Intelligent Analysis System for Seamless Steel Tubes Based on Action Collaboration

Abstract

The present invention relates to an intelligent analysis system for seamless steel pipes based on action coordination, including: a model application device for parallelly inputting various visualization parameters of the seamless steel pipe into an artificial intelligence model to obtain the current length value corresponding to the seamless steel pipe output therefrom; a demand judgment device for issuing a on-site pipe length mismatch instruction when the current length value corresponding to the received seamless steel pipe is inconsistent with the designed required length of the seamless steel pipe, or otherwise issuing a on-site pipe length matching instruction. Through the present invention, the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpened image can be parallelly input into the artificial intelligence model to obtain the current length value corresponding to the seamless steel pipe, thereby providing key reference information for whether the seamless steel pipe on site meets the designed length requirement.

Description

Technical Field

[0001] The present invention relates to the field of pipe materials, and particularly to an intelligent analysis system for seamless steel pipes based on action coordination. Background Art

[0002] Seamless steel pipe is made by piercing a whole round steel, and the steel pipe without weld on the surface is called seamless steel pipe. According to the production method, seamless steel pipes can be divided into hot-rolled seamless steel pipes, cold-rolled seamless steel pipes, cold-drawn seamless steel pipes, extruded seamless steel pipes, pipe jacking, etc. According to the cross-sectional shape, seamless steel pipes are divided into two types: circular and special-shaped. Special-shaped pipes have various complex shapes such as square, oval, triangular, hexagonal, sunflower-shaped, star-shaped, and finned pipes. The maximum diameter reaches 900mm, and the minimum diameter is 4mm. According to different uses, there are thick-walled seamless steel pipes and thin-walled seamless steel pipes. Seamless steel pipes are mainly used as oil geological drilling pipes, cracking pipes for petrochemical industry, boiler pipes, bearing pipes, and high-precision structural steel pipes for automobiles, tractors, and aviation.

[0003] However, even among the seamless steel pipes of the same batch produced by the same production line, there are still size differences. Once such differences exceed the range allowed by quality inspection, inferior seamless steel pipe products that cannot be put into market use will be formed. Therefore, it is necessary to conduct customized detection and differentiation on the above-mentioned size differences, such as the length value differences of seamless steel pipes, to identify inferior seamless steel pipe products. Summary of the Invention

[0004] In order to solve the technical problems in the related fields, the present invention provides an intelligent analysis system for seamless steel pipes based on action coordination. By performing multiple learning on a feedforward neural network to obtain a corresponding artificial intelligence model, the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the received sharpened image, the overall depth of field value of the seamless steel pipe in the sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the sharpened image are input into the artificial intelligence model in parallel to run the artificial intelligence model, and the corresponding current length value of the seamless steel pipe output by it is obtained, so as to provide key reference information for whether the seamless steel pipe on site meets the design length requirements.

[0005] According to the present invention, there is provided an intelligent analysis system for seamless steel pipes based on action coordination, the system comprising:

[0006] An action coordination mechanism, respectively connected to the steel pipe pushing mechanism and the wireless camera mechanism, and configured to send a capture start instruction to the wireless camera mechanism once every time it detects that the steel pipe pushing mechanism has completed a pushing operation of a seamless steel pipe to the detection station directly below the wireless camera mechanism;

[0007] A wireless camera is used to perform an image signal capture action on the detection station directly below it every time a capture start instruction is received, so as to obtain and output a corresponding instant capture image;

[0008] A morphological processing device is connected to the wireless camera and is used to perform morphological processing of first dilation and then erosion on the received instant capture image in sequence, so as to obtain and output a corresponding morphological processing image;

[0009] A content filtering device is connected to the morphological processing device and is used to perform Butterworth low-pass filtering processing on the received morphological processing image, so as to obtain and output a corresponding content filtering image;

[0010] A sharpening processing device is connected to the content filtering device and is used to perform sharpening processing based on the USM filter on the received content filtering image, so as to obtain and output a corresponding instant sharpening image;

[0011] A model application device is connected to the sharpening processing device and is used to perform multiple learning on a feedforward neural network to obtain a corresponding artificial intelligence model, and parallelly input the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpening image, the overall depth of field value of the seamless steel pipe in the instant sharpening image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpening image into the artificial intelligence model, so as to run the artificial intelligence model and obtain the corresponding current length value of the seamless steel pipe output by it;

