CN117455867A - Magnetic core performance optimization management system based on neural network - Google Patents
Magnetic core performance optimization management system based on neural network Download PDFInfo
- Publication number
- CN117455867A CN117455867A CN202311417724.1A CN202311417724A CN117455867A CN 117455867 A CN117455867 A CN 117455867A CN 202311417724 A CN202311417724 A CN 202311417724A CN 117455867 A CN117455867 A CN 117455867A
- Authority
- CN
- China
- Prior art keywords
- magnetic core
- unit
- equalization processing
- thickness
- parameter identification
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 26
- 238000001514 detection method Methods 0.000 claims abstract description 39
- 238000007726 management method Methods 0.000 claims abstract description 34
- 238000006243 chemical reaction Methods 0.000 claims abstract description 21
- 238000003384 imaging method Methods 0.000 claims abstract description 19
- 230000001360 synchronised effect Effects 0.000 claims abstract description 15
- 238000003062 neural network model Methods 0.000 claims abstract description 13
- 238000004519 manufacturing process Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims description 65
- 230000000875 corresponding effect Effects 0.000 claims description 58
- 238000001914 filtration Methods 0.000 claims description 58
- 238000005259 measurement Methods 0.000 claims description 27
- 230000000007 visual effect Effects 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 21
- 238000003860 storage Methods 0.000 claims description 18
- 230000007246 mechanism Effects 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 8
- 230000002596 correlated effect Effects 0.000 claims description 5
- 238000000034 method Methods 0.000 claims description 3
- 239000011162 core material Substances 0.000 description 123
- 238000010586 diagram Methods 0.000 description 7
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 5
- 229910052802 copper Inorganic materials 0.000 description 5
- 239000010949 copper Substances 0.000 description 5
- 230000004907 flux Effects 0.000 description 4
- 230000002950 deficient Effects 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- UQSXHKLRYXJYBZ-UHFFFAOYSA-N Iron oxide Chemical compound [Fe]=O UQSXHKLRYXJYBZ-UHFFFAOYSA-N 0.000 description 2
- 229910001053 Nickel-zinc ferrite Inorganic materials 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 229910000859 α-Fe Inorganic materials 0.000 description 2
- 229910001289 Manganese-zinc ferrite Inorganic materials 0.000 description 1
- JIYIUPFAJUGHNL-UHFFFAOYSA-N [O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[Mn++].[Mn++].[Mn++].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Zn++].[Zn++] Chemical compound [O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[O--].[Mn++].[Mn++].[Mn++].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Fe+3].[Zn++].[Zn++] JIYIUPFAJUGHNL-UHFFFAOYSA-N 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000004941 influx Effects 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000011701 zinc Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a magnetic core performance optimization management system based on a neural network, which comprises: the synchronous driving device is used for conveying the produced magnetic core products to the vision acquisition station in a vertical state when the magnetic core production assembly finishes the production of the magnetic core products every time; and the error alarm device is used for sending a thickness error detection signal when the matching identifier output by the deep neural network model of the targeted structural design indicates that the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core. According to the invention, the imaging pattern of the magnetic core product in the interpolation conversion image obtained after the corresponding product delivery picture is optimized in a targeted manner can be detected, and whether the thickness of the current magnetic core product meets the standard or not is intelligently analyzed by adopting the deep neural network model based on various basic data of the imaging pattern and the corresponding reference thickness of the magnetic core, so that the intelligent level and the automatic level of the magnetic core thickness identification management are improved, and meanwhile, the magnetic core performance optimization management system is satisfied.
Description
Technical Field
The invention relates to the field of magnetic cores, in particular to a magnetic core performance optimization management system based on a neural network.
Background
The magnetic core refers to a sintered magnetic metal oxide composed of various iron oxide mixtures. For example, manganese-zinc ferrite and nickel-zinc ferrite are typical core materials. The Mn-Zn ferrite has the characteristics of high magnetic permeability and high magnetic flux density, and has the characteristic of low loss. The nickel-zinc ferrite has the characteristics of extremely high impedance rate, low magnetic permeability of less than hundreds, and the like. Ferrite cores are used in coils and transformers for various electronic devices.
