A gas meter production line transmission equipment operation detection system based on image recognition

Abstract

The present invention relates to the technical field of conveying equipment operation detection, and specifically to a gas meter production line conveying equipment operation detection system based on image recognition, comprising a position information acquisition unit, an electronic signal acquisition unit, an equipment image acquisition unit, an aging assessment unit, a fault assessment unit, an operation early warning unit and an operation analysis and processing unit; the present invention collects images and analyzes the collected images to judge the operation status of gas meter components on the conveying equipment, obtains control parameters through analysis as the basis for controlling the operation of the conveying equipment, appropriately changes the operation speed of the conveying equipment to meet the production requirements, and ensures the smooth operation of the conveying equipment; warns of the failure rate and aging rate, and if they exceed the preset allowable range, promptly issues an alarm message to reduce the probability of production accidents, and can promptly maintain the equipment; otherwise, the failure rate and aging rate are introduced as influencing factors of the control parameters to ensure the normal operation of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of conveying equipment operation detection, and in particular to a gas meter production line conveying equipment operation detection system based on image recognition. Background Art

[0002] Gas meters are commonly used instruments in daily life. The production of gas meters is very strict. Their quality and accuracy are related to the vital interests and safety of every family. Therefore, the current production of gas meters relies heavily on automated production. The conveying equipment of the gas meter production line generally includes conveyor belts, automated handling systems, detection devices, etc. In order to ensure the smooth operation of the entire gas meter production line, the conveying equipment of the gas meter production line needs to be kept in good condition so that it can continuously and stably convey gas meter components to eventually form finished products.

[0003] The conveying equipment of the gas meter production line is a kind of equipment that is used for a long time and has wear and tear. Therefore, in order to prevent unexpected problems from occurring in the conveying equipment of the gas meter production line, it is necessary to detect the operation of the conveying equipment to avoid accumulation and offset of conveyor belt products, and reduce the probability of problems such as setbacks and stops in the conveying equipment. In the actual production process, the equipment needs to be shut down for maintenance regularly, but shutdown maintenance will cause the entire production line to stagnate, which has a great impact on the production progress. Therefore, real-time adjustment and detection are also required during the operation of the equipment to ensure the smooth progress of production and provide early warning information. At present, the existing technology cannot effectively detect the conveying equipment of the gas meter production line. In view of the above technical defects, a solution is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a gas meter production line conveying equipment operation detection system based on image recognition to solve the above-mentioned technical defects. The present invention identifies the gas meter components on the conveying equipment through image recognition, and judges the angles between the gas meter components to prevent the gas meter components on the conveying equipment from being arranged closely and affecting subsequent processes; and detects the placement of the gas meter components on the conveying equipment, and issues an early warning if it does not meet the requirements so that the equipment can be adjusted in time; when adjusting the operating speed of the conveying equipment, the failure rate and aging rate of the conveying equipment are taken into consideration, so that the conveying equipment can work more smoothly and reduce the failure probability and aging rate, and when the failure probability and aging rate of the conveying equipment are higher than the preset threshold, the operation analysis and processing unit generates an alarm message to remind the operation and maintenance personnel to perform equipment maintenance and further inspection to avoid causing greater losses.

[0005] The object of the present invention can be achieved by the following technical solution: a gas meter production line conveying equipment operation detection system based on image recognition, wherein the gas meter production line conveying equipment includes a conveying device and a mechanical arm; the mechanical arm is used to grab the gas meter components and place them on the conveying device;

[0006] The gas meter production line conveying equipment operation detection system mentioned in the present invention includes a position information acquisition unit, an electronic signal acquisition unit, an equipment image acquisition unit, an aging evaluation unit, a fault evaluation unit, an operation warning unit and an operation analysis and processing unit.

[0007] When the conveying equipment of the gas meter production line is running, the robotic arm needs to grab the gas meter components and place them on the conveying device for transmission. In actual operation, the robotic arm will only grab the components after recognizing the grabbing signal. Therefore, the time when the robotic arm grabs the gas meter components and places them is not necessarily consistent. Therefore, there are also differences in the distance between two adjacent gas meter components on the conveying device. In order to prevent the gas meter components on the conveying device from being too concentrated or too dispersed, timely detection and early warning are required to timely adjust the speed of the conveying device to ensure the stable operation of the gas meter production line.