[0012] A demand judgment device is connected to the model application device and is used to issue a on-site pipe length mismatch instruction when the current length value of the seamless steel pipe output by the artificial intelligence model is inconsistent with the designed required length of the seamless steel pipe, otherwise, issue a on-site pipe length matching instruction;

[0013] Among them, performing multiple learning on a feedforward neural network to obtain a corresponding artificial intelligence model, and parallelly inputting the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpening image, the overall depth of field value of the seamless steel pipe in the instant sharpening image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpening image into the artificial intelligence model, so as to run the artificial intelligence model and obtain the corresponding current length value of the seamless steel pipe output by it includes: the total number of times of learning performed on the feedforward neural network is monotonically and positively correlated with the resolution of the image signal captured by the wireless camera.

[0014] The present invention has at least the following three important inventive points:

[0015] Inventive Point A: Introduce a model application device with a targeted design including a network learning device, a multi-parameter analysis device, and a length identification device, which is used to complete the intelligent identification process of the current length value corresponding to the seamless steel pipe by using an artificial intelligence model;

[0016] Inventive Point B: Specifically, perform multiple learning on the feedforward neural network to obtain the corresponding artificial intelligence model, and parallelly input the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the received sharpened image, the overall depth of field value of the seamless steel pipe in the sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the sharpened image into the artificial intelligence model to run the artificial intelligence model and obtain the current length value corresponding to the seamless steel pipe output by it, so as to provide key reference information for whether the seamless steel pipe on site meets the design length requirements;

[0017] Inventive Point C: The total number of times of learning performed on the feedforward neural network is monotonically and positively correlated with the resolution of the image signal captured by the wireless camera, thereby realizing the structural customization of the artificial intelligence model and ensuring the stability and effectiveness of the intelligent analysis results of the artificial intelligence model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The following will describe the implementation embodiments of the present invention in conjunction with the drawings, where:

[0019] Figure 1 FIG. is a schematic internal structure diagram of a seamless steel pipe intelligent analysis system based on action coordination shown according to an implementation embodiment of the present invention.

[0020] Figure 2 FIG. is a schematic internal structure diagram of a seamless steel pipe intelligent analysis system based on action coordination shown according to Implementation Embodiment B of the present invention.

[0021] Figure 3 FIG. is a schematic internal structure diagram of a seamless steel pipe intelligent analysis system based on action coordination shown according to Implementation Embodiment C of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0022] The following will refer to the drawings to detail the implementation embodiments of the seamless steel pipe intelligent analysis system based on action coordination of the present invention.

[0023] Figure 1 FIG. is a schematic internal structure diagram of a seamless steel pipe intelligent analysis system based on action coordination shown according to an implementation embodiment of the present invention for seamless steel pipes, and the system includes:

[0024] The action coordination mechanism is respectively connected to the steel pipe pushing mechanism and the wireless camera mechanism, and is used to send a capture start instruction to the wireless camera mechanism every time it detects that the steel pipe pushing mechanism has completed a pushing operation of a seamless steel pipe to the detection station directly below the wireless camera mechanism;

[0025] Exemplarily, the action coordination mechanism is respectively connected to the steel pipe pushing mechanism and the wireless camera mechanism, and is used to send a capture start instruction to the wireless camera mechanism every time it detects that the steel pipe pushing mechanism has completed a pushing operation of a seamless steel pipe to the detection station directly below the wireless camera mechanism, including: selecting to use an ASIC chip to implement the action coordination mechanism, which is respectively connected to the steel pipe pushing mechanism and the wireless camera mechanism, and is used to send a capture start instruction to the wireless camera mechanism every time it detects that the steel pipe pushing mechanism has completed a pushing operation of a seamless steel pipe to the detection station directly below the wireless camera mechanism;

[0026] The wireless camera mechanism is used to execute an image signal capture action on the detection station directly below it every time it receives a capture start instruction, so as to obtain and output a corresponding instant capture image;

[0027] The morphological processing device is connected to the wireless camera mechanism, and is used to sequentially perform morphological processing of dilation first and then erosion on the received instant capture image, so as to obtain and output a corresponding morphological processing image;

[0028] The content filtering device is connected to the morphological processing device, and is used to perform Butterworth low-pass filtering processing on the received morphological processing image, so as to obtain and output a corresponding content filtering image;