The magnetic core detection technology disclosed in the prior art comprises the following steps: the invention discloses a core detection device, which comprises a core conveying assembly, a detection assembly and a screening assembly, wherein the core conveying assembly comprises a first conveying belt and a second conveying belt which are arranged at two ends of the detection assembly, the detection assembly comprises a detection head and a detection copper plate, the detection head is connected with a lifting cylinder for driving the detection head to move up and down, the first conveying belt and the second conveying belt are arranged at the front end and the rear end of the detection copper plate, a guide structure is arranged on the first conveying belt, a positioning structure, a good product channel and a defective product channel are arranged on the second conveying belt, when the core is positioned at the position of the positioning structure, the middle part of the core is positioned right above the detection copper plate, and the screening assembly is arranged at two sides of the detection copper plate and used for pushing the core to the good product channel or the defective product channel. According to the technology, the performance of the magnetic core is tested through the detection copper plate and the detection head, the magnetic core product is divided into good products and defective products according to the detection result after the detection is finished, and the magnetic core product is conveyed to the next working procedure through the conveying belt for processing. The invention discloses a detecting device and a detecting method for performance of a magnetic core of a fluxgate, wherein the detecting device comprises: the shielding cylinder is used for shielding signals of an external magnetic field and forming a zero-field environment in the shielding cylinder; the rotating unit is used for providing rotating power for the detection platform; the detection platform is used for installing the magnetic core to be detected; the circuit module is used for supplying power to the rotating unit and the detection platform and providing an electric signal path; and the bracket structure is used for providing a supporting structure for the whole fluxgate magnetic core detection device.
The magnetic flux of the magnetic core required by different application scenes is fixed, and the required magnetic core thickness is different. The magnetic core thickness with strictly limited numerical error is required for a fixed application scene, so that the required magnetic flux can be accurately obtained in practical application. However, due to the limitation of the magnetic core production process and the random parameter difference of the individuation of the magnetic core, the thicknesses of the magnetic cores corresponding to the magnetic core products produced by the same magnetic core production line are not necessarily all required, and the magnetic core products which do not meet the requirements need to be screened.
The difference in thickness of the fine core brings about a large change in the magnetic flux, and therefore, accuracy is required for identification and management of the core thickness of the core product. The lack of a technical solution for core thickness identification management that is highly accurate and automated and intelligent while meeting current core product production and identification requirements in the prior art results in the possible influx of partially disqualified core products into the market.
Disclosure of Invention
In order to solve the technical problems in the related art, the invention provides a magnetic core performance optimization management system based on a neural network, which comprises:
the synchronous driving device is used for conveying the magnetic core product which is produced newly to the vision acquisition station in a vertical state when the magnetic core production assembly finishes the production of the magnetic core product every time;
the content acquisition equipment is connected with the synchronous driving equipment and is used for executing visual data acquisition of product delivery environment on the visual acquisition station in the environment where the newly produced magnetic core product is positioned after the synchronous driving equipment finishes the vertical state transmission action of the newly produced magnetic core product every time so as to obtain a corresponding product delivery picture;
the step-by-step conversion device comprises a mean value filtering unit, an equalization processing unit and a data interpolation unit, wherein the equalization processing unit is respectively connected with the mean value filtering unit and the data interpolation unit, the mean value filtering unit is also connected with the content acquisition equipment and is used for performing arithmetic mean value filtering processing on received product factory pictures to obtain and output corresponding arithmetic mean value filtering images, the equalization processing unit is used for performing histogram equalization processing on the received arithmetic mean value filtering images to obtain and output corresponding equalization processing images, and the data interpolation equipment is used for performing nearest neighbor interpolation processing on the received equalization processing images to obtain and output corresponding interpolation conversion images;
the parameter identification device is connected with the step-by-step conversion device and comprises a main control chip, a pattern detection device, a depth analysis device and a thickness judgment device, wherein the main control chip respectively provides configuration services of working data for the pattern detection device, the depth analysis device and the thickness judgment device, the thickness judgment device is respectively connected with the pattern detection device and the depth analysis device, the parameter identification device is used for detecting an imaging pattern of a magnetic core in an interpolation conversion image, analyzing the number of pixel points traversed by the imaging pattern in the thickness direction of the magnetic core, and intelligently analyzing a matching identifier for representing whether the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core by adopting a depth neural network model based on the number of pixel points, each depth data corresponding to each pixel point in the imaging pattern corresponding to the magnetic core and the reference thickness corresponding to the magnetic core, and the learning number of the depth neural network model is positively correlated with the resolution of the interpolation conversion image;
and the error alarm device is connected with the parameter identification device and is used for sending a thickness error detection signal when the received matching identifier indicates that the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and sending a thickness reliable signal when the received matching identifier indicates that the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core.
