detection unit 110 detects the size of the component in real time and outputs a detection signal to the data processing unit 120;
s2.2, data processing: the data processing unit 120 collects the detection signals and converts the collected detection signals into digital signals;
s2.3, data recording: the recording unit 130 records and stores the data detected by the parts;
s2.4, establishing a network: the network unit 140 establishes a network of data transmission between the respective units;
the size detection unit 110 includes a size detection module 111, a signal acquisition module 112, a signal processing module 113, a signal output module 114, and a signal display module 115; the size detection module 111 is used for controlling the sensor to measure the size of the part in real time; the signal acquisition module 112 is used for acquiring the signal data measured by the size detection module 111; the signal processing module 113 is configured to process the signal data acquired by the signal acquisition module 112; the signal processing module 113 detects, converts, filters, amplifies, etc. the weak signal output by the sensor during detection, so as to facilitate subsequent processing or display of the detection system, for example, the output signal of the laser ranging sensor changes in an array during measurement of the size of a component, and for facilitating subsequent processing, a four-arm bridge is usually required to be designed, so as to convert the change along with the measured array into a voltage signal; because a noise voltage is often mixed in a signal, a signal conditioning circuit of the signal conditioning circuit usually comprises links of filtering, amplifying, linearizing and the like.
If the remote transmission is needed, the voltage signal obtained by a D/A or V/I circuit is converted into a standard 4-20 m A current signal and then transmitted remotely.
The signal output module 114 is configured to output the processed signal data to the device side; the signal display module 115 is configured to display output signal data through a device side, where the device side includes:
the value of the measured parameter is represented by the relative position of an optical indicator or a pointer on a scale. It is convenient and intuitive for a tangible pointer displacement to be used to simulate an intangible measurand. The indicating instrument has various moving coil and moving magnet forms, but has simple structure, low cost and visual display, and is applied to single parameter measurement and display occasions with low detection precision requirement;
the digital display equipment end directly displays the value of the measured parameter in a digital form; under normal conditions, the digital display thoroughly eliminates display driving errors, can effectively overcome subjective errors of reading, improves display and reading precision, and can be conveniently connected with a computer and carry out data transmission;
the screen display device has the advantages of image and easy reading, and can simultaneously display one or more measured change curves on the same screen, thereby facilitating the comparison and analysis of the measured change curves, and the display of the change curves is usually controlled by a computer.
The network unit 140 includes a neural network module 141 and an excitation module 142; the neural network module 141 is used for establishing data transmission among the units or modules; the excitation module 142 is used to convert a plurality of linear inputs in the neural network module 141 into nonlinear inputs.
In this embodiment, the data processing unit 120 includes a data acquisition module 121 and a data conversion module 122; the data acquisition module 121 is configured to acquire signal data output by the signal output module 114, discretize a continuous analog signal conditioned by the data acquisition module 121 and convert the discretized continuous analog signal into a series of numerical information corresponding to the voltage amplitude of the analog signal, and simultaneously transmit the converted data to the microprocessor in a certain manner or automatically store the converted data in sequence, and transmit the acquired signal data to the data conversion module 122; the data conversion module 122 is configured to receive the signal data of the data acquisition module 121, and convert an analog signal of the signal data into a digital signal through an a/D converter.
Further, the conversion steps of the A/D converter are as follows:
s2.1, extracting an instantaneous amplitude value of the detection signal at regular time intervals, wherein a series of sampling values which are discrete in time are obtained after sampling and are called sample value sequences, and the sample value sequences after sampling are discrete in time and can be subjected to time division multiplexing or each sampling value can be converted into a binary digital signal through quantization and coding;
s2.2, performing quantization rounding on the extracted instantaneous amplitude value, namely outputting 0V for all input voltages between 0 and 1V, outputting 1V for all input voltages between 1 and 2V and the like, wherein by adopting the quantization mode, the input voltage is always greater than the output voltage, so that the generated quantization error is always positive, and the maximum quantization error is equal to the interval between two adjacent quantization levels.
S2.3, coding the quantized signal, wherein during specific coding, the quantized sample values are represented by n-bit binary codes, each binary number corresponds to a quantized value, and then the binary values are arranged to obtain a digital information stream consisting of binary pulses.
Specifically, the recording unit 130 includes a recording module 131 and a storage module 132; the recording module 131 is used for recording the detected data and storing the data through the storage module 132.