[0029] The sharpening processing device is connected to the content filtering device, and is used to perform sharpening processing based on the USM filter on the received content filtering image, so as to obtain and output a corresponding instant sharpening image;

[0030] The model application device is connected to the sharpening processing device, and is used to perform multiple learning on the feedforward neural network to obtain a corresponding artificial intelligence model, and parallelly input the horizontal coordinate value and vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpening image, the overall depth of field value of the seamless steel pipe in the instant sharpening image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpening image into the artificial intelligence model, so as to run the artificial intelligence model and obtain the corresponding current length value of the seamless steel pipe output by it;

[0031] The demand judgment device is connected to the model application device, and is used to issue a on-site pipe length mismatch instruction when the current length value of the seamless steel pipe output by the artificial intelligence model is inconsistent with the designed required length of the seamless steel pipe, otherwise, issue a on-site pipe length matching instruction;

[0032] Among them, the feedforward neural network is trained multiple times to obtain the corresponding artificial intelligence model. The horizontal coordinate values and vertical coordinate values of each pixel point of the seamless steel pipe in the received instantaneously sharpened image, the overall depth of field value of the seamless steel pipe in the instantaneously sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instantaneously sharpened image are input into the artificial intelligence model in parallel to run the artificial intelligence model. The obtained current length value corresponding to the seamless steel pipe output by it includes: the total number of times of training performed on the feedforward neural network is monotonically and positively correlated with the resolution of the image signal captured by the wireless camera.

[0033] Among them, the feedforward neural network is trained multiple times to obtain the corresponding artificial intelligence model. The horizontal coordinate values and vertical coordinate values of each pixel point of the seamless steel pipe in the received instantaneously sharpened image, the overall depth of field value of the seamless steel pipe in the instantaneously sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instantaneously sharpened image are input into the artificial intelligence model in parallel to run the artificial intelligence model. The obtained current length value corresponding to the seamless steel pipe output by it further includes: after performing numerical normalization processing on the horizontal coordinate values and vertical coordinate values of each pixel point of the seamless steel pipe in the received instantaneously sharpened image, the overall depth of field value of the seamless steel pipe in the instantaneously sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instantaneously sharpened image respectively, they are input into the artificial intelligence model in parallel.

[0034] Figure 2 It is a schematic internal structure diagram of the intelligent analysis system for seamless steel pipes based on action coordination shown in Embodiment B of the present invention.

[0035] Compared with the seamless steel pipe in the embodiment, the intelligent analysis system for seamless steel pipes based on action coordination shown in Embodiment B may further include the following components:

[0036] A humidity measurement mechanism, including a plurality of humidity measurement units, is used to measure the current surface humidity values of the morphological processing device, the content filtering device, the sharpening processing device, and the model application device respectively.

[0037] Among them, the humidity measurement mechanism, including a plurality of humidity measurement units, is used to measure the current surface humidity values of the morphological processing device, the content filtering device, the sharpening processing device, and the model application device respectively, including: the plurality of humidity measurement units respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device are a plurality of non-contact humidity sensors.

[0038] Among them, the humidity measurement mechanism includes multiple humidity measurement units for respectively measuring the current surface humidity values of the morphological processing device, the content filtering device, the sharpening processing device, and the model application device. It also includes: the internal structures of the multiple non-contact humidity sensors respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device are the same;

[0039] Among them, the humidity measurement mechanism includes multiple humidity measurement units for respectively measuring the current surface humidity values of the morphological processing device, the content filtering device, the sharpening processing device, and the model application device. It also includes: the multiple non-contact humidity sensors respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device have the same humidity measurement upper limit threshold and humidity measurement lower limit threshold;

[0040] Among them, the humidity measurement mechanism includes multiple humidity measurement units for respectively measuring the current surface humidity values of the morphological processing device, the content filtering device, the sharpening processing device, and the model application device. It also includes: the distances from the multiple non-contact humidity sensors respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device to the morphological processing device, the content filtering device, the sharpening processing device, and the model application device are equal.

[0041] Figure 3 It is a schematic internal structure diagram of the seamless steel pipe intelligent analysis system based on action coordination shown in Embodiment C of the present invention.