According to the magnetic core performance optimization management system based on the neural network, the imaging pattern in the interpolation conversion image obtained after the corresponding product delivery picture is optimized in a targeted mode of the magnetic core product can be detected, the number of pixels traversed by the imaging pattern in the thickness direction of the magnetic core is analyzed, and based on the number of pixels, depth data corresponding to each pixel in the imaging pattern corresponding to the magnetic core and reference thickness corresponding to the magnetic core, a matching identifier for intelligently analyzing and representing whether the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core is adopted by the deep neural network model, and the learning number of the deep neural network model is positively correlated with the resolution of the interpolation conversion image, so that the high-precision requirement of the magnetic core thickness identification management is met while the intelligent level and the automatic level of the magnetic core thickness identification management are improved.
Drawings
Embodiments of the present invention will be described below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of a magnetic core structure according to various embodiments of the present invention.
Fig. 2 is a block diagram showing the structure of a neural network-based core performance optimization management system according to embodiment 1 of the present invention.
Fig. 3 is a block diagram showing the structure of a neural network-based core performance optimization management system according to embodiment 2 of the present invention.
Fig. 4 is a block diagram showing the structure of a neural network-based core performance optimization management system according to embodiment 3 of the present invention.
Detailed Description
Embodiments of the neural network-based core performance optimization management system of the present invention will be described in detail below with reference to the accompanying drawings.
Example 1
Fig. 2 is a block diagram showing a neural network-based core performance optimization management system according to embodiment 1 of the present invention, the system including:
the synchronous driving device is used for conveying the magnetic core product which is produced newly to the vision acquisition station in a vertical state when the magnetic core production assembly finishes the production of the magnetic core product every time;
the content acquisition equipment is connected with the synchronous driving equipment and is used for executing visual data acquisition of product delivery environment on the visual acquisition station in the environment where the newly produced magnetic core product is positioned after the synchronous driving equipment finishes the vertical state transmission action of the newly produced magnetic core product every time so as to obtain a corresponding product delivery picture;
the step-by-step conversion device comprises a mean value filtering unit, an equalization processing unit and a data interpolation unit, wherein the equalization processing unit is respectively connected with the mean value filtering unit and the data interpolation unit, the mean value filtering unit is also connected with the content acquisition equipment and is used for performing arithmetic mean value filtering processing on received product factory pictures to obtain and output corresponding arithmetic mean value filtering images, the equalization processing unit is used for performing histogram equalization processing on the received arithmetic mean value filtering images to obtain and output corresponding equalization processing images, and the data interpolation equipment is used for performing nearest neighbor interpolation processing on the received equalization processing images to obtain and output corresponding interpolation conversion images;
the parameter identification device is connected with the step-by-step conversion device and comprises a main control chip, a pattern detection device, a depth analysis device and a thickness judgment device, wherein the main control chip respectively provides configuration services of working data for the pattern detection device, the depth analysis device and the thickness judgment device, the thickness judgment device is respectively connected with the pattern detection device and the depth analysis device, the parameter identification device is used for detecting an imaging pattern of a magnetic core in an interpolation conversion image, analyzing the number of pixel points traversed by the imaging pattern in the thickness direction of the magnetic core, and intelligently analyzing a matching identifier for representing whether the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core by adopting a depth neural network model based on the number of pixel points, each depth data corresponding to each pixel point in the imaging pattern corresponding to the magnetic core and the reference thickness corresponding to the magnetic core, and the learning number of the depth neural network model is positively correlated with the resolution of the interpolation conversion image;
the error alarm device is connected with the parameter identification device and is used for sending a thickness error detection signal when the received matching identifier indicates that the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and sending a thickness reliable signal when the received matching identifier indicates that the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core;
wherein, when the received matching identifier indicates that the current thickness of the magnetic core does not match with the reference thickness corresponding to the magnetic core, executing the thickness error detection signal, and further configured to, when the received matching identifier indicates that the current thickness of the magnetic core matches with the reference thickness corresponding to the magnetic core, execute the thickness reliability signal includes: when the value of the matching mark is 0B01, the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and when the value of the matching mark is 0B00, the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core;
after the synchronous driving device completes the vertical state transmission action of the newly produced magnetic core product, the visual data acquisition of the product factory environment is executed for the environment where the newly produced magnetic core product is located on the visual acquisition station, so as to obtain a corresponding product factory picture, wherein the steps of: the synchronous driving device adopts the falling edge of square wave to realize the representation of the trigger signal of the transmission action of the vertical state of the magnetic core product which is produced up to date every time.