In addition, the sensor adopts a laser ranging sensor, and the measuring method comprises the following steps:
s3.1, opening the semiconductor laser, and focusing the laser generated by the semiconductor laser on the tested part through a focusing lens;
s3.2, emitting laser focused on the part to be detected, and collecting reflected light by a reflecting mirror;
s3.3, projecting the laser collected by the reflecting mirror onto a CMOS array, wherein the CMOS array is a 4 x 4 array and consists of 8 green pixels, 4 blue pixels and 4 red pixels, and performing 9 times of operation by using a 2 x 2 matrix when converting the gray level image into a color image to finally generate a color image;
and S3.4, the signal processor calculates the position of the light spot on the array through a trigonometric function to obtain the distance from the light spot to the object.
In addition, the neural network module 141 includes an input layer, an output layer, and a hidden layer, the nodes of the input layer and the output layer transmit data signals of each unit or module by adjusting the interconnection relationship, and the input and output of the nodes of the hidden layer and the output layer have a relationship established by the excitation module 142.
Further, the neural network module 141 employs an RBF neural network, and the work flow thereof is as follows:
s4.1, firstly, using RBF as a 'base' of a hidden unit to form a hidden layer space, so that an input vector can be directly mapped to the hidden space without being connected through a right;
s4.2, when the central point of the RBF is determined, the output of the network becomes a linear weighted sum of the output of the hidden unit;
s4.3, the hidden layer maps the linear weighted sum function from the low dimension to the high dimension, so that the low dimension linear inseparable condition can become linearly separable to the high dimension, then the mapping from the input to the output of the network is nonlinear, the network output is linear for the adjustable parameters, and the weight of the network can be directly solved by a linear equation set, thereby greatly accelerating the learning speed and avoiding the local minimum problem.
Specifically, the function formula of the linear weighted sum function in S4.3 is as follows:
wherein, is an approximation function;
are approximation coefficients;
a different center is assigned to the RBF.
The approximation function represents a weighted sum of RBFs, each RBFCorresponding to a different center
And the weight is formed by an approximation coefficient
That is, since the approximation function is a linear weighted average of the RBFs, the coefficients are approximated
The arrival of a linear system of equations can be solved using least squares and the nonlinear function is RBF, whose parameters depend on the euclidean norm of the dimensional space, as follows:
in addition, the excitation module 142 employs a leaky ReLu function, whose formula is as follows:
when the neuron is in an inactive state, the LEAKY ReLu function allows a non-0 gradient to exist, so that the situations of gradient disappearance and high convergence speed cannot occur, the forward calculation amount is small, the back propagation calculation is fast, the derivative calculation is simple, the index and starting calculation are not needed, and overfitting is reduced.
Besides, the signal processing module 113 includes a pulse width modulation module, and its working steps are as follows:
s5.1, electrifying the switching device of the inverter circuit to enable the output end to obtain a series of pulses with equal amplitude;
s5.2, replacing a sine wave or a required waveform with a pulse output by an inverter circuit switching device, namely generating a plurality of pulses in a half cycle of an output waveform to enable equivalent voltage of each pulse to be a sine waveform, and obtaining smooth output and less low-order harmonic waves;
and S5.3, modulating the width of each pulse, so that the output voltage of the inverter circuit is changed, and the change of the output frequency is realized.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and the preferred embodiments of the present invention are described in the above embodiments and the description, and are not intended to limit the present invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. A virtual computing product part size detection method is characterized by comprising a part size detection platform (100), and the method comprises the following steps:
s1.1, size detection: the size detection unit (110) detects the size of the part in real time and outputs a detection signal to the data processing unit (120);
s1.2, data processing: the data processing unit (120) collects the detection signals and converts the collected detection signals into digital signals;
s1.3, data recording: the recording unit (130) records and stores the data detected by the parts;
s1.4, establishing a network: the network unit (140) establishes a network of data transmissions between the various units;
the size detection unit (110) comprises a size detection module (111), a signal acquisition module (112), a signal processing module (113), a signal output module (114) and a signal display module (115); the size detection module (111) is used for controlling the sensor to measure the size of the part in real time; the signal acquisition module (112) is used for acquiring the signal data measured by the size detection module (111); the signal processing module (113) is used for processing the signal data acquired by the signal acquisition module (112); the signal output module (114) is used for outputting the processed signal data to a device end; the signal display module (115) is used for displaying the output signal data through a device end;
the network unit (140) comprises a neural network module (141) and an excitation module (142); the neural network module (141) is used for establishing data transmission among various units or modules; the excitation module (142) is used for converting a plurality of linear inputs in the neural network module (141) into nonlinear inputs.
2. The virtual-computing product-part-size detecting method according to claim 1, characterized in that: the data processing unit (120) comprises a data acquisition module (121) and a data conversion module (122); the data acquisition module (121) is used for acquiring the signal data output by the signal output module (114) and transmitting the acquired signal data to the data conversion module (122); the data conversion module (122) is used for receiving the signal data of the data acquisition module (121) and converting the analog signal of the signal data into a digital signal through an A/D converter.