[0042] Compared with the seamless steel pipe of the embodiment, the seamless steel pipe intelligent analysis system based on action coordination shown in Embodiment C may further include the following components:

[0043] An instant notification mechanism is respectively connected to the multiple non-contact humidity sensors respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device, and is used to perform corresponding humidity alarm actions based on the humidity measurement results of the multiple non-contact humidity sensors respectively adopted for the morphological processing device, the content filtering device, the sharpening processing device, and the model application device;

[0044] Among them, the instant notification mechanism is respectively connected to multiple non-contact humidity sensors adopted by the morphological processing device, the content filtering device, the sharpening processing device, and the model application device, and is used to perform corresponding humidity alarm actions based on the humidity measurement results of the multiple non-contact humidity sensors adopted by the morphological processing device, the content filtering device, the sharpening processing device, and the model application device, including: the instant notification mechanism includes a humidity receiving unit, a humidity judging unit, and a notification execution unit built therein;

[0045] And among them, the instant notification mechanism is respectively connected to multiple non-contact humidity sensors adopted by the morphological processing device, the content filtering device, the sharpening processing device, and the model application device, and is used to perform corresponding humidity alarm actions based on the humidity measurement results of the multiple non-contact humidity sensors adopted by the morphological processing device, the content filtering device, the sharpening processing device, and the model application device, and further includes: in the instant notification mechanism, the humidity receiving unit, the humidity judging unit, and the notification execution unit are connected in sequence.

[0046] In addition, in the seamless steel pipe intelligent analysis system based on action coordination, the feedforward neural network is trained multiple times to obtain a corresponding artificial intelligence model, and the horizontal coordinate values and vertical coordinate values of each pixel point of the seamless steel pipe in the received instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpened image are input into the artificial intelligence model in parallel to run the artificial intelligence model, and the obtained current length value of the seamless steel pipe corresponding to the output thereof further includes: running the artificial intelligence model to obtain the current length value of the seamless steel pipe corresponding to the output thereof in a numerically normalized form.

[0047] By using the seamless steel pipe intelligent analysis system based on action coordination of the present invention, aiming at the technical problem that it is difficult to specifically identify the subtle numerical changes in the length of seamless steel pipes in the prior art, by inputting the horizontal coordinate values and vertical coordinate values of each pixel point of the seamless steel pipe in the instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixel points occupied by the seamless steel pipe in the instant sharpened image into the artificial intelligence model in parallel, the current length value of the seamless steel pipe corresponding to it is obtained, so as to provide key reference information for whether the seamless steel pipe on site meets the design length requirements.

[0048] Although the present invention has been described by preferred embodiments, the present invention is not limited to the specific embodiments given, and those skilled in the art can make other embodiments and modifications without departing from the spirit and scope of the present invention.

Claims

1. An intelligent analysis system for seamless steel pipes based on action coordination, characterized in that: The system comprises: The action coordination mechanism is connected to the steel pipe pushing mechanism and the wireless camera mechanism respectively, and is used to send a capture start instruction to the wireless camera mechanism every time it detects that the steel pipe pushing mechanism completes a pushing operation of the seamless steel pipe to the inspection station directly below the wireless camera mechanism; The wireless camera mechanism is used to execute the image signal capturing action of the inspection station directly below it after receiving a capture start instruction, so as to obtain and output the corresponding real-time captured image; A morphological processing device, connected to the wireless camera mechanism, for sequentially performing a morphological processing of first dilation and then erosion on the received instant captured image to obtain and output a corresponding morphologically processed image; a content filtering device connected to the morphological processing device, and configured to perform Butterworth low-pass filtering on the received morphologically processed image to obtain and output a corresponding content filtered image; A sharpening processing device, connected to the content filtering device, for performing a sharpening process based on a USM filter on the received content filtering image to obtain and output a corresponding instant sharpened image; A model application device is connected to the sharpening processing device and is used to perform multiple learning on the feedforward neural network to obtain a corresponding artificial intelligence model, and the horizontal coordinate value and the vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixels occupied by the seamless steel pipe in the instant sharpened image are input to the artificial intelligence model in parallel to run the artificial intelligence model to obtain the current length value corresponding to the seamless steel pipe in the normalized form of the output of the artificial intelligence model; A demand judgment device is connected to the model application device, and is used to issue an on-site pipe length mismatch instruction when the current length value corresponding to the seamless steel pipe output by the artificial intelligence model is inconsistent with the design required length of the seamless steel pipe, otherwise, issue an on-site pipe length matching instruction; wherein, performing multiple learning on the feedforward neural network to obtain a corresponding artificial intelligence model, inputting the horizontal coordinate value and the vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixels occupied by the seamless steel pipe in the instant sharpened image into the artificial intelligence model in parallel, so as to run the artificial intelligence model, and obtaining the current length value corresponding to the seamless steel pipe outputted by the artificial intelligence model includes: the total number of learning performed on the feedforward neural network is monotonically positively correlated with the resolution of the image signal captured by the wireless camera mechanism; The feedforward neural network is subjected to multiple learning to obtain a corresponding artificial intelligence model, and the horizontal coordinate value and the vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixels occupied by the seamless steel pipe in the instant sharpened image are input into the artificial intelligence model in parallel to run the artificial intelligence model to obtain the current length value corresponding to the seamless steel pipe outputted by the model, which also includes: performing numerical normalization processing on the horizontal coordinate value and the vertical coordinate value of each pixel point of the seamless steel pipe in the received instant sharpened image, the overall depth of field value of the seamless steel pipe in the instant sharpened image, and the total number of pixels occupied by the seamless steel pipe in the instant sharpened image, and then inputting them into the artificial intelligence model in parallel.