Example 2
Fig. 3 is a block diagram showing the structure of a neural network-based core performance optimization management system according to embodiment 2 of the present invention.
In comparison with fig. 2, the neural network-based core performance optimization management system in fig. 3 may further include:
the area analysis mechanism is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device and is used for respectively measuring the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device;
the area analysis mechanism is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and is used for respectively measuring the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following components: the area analysis mechanism comprises a plurality of area measurement units which are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device;
the area analysis mechanism comprises a plurality of area measurement units, which are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following steps: the area measurement units are visual detectors and are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device;
the area measurement units are visual detectors and are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and the respective measurement comprises the following steps: the plurality of visual detectors are identical in structure;
and wherein the plurality of area measurement units are a plurality of visual detectors, and are configured to be respectively connected to the average filtering unit, the equalization processing unit, the data interpolation unit, and the parameter identification device, so as to complete respective measurement of current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit, and the parameter identification device, and further include: each of the plurality of visual detectors includes a planar camera.
Example 3
Fig. 4 is a block diagram showing the structure of a neural network-based core performance optimization management system according to embodiment 3 of the present invention.
In comparison with fig. 3, the neural network-based core performance optimization management system in fig. 4 may further include:
the information storage device is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the plurality of area measurement units of the parameter identification device and is used for executing the instant storage of the real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device;
the information storage device is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the plurality of area measurement units of the parameter identification device, and is used for executing the real-time storage of the real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following steps: the information storage device is a dynamic storage device;
and an information storage device connected to the average filtering unit, the equalization processing unit, the data interpolation unit, and the plurality of area measurement units of the parameter identification device, respectively, for performing an instant storage of real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit, and the parameter identification device, including: the information storage device is a TF memory card.
In addition, in the magnetic core performance optimization management system based on the neural network, after the synchronous driving device completes the transmission action of the vertical state of the newly produced magnetic core product every time, performing visual data acquisition of the product factory environment on the environment where the newly produced magnetic core product is located on the visual acquisition station, so as to obtain a corresponding product factory image, and further including: the trigger signal is used for triggering the visual data acquisition of the product delivery environment of the environment where the newly produced magnetic core product is located on the visual acquisition station.
The technical innovation and the progress of the invention are as follows:
and (3) a step of: detecting an imaging pattern of a magnetic core in an interpolation conversion image, analyzing the number of pixel points traversed by the imaging pattern in the thickness direction of the magnetic core, intelligently analyzing a matching identifier for representing whether the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core by adopting a depth neural network model based on the number of pixel points, each depth data corresponding to each pixel point in the imaging pattern corresponding to the magnetic core and the reference thickness corresponding to the magnetic core, wherein the learning number of the depth neural network model is positively correlated with the resolution of the interpolation conversion image;
and II: introducing a parameter identification device comprising a main control chip, a pattern detection device, a depth of field analysis device and a thickness judgment device, wherein the main control chip respectively provides configuration services of working data for the pattern detection device, the depth of field analysis device and the thickness judgment device, and the thickness judgment device is respectively connected with the pattern detection device and the depth of field analysis device so as to provide hardware resources for intelligent analysis of whether the current thickness of a magnetic core is matched with the reference thickness corresponding to the magnetic core;
thirdly,: and when the received matching identifier indicates that the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, sending a thickness error detection signal, and when the received matching identifier indicates that the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core, sending a thickness reliable signal, wherein when the value of the matching identifier is 0B01, the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and when the value of the matching identifier is 0B00, the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core.