[0008] The position information acquisition unit of the gas meter production line conveying equipment operation detection system mentioned in the present invention is used to collect the gas meter position image data and gas meter status image data on the gas meter production line conveying equipment, and then send the gas meter position image data to the operation warning unit, and send the gas meter status image data to the fault assessment unit; the operation warning unit calculates the gas meter position image data to obtain the gas meter position angle E, and compares the gas meter position angle E with the set angle threshold L to obtain the operation risk coefficient ZS, and sends the operation risk coefficient ZS to the operation analysis processing unit; the operation analysis processing unit obtains the operation status level CM according to the operation risk coefficient ZS; the fault assessment unit compares the gas meter status with the standard status, calculates the fault occurrence rate Ry, and determines whether the requirements are met. If the requirements are not met, a fault signal Gb is generated, and the fault signal Gb is sent to the operation analysis processing unit; wherein, the gas meter position angle E is the plane angle formed by the line connecting the coordinates of two adjacent gas meter components and the coordinates of the first image acquisition camera.

[0009] The electronic signal acquisition unit of the gas meter production line conveying equipment operation detection system mentioned in the present invention is used to collect the operating speed Vt data, vibration frequency F data, and motor speed Vr data of the gas meter production line conveying equipment, and send the operating speed Vt data, vibration frequency F data, and motor speed Vr data to the operation analysis and processing unit; the analysis and processing unit obtains the frequency modulation parameter CL according to the vibration frequency F.

[0010] The equipment image acquisition unit of the gas meter production line conveying equipment operation detection system mentioned in the present invention is used to collect image data of the end of the gas meter production line conveying equipment, and send the collected image data of the end of the gas meter production line conveying equipment to the aging evaluation unit. The aging evaluation unit analyzes the image data of the end of the gas meter production line conveying equipment, calculates the aging rate Rl, and identifies whether there is a defect. If there is a defect, an aging signal Gc is generated, and the aging signal Gc is sent to the operation analysis processing unit.

[0011] The operation analysis and processing unit of the gas meter production line transportation equipment operation detection system mentioned in the present invention generates a control signal and / or alarm information according to the fault occurrence rate Ry, the aging rate Rl, and the operation status level CM; wherein the control signal is used to adjust the motor speed Vr, and the alarm information is displayed through the terminal, wherein the control signal includes a control parameter K, and the control parameter K is used to calculate the operating speed Vm.

[0012] The position information acquisition unit includes a first image acquisition camera, which is located directly above the gas meter production line conveying equipment. The operation process of the position information acquisition unit is as follows:

[0013] An angle threshold L is set, and the first image acquisition camera acquires image data of the gas meter assembly on the transmission device to form gas meter position image data, and sends the acquired gas meter position image data to the operation warning unit, and the operation warning unit divides the gas meter position image data into n independent position coordinate data, where n is a positive integer; the position coordinate data includes three-dimensional coordinate data of adjacent gas meter assemblies; the operation warning unit calculates the position angle E of two adjacent gas meters in the position coordinate data, and then compares the obtained gas meter position angle E with the set angle threshold L to obtain the operation risk coefficient , i=1~n.

[0014] The operation analysis processing unit compares the operation risk factor ZS to see if it is within the preset risk factor threshold. If the operating risk factor ZS exceeds the risk factor threshold, , the operation analysis and processing unit generates an alarm message to notify the operation and maintenance personnel to perform timely maintenance; if the operation risk factor ZS does not exceed the risk factor threshold , then the operation analysis processing unit obtains the operation status level CM according to the operation risk coefficient ZS; if the operation risk coefficient ZS is less than 1, then , where Y is the state parameter, Y is greater than or equal to 2; if the operation risk coefficient ZS is greater than 1, then CM=1; the operation warning unit sends the operation state level CM to the operation analysis processing unit, and the operation analysis processing unit generates a control signal for adjusting the motor speed Vr to reach the calculated operation speed Vm.

[0015] In order to evaluate the status of the gas meter component on the transmission device, the first image acquisition camera collects gas meter status image data and sends the gas meter status image data to the fault assessment unit; the fault assessment unit annotates the gas meter status image data, adopts the pre-trained convolutional neural network model ResNet, uses the annotated gas meter status image data to train the convolutional neural network model ResNet, and adjusts the hyperparameters; then uses the cross entropy loss function to measure the difference between the model evaluation and the true label; the trained model is integrated into the fault assessment unit to realize real-time image analysis.