3. The virtual-computing product-part-size detecting method according to claim 2, characterized in that: the conversion steps of the A/D converter are as follows:
s2.1, extracting an instantaneous amplitude value of the detection signal at regular time intervals;
s2.2, performing quantization rounding on the extracted instantaneous amplitude value, namely outputting 0V by all input voltages between 0 and 1V, outputting 1V by all input voltages between 1 and 2V and the like;
and S2.3, coding the quantized signal.
4. The virtual-computing product-part-size detecting method according to claim 1, characterized in that: the recording unit (130) comprises a recording module (131) and a storage module (132); the recording module (131) is used for recording the detected data and storing the data through the storage module (132).
5. The virtual-computing product-part-size detecting method according to claim 1, characterized in that: the sensor adopts a laser ranging sensor, and the measuring method comprises the following steps:
s3.1, opening the semiconductor laser, and focusing the laser generated by the semiconductor laser on the tested part through a focusing lens;
s3.2, emitting laser focused on the part to be detected, and collecting reflected light by a reflecting mirror;
s3.3, projecting the laser collected by the reflecting mirror onto the CMOS array;
and S3.4, the signal processor calculates the position of the light spot on the array through a trigonometric function to obtain the distance from the light spot to the object.
6. The virtual-computing product-part-size detecting method according to claim 1, characterized in that: the neural network module (141) comprises an input layer, an output layer and a hidden layer, the nodes of the input layer and the output layer transmit data signals of all units or modules by adjusting the mutual connection relationship, and the input and the output of the nodes of the hidden layer and the output layer are in a relationship established by an excitation module (142).
7. The virtual-computing product-part-size detecting method according to claim 6, characterized in that: the neural network module (141) adopts an RBF neural network, and the working process is as follows:
s4.1, firstly, using RBF as a base of a hidden unit to form a hidden layer space;
s4.2, when the central point of the RBF is determined, the output of the network becomes a linear weighted sum of the output of the hidden unit;
s4.3, the hidden layer maps the linear weighted sum function from the low-dimensional one to the high-dimensional one.
8. The virtual-computing product-part-size detecting method according to claim 7, characterized in that: the functional formula of the linear weighted sum function in S4.3 is as follows:
wherein the content of the first and second substances,
is an approximation function;
are approximation coefficients;
a different center is assigned to the RBF.
9. The virtual-computing product-part-size detecting method according to claim 1, characterized in that: the excitation module (142) adopts a 1eaky ReLu function, and the formula is as follows:
10. the virtual-computing product-part-size detecting method according to claim 1, characterized in that: the signal processing module (113) comprises a pulse width modulation module, and the working steps are as follows:
s5.1, electrifying the switching device of the inverter circuit to enable the output end to obtain a series of pulses with equal amplitude;
s5.2, replacing sine waves or required waveforms with pulses output by a switching device of the inverter circuit;
and S5.3, modulating the width of each pulse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011468463.2A CN112648923A (en) | 2020-12-12 | 2020-12-12 | Virtual calculation method for detecting size of product part |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011468463.2A CN112648923A (en) | 2020-12-12 | 2020-12-12 | Virtual calculation method for detecting size of product part |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112648923A true CN112648923A (en) | 2021-04-13 |
Family
ID=75353907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011468463.2A Pending CN112648923A (en) | 2020-12-12 | 2020-12-12 | Virtual calculation method for detecting size of product part |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112648923A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09305761A (en) * | 1996-05-09 | 1997-11-28 | Mitsubishi Electric Corp | Object shape extraction device |
| CN202119401U (en) * | 2011-07-12 | 2012-01-18 | 上海徕肯工贸有限公司 | Three-dimensional laser real-time measurement system for large-scale stock ground |
| CN102506719A (en) * | 2011-11-01 | 2012-06-20 | 浙江大学 | Height gauge for measuring height of trees |
| CN103591903A (en) * | 2013-11-22 | 2014-02-19 | 苏州大学 | System and method for measuring distance between elevator lift car sill and well inner surface |