The magnetic core performance optimization management system based on the neural network, disclosed by the invention, is used for solving the technical problem that a magnetic core thickness identification management mechanism which is high in accuracy, high in automation level and high in intellectualization level and simultaneously meets the production requirement and the identification requirement of the current magnetic core product is lacking in the prior art, detecting the imaging pattern in an interpolation conversion image obtained by the magnetic core product after the corresponding product delivery picture is optimized in a targeted manner, and intelligently analyzing whether the thickness of the current magnetic core product meets the standard or not by adopting a deep neural network model based on various basic data of the imaging pattern and the corresponding reference thickness of the magnetic core, so that the magnetic core performance optimization management system is met while the intellectualization level and the automation level of the magnetic core thickness identification management are improved.
The foregoing description of the exemplary embodiments of the invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations will be apparent to practitioners skilled in the art. The exemplary embodiments were chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling others skilled in the art to understand the invention for various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalents.
Claims (9)
1. A neural network-based magnetic core performance optimization management system, the system comprising:
the synchronous driving device is used for conveying the magnetic core product which is produced newly to the vision acquisition station in a vertical state when the magnetic core production assembly finishes the production of the magnetic core product every time;
the content acquisition equipment is connected with the synchronous driving equipment and is used for executing visual data acquisition of product delivery environment on the visual acquisition station in the environment where the newly produced magnetic core product is positioned after the synchronous driving equipment finishes the vertical state transmission action of the newly produced magnetic core product every time so as to obtain a corresponding product delivery picture;
the step-by-step conversion device comprises a mean value filtering unit, an equalization processing unit and a data interpolation unit, wherein the equalization processing unit is respectively connected with the mean value filtering unit and the data interpolation unit, the mean value filtering unit is also connected with the content acquisition equipment and is used for performing arithmetic mean value filtering processing on received product factory pictures to obtain and output corresponding arithmetic mean value filtering images, the equalization processing unit is used for performing histogram equalization processing on the received arithmetic mean value filtering images to obtain and output corresponding equalization processing images, and the data interpolation equipment is used for performing nearest neighbor interpolation processing on the received equalization processing images to obtain and output corresponding interpolation conversion images;
the parameter identification device is connected with the step-by-step conversion device and comprises a main control chip, a pattern detection device, a depth analysis device and a thickness judgment device, wherein the main control chip respectively provides configuration services of working data for the pattern detection device, the depth analysis device and the thickness judgment device, the thickness judgment device is respectively connected with the pattern detection device and the depth analysis device, the parameter identification device is used for detecting an imaging pattern of a magnetic core in an interpolation conversion image, analyzing the number of pixel points traversed by the imaging pattern in the thickness direction of the magnetic core, and intelligently analyzing a matching identifier for representing whether the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core by adopting a depth neural network model based on the number of pixel points, each depth data corresponding to each pixel point in the imaging pattern corresponding to the magnetic core and the reference thickness corresponding to the magnetic core, and the learning number of the depth neural network model is positively correlated with the resolution of the interpolation conversion image;
and the error alarm device is connected with the parameter identification device and is used for sending a thickness error detection signal when the received matching identifier indicates that the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and sending a thickness reliable signal when the received matching identifier indicates that the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core.
2. The neural network-based core performance optimization management system of claim 1, wherein:
executing the thickness error detection signal when the received matching identification indicates that the current thickness of the magnetic core does not match the reference thickness corresponding to the magnetic core, and further for executing the thickness reliability signal when the received matching identification indicates that the current thickness of the magnetic core matches the reference thickness corresponding to the magnetic core comprises: when the value of the matching mark is 0B01, the current thickness of the magnetic core is not matched with the reference thickness corresponding to the magnetic core, and when the value of the matching mark is 0B00, the current thickness of the magnetic core is matched with the reference thickness corresponding to the magnetic core;
after the synchronous driving device completes the vertical state transmission action of the newly produced magnetic core product, the visual data acquisition of the product factory environment is executed for the environment where the newly produced magnetic core product is located on the visual acquisition station, so as to obtain a corresponding product factory picture, wherein the steps of: the synchronous driving device adopts the falling edge of square wave to realize the representation of the trigger signal of the transmission action of the vertical state of the magnetic core product which is produced up to date every time.