2. The seamless steel pipe intelligent analysis system based on action coordination according to claim 1 is characterized in that: The system further comprises: A humidity measuring mechanism, comprising a plurality of humidity measuring units, for respectively measuring current surface humidity values ​​of the morphological processing device, the content filtering device, the sharpening processing device and the model application device; Among them, the humidity measurement mechanism includes multiple humidity measurement units, which are used to respectively measure the current surface humidity values ​​of the morphological processing device, the content filtering device, the sharpening processing device and the model application device, including: the multiple humidity measurement units respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device are multiple non-contact humidity sensors.

3. The seamless steel pipe intelligent analysis system based on action coordination according to claim 2 is characterized in that: The humidity measuring mechanism includes a plurality of humidity measuring units for respectively measuring the current surface humidity values ​​of the morphological processing device, the content filtering device, the sharpening processing device and the model application device. The mechanism also includes: the internal structures of the plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device are the same.

4. The seamless steel pipe intelligent analysis system based on action coordination according to claim 2 is characterized in that: The humidity measuring mechanism includes a plurality of humidity measuring units for respectively measuring the current surface humidity values ​​of the morphological processing device, the content filtering device, the sharpening processing device and the model application device, and further includes: a plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device have the same humidity measurement upper limit threshold and humidity measurement lower limit threshold.

5. The seamless steel pipe intelligent analysis system based on action coordination as claimed in claim 4 is characterized in that: The humidity measuring mechanism includes a plurality of humidity measuring units for respectively measuring the current surface humidity values ​​of the morphological processing device, the content filtering device, the sharpening processing device and the model application device, and further includes: a plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device are at equal distances from the morphological processing device, the content filtering device, the sharpening processing device and the model application device.

6. The seamless steel pipe intelligent analysis system based on action coordination according to any one of claims 2 to 5, characterized in that: The system further comprises: The instant notification mechanism is respectively connected to a plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device, and is used to perform corresponding humidity alarm actions based on the humidity measurement results of the plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device.

7. The seamless steel pipe intelligent analysis system based on action coordination according to claim 6 is characterized in that: The instant notification mechanism is respectively connected to a plurality of non-contact humidity sensors respectively used by the morphological processing device, the content filtering device, the sharpening processing device and the model application device, and is used to perform corresponding humidity alarm actions based on humidity measurement results of the plurality of non-contact humidity sensors respectively used by the morphological processing device, the content filtering device, the sharpening processing device and the model application device, including: the instant notification mechanism has a built-in humidity receiving unit, a humidity judgment unit and a notification execution unit.

8. The seamless steel pipe intelligent analysis system based on action coordination according to claim 7 is characterized in that: An instant notification mechanism is respectively connected to a plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device, and is used to perform corresponding humidity alarm actions based on humidity measurement results of the plurality of non-contact humidity sensors respectively used for the morphological processing device, the content filtering device, the sharpening processing device and the model application device. The instant notification mechanism also includes: within the instant notification mechanism, the humidity receiving unit, the humidity judgment unit and the notification execution unit are connected in sequence.

Images