| CN105115431A (en) * | 2015-06-02 | 2015-12-02 | 绵阳东福高科技有限公司 | Novel CCD laser displacement sensor paper sheet thickness meter |
| CN108569607A (en) * | 2018-06-22 | 2018-09-25 | 西安理工大学 | Elevator faults method for early warning based on bidirectional valve controlled Recognition with Recurrent Neural Network |
| CN109190834A (en) * | 2018-09-12 | 2019-01-11 | 百色学院 | Stock price trend prediction method and system neural network based |
| CN109884530A (en) * | 2019-04-02 | 2019-06-14 | 西安交通大学 | A kind of battery management system and its operating method based on neural network algorithm |
| CN110569593A (en) * | 2019-09-05 | 2019-12-13 | 武汉纺织大学 | Method and system for measuring three-dimensional size of dressed human body, storage medium and electronic equipment |
| CN111578855A (en) * | 2020-06-30 | 2020-08-25 | 河海大学 | Road marking thickness continuous measurement equipment and method |
-
2020
- 2020-12-12 CN CN202011468463.2A patent/CN112648923A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09305761A (en) * | 1996-05-09 | 1997-11-28 | Mitsubishi Electric Corp | Object shape extraction device |
| CN202119401U (en) * | 2011-07-12 | 2012-01-18 | 上海徕肯工贸有限公司 | Three-dimensional laser real-time measurement system for large-scale stock ground |
| CN102506719A (en) * | 2011-11-01 | 2012-06-20 | 浙江大学 | Height gauge for measuring height of trees |
| CN103591903A (en) * | 2013-11-22 | 2014-02-19 | 苏州大学 | System and method for measuring distance between elevator lift car sill and well inner surface |
| CN105115431A (en) * | 2015-06-02 | 2015-12-02 | 绵阳东福高科技有限公司 | Novel CCD laser displacement sensor paper sheet thickness meter |
| CN108569607A (en) * | 2018-06-22 | 2018-09-25 | 西安理工大学 | Elevator faults method for early warning based on bidirectional valve controlled Recognition with Recurrent Neural Network |
| CN109190834A (en) * | 2018-09-12 | 2019-01-11 | 百色学院 | Stock price trend prediction method and system neural network based |
| CN109884530A (en) * | 2019-04-02 | 2019-06-14 | 西安交通大学 | A kind of battery management system and its operating method based on neural network algorithm |
| CN110569593A (en) * | 2019-09-05 | 2019-12-13 | 武汉纺织大学 | Method and system for measuring three-dimensional size of dressed human body, storage medium and electronic equipment |
| CN111578855A (en) * | 2020-06-30 | 2020-08-25 | 河海大学 | Road marking thickness continuous measurement equipment and method |
Non-Patent Citations (2)
| Title |
|---|
| 单清群等: "基于三角激光测量的地铁制动盘辅助测量工装", 《机械制造》 * |
| 田树钰: "《电子电工基础》", 31 May 2009 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109813544B (en) | A kind of rotating machinery Incipient Fault Diagnosis method and system based on on-line monitoring | |
| CN209841947U (en) | Intelligent multimeter with automatic range switching function | |
| CN105321254A (en) | Equipment, method and device for detecting thickness of measured object | |
| CN109584232A (en) | Equipment use state on-line monitoring method, system and terminal based on image recognition | |
| CN109855796A (en) | A kind of testing machine force value automated calibration system and method | |
| CN108279317A (en) | A kind of space filtering tachogenerator device and the method for improving rate accuracy | |
| CN112648923A (en) | Virtual calculation method for detecting size of product part | |
| CN209745322U (en) | multi-parameter measuring instrument for fluid | |
| CN1226682A (en) | RMS converter for obtaining fast RMS measurements | |
| CN214372307U (en) | Digital sphere diameter instrument | |
| CN114500615A (en) | Intelligent terminal based on thing allies oneself with sensing technology | |
| CN111060731A (en) | Digital oscilloscope, and signal processing method and system of digital oscilloscope | |
| CN2849662Y (en) | Integrated metering calibrator for detector | |
| CN203396461U (en) | Differential pressure flowmeter verification system | |
| CN113324938B (en) | Infrared oil measurement method and oil measurement instrument based on pressure-frequency conversion phase-sensitive detector | |
| CN115792141B (en) | Method and system for improving balance detection efficiency of water activity meter | |
| CN116817700B (en) | Caliper quality detection method and system | |
| CN212903242U (en) | Metering device of speedometer | |
| CN216116288U (en) | Monitoring device applied to mechanical flowmeter | |
| CN113945253B (en) | Water level measuring method for rail traffic track area | |
| CN116381321B (en) | Pole-mounted switch voltage signal measuring device and pole-mounted switch power-off control method | |
| CN216484461U (en) | On-line intelligent detection regulating valve | |
| CN111064532B (en) | Transmission delay test method and system for unmanned platform remote control and remote measurement data system | |
| CN114791317B (en) | High-sensitivity weighing system based on sensor | |
| CN2424451Y (en) | Resistance type intelligent wet temp detector |
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: 20210413 |