3. The neural network-based core performance optimization management system of claim 2, wherein the system further comprises:
the area analysis mechanism is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device and is used for respectively measuring the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device;
the area analysis mechanism is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and is used for respectively measuring the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following components: the area analysis mechanism comprises a plurality of area measurement units which are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device so as to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device.
4. The neural network-based core performance optimization management system of claim 3, wherein:
the area analysis mechanism comprises a plurality of area measurement units, which are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following steps: the area measuring units are visual detectors and are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device.
5. The neural network-based core performance optimization management system of claim 4, wherein:
the area measurement units are visual detectors and are respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and the respective measurement comprises the following steps: the plurality of visual detectors are identical in structure.
6. The neural network-based core performance optimization management system of claim 5, wherein:
the plurality of area measurement units are a plurality of visual detectors and are used for being respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device so as to finish the respective measurement of the current real-time area values of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and the method further comprises the following steps: each of the plurality of visual detectors includes a planar camera.
7. The neural network-based core performance optimization management system of any of claims 3-6, further comprising:
and the information storage device is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the plurality of area measurement units of the parameter identification device and is used for executing the instant storage of the real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device.
8. The neural network-based core performance optimization management system of claim 7, wherein:
the information storage device is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the plurality of area measurement units of the parameter identification device, and is used for executing the real-time storage of the real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following steps: the information storage device is a dynamic storage device.
9. The neural network-based core performance optimization management system of claim 7, wherein:
the information storage device is respectively connected with the average filtering unit, the equalization processing unit, the data interpolation unit and the plurality of area measurement units of the parameter identification device, and is used for executing the real-time storage of the real-time state parameters of the average filtering unit, the equalization processing unit, the data interpolation unit and the parameter identification device, and comprises the following steps: the information storage device is a TF memory card.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311417724.1A CN117455867A (en) | 2023-10-30 | 2023-10-30 | Magnetic core performance optimization management system based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311417724.1A CN117455867A (en) | 2023-10-30 | 2023-10-30 | Magnetic core performance optimization management system based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117455867A true CN117455867A (en) | 2024-01-26 |
Family
ID=89596140
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311417724.1A Pending CN117455867A (en) | 2023-10-30 | 2023-10-30 | Magnetic core performance optimization management system based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117455867A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118212190A (en) * | 2024-03-14 | 2024-06-18 | 卜筛(上海)信息技术有限公司 | Working performance analysis system of power transmission line |
CN118294595A (en) * | 2024-04-08 | 2024-07-05 | 南京利奕辽机械科技有限公司 | Directional visual data analysis system for stainless steel gear body |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150213587A1 (en) * | 2014-01-30 | 2015-07-30 | Nuflare Technology, Inc. | Inspection apparatus and inspection method |
CN106840258A (en) * | 2017-01-23 | 2017-06-13 | 国网山东省电力公司电力科学研究院 | The full state electromagnetic environment monitoring system of wide area and method based on multi-parameter synergic monitoring |
US20180035606A1 (en) * | 2016-08-05 | 2018-02-08 | Romello Burdoucci | Smart Interactive and Autonomous Robotic Property Maintenance Apparatus, System, and Method |
CN110796697A (en) * | 2019-11-04 | 2020-02-14 | 刘勇 | General surgery ward object size identification platform and method |
CN113095145A (en) * | 2021-03-15 | 2021-07-09 | 南京理工大学 | Hyperspectral anomaly detection deep learning method based on pixel pair matching and double-window discrimination |
CN113379957A (en) * | 2020-02-25 | 2021-09-10 | 李绍辉 | Banknote thickness measurement system one by one |
CN116156127A (en) * | 2022-10-14 | 2023-05-23 | 易正芳 | Image matching system and method |
EP4246458A1 (en) * | 2022-03-17 | 2023-09-20 | Guangzhou Xiaopeng Autopilot Technology Co., Ltd. | System for three-dimensional geometric guided student-teacher feature matching (3dg-stfm) |
-
2023
- 2023-10-30 CN CN202311417724.1A patent/CN117455867A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150213587A1 (en) * | 2014-01-30 | 2015-07-30 | Nuflare Technology, Inc. | Inspection apparatus and inspection method |
US20180035606A1 (en) * | 2016-08-05 | 2018-02-08 | Romello Burdoucci | Smart Interactive and Autonomous Robotic Property Maintenance Apparatus, System, and Method |
CN106840258A (en) * | 2017-01-23 | 2017-06-13 | 国网山东省电力公司电力科学研究院 | The full state electromagnetic environment monitoring system of wide area and method based on multi-parameter synergic monitoring |
CN110796697A (en) * | 2019-11-04 | 2020-02-14 | 刘勇 | General surgery ward object size identification platform and method |
CN113379957A (en) * | 2020-02-25 | 2021-09-10 | 李绍辉 | Banknote thickness measurement system one by one |
CN113095145A (en) * | 2021-03-15 | 2021-07-09 | 南京理工大学 | Hyperspectral anomaly detection deep learning method based on pixel pair matching and double-window discrimination |
EP4246458A1 (en) * | 2022-03-17 | 2023-09-20 | Guangzhou Xiaopeng Autopilot Technology Co., Ltd. | System for three-dimensional geometric guided student-teacher feature matching (3dg-stfm) |
CN116156127A (en) * | 2022-10-14 | 2023-05-23 | 易正芳 | Image matching system and method |
Non-Patent Citations (2)
Title |
---|
余永维;杜柳青;闫哲;许贺作;: "基于深度学习特征的铸件缺陷射线图像动态检测方法", 农业机械学报, no. 07, 12 May 2016 (2016-05-12) * |
刘东编著: "《工业机器视觉:基于闪灵平台的开发及应用》", vol. 1, 30 November 2020, 上海教育出版社, pages: 83 - 84 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118212190A (en) * | 2024-03-14 | 2024-06-18 | 卜筛(上海)信息技术有限公司 | Working performance analysis system of power transmission line |
CN118294595A (en) * | 2024-04-08 | 2024-07-05 | 南京利奕辽机械科技有限公司 | Directional visual data analysis system for stainless steel gear body |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117455867A (en) | Magnetic core performance optimization management system based on neural network | |
CN114994061B (en) | Machine vision-based steel rail intelligent detection method and system | |
CN105510348A (en) | Flaw detection method and device of printed circuit board and detection equipment | |
CN102169156B (en) | Method for detecting high-density electronic circuit fault by using EMScan technology | |
CN103091331A (en) | System and method for visual inspection on burrs and stain defects of radio frequency identification (RFID) antennae | |
CN101303226A (en) | Method for measuring circuit board line width based on largest communication domain | |
CN106651849A (en) | Area-array camera-based PCB bare board defect detection method | |
CN113723325A (en) | Tool defect detection system for prefabricated parts | |
CN107290653B (en) | Detection device and method for identifying PCB based on broadband magnetic induction channel characteristics | |
CN115015286B (en) | Chip detection method and system based on machine vision | |
CN111310727A (en) | Object detection method and device, storage medium and electronic device | |
CN101592620A (en) | Circuit substrate pick-up unit and method | |
CN107967679B (en) | Method for automatically selecting positioning core based on vector graph of PCB product | |
CN105427278A (en) | PCB positioning point determining method and system | |
CN111310402B (en) | Method for detecting defects of bare printed circuit board based on surface-to-surface parallelism | |
CN112634259A (en) | Automatic modeling and positioning method for keyboard keycaps | |
CN109544514B (en) | Sawn timber identity identification method, device and equipment integrating apparent features | |
CN116993654B (en) | Camera module defect detection method, device, equipment, storage medium and product | |
CN111339729A (en) | Balance layout method, equipment and readable storage medium for automatically positioning screw hole position | |
CN113723841A (en) | Online detection method for tool missing in assembly type prefabricated part | |
CN104407276A (en) | Signal acquisition device for laboratory partial discharge tests | |
CN110956640B (en) | Heterogeneous image edge point detection and registration method | |
CN112529902A (en) | Hole checking method of PCB (printed circuit board) | |
CN112561992A (en) | Position determination method and device, storage medium and electronic device | |
KR102129970B1 (en) | Method And Apparatus for Matching inspection Data Electronic Component |
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 |