[0016] The collected gas meter status image data is divided into independent image data, which only contains one gas meter component. Then, the collected gas meter status image data is annotated as normal, offset, and tilted respectively. The annotated data is divided into a training set and a test set, where the training set accounts for 80% of the annotated data and the rest is a test set. The training set is used to adjust the hyperparameters of the convolutional neural network model ResNet. The adjusted convolutional neural network model ResNet is then used for evaluation to obtain the model-evaluated fault rate Ry. The model-evaluated fault rate Ry is compared with the true label fault rate Rr, and the cross-entropy loss function is used to measure the difference between the model evaluation and the true label, where the cross-entropy loss function is , the smaller L is, the closer the evaluation of the convolutional neural network model ResNet is to the real distribution; when L is less than or equal to the set threshold, the trained model is integrated into the fault evaluation unit; the detection time is set to Tb, and the fault occurrence rate Ry is calculated within the time range of Tb. If the fault occurrence rate Ry is greater than the threshold Ra, a fault signal Gb is generated and sent to the operation analysis processing unit, and the operation analysis processing unit generates an alarm message; if the fault occurrence rate Ry is less than the threshold Ra, the fault occurrence rate Ry is sent to the operation analysis processing unit.

[0017] The electronic signal acquisition unit of the gas meter production line conveying equipment operation detection system mentioned in the present invention includes a sensor for collecting operation speed Vt data, vibration frequency F data, and motor speed Vr data; when the collected vibration frequency F data is sent to the operation analysis processing unit, if the vibration frequency F is greater than the set vibration frequency threshold, the operation analysis processing unit generates a feedback signal and sends the feedback signal to the fault assessment unit;

[0018] After receiving the feedback signal, the fault assessment unit sets the detection time again and calculates the fault occurrence rate Ry within the corresponding time. If the recalculated fault occurrence rate Ry is less than the threshold value Ra, the vibration frequency threshold is reset. If the recalculated fault occurrence rate Ry is greater than the threshold value Ra, a fault signal Gb is generated and sent to the operation analysis and processing unit. The operation analysis and processing unit generates an alarm message to notify the maintenance personnel to maintain the equipment. After maintenance, the above steps are repeated until the vibration frequency F is less than the set vibration frequency threshold or the fault occurrence rate Ry is less than the threshold value Ra.

[0019] The equipment image acquisition unit of the gas meter production line conveying equipment operation detection system mentioned in the present invention includes a second image acquisition camera, which collects data on the end image of the gas meter production line conveying equipment and sends it to the aging assessment unit; the aging assessment unit divides the collected gas meter production line conveying equipment end image data into M sub-areas, and then extracts features from the M sub-area data respectively, establishes a grayscale co-occurrence matrix GL, calculates the contrast Dt and entropy value St of the M sub-areas respectively according to the grayscale co-occurrence matrix GL, and then calculates the aging number m; calculates the contrast Dt multiplied by the entropy value St to obtain the value SD, calculates the product of the contrast Dt and the entropy value St of the standard image to obtain the value LD, if SD is greater than LD, it is considered that the sub-area has a defect and is included in the aging number m; if SD is less than LD, it is considered that the sub-area is normal and is not included in the aging number m; calculates the aging rate Rl=m / M*100%; the aging assessment unit generates an aging signal Gc, and sends the aging signal Gc and the aging rate Rl to the operation analysis processing unit, if Rl is greater than the threshold Rb, the operation analysis processing unit generates an alarm message.

[0020] Establish the gray-level co-occurrence matrix GL and calculate the contrast Dt and entropy value St. The specific steps are as follows:

[0021] The M sub-region data are grayed out separately to create a matrix with a size of N*N, where N is the number of gray levels and the contrast ;

[0022] Entropy ;

[0023] Each element (i, j) represents the number of occurrences of a pixel pair with grayscale values ​​i and j in a specified direction and distance; A decimal number greater than 0.

[0024] Among them, the frequency modulation parameters ,in is the standard vibration frequency;

[0025] Control parameters ,in is the current running speed, is the ideal operating speed, Ry is the failure rate, and Rl is the aging rate;

[0026] , , are weight coefficients and are all greater than 0. ;

[0027] The operation analysis processing unit adjusts the motor speed Vr so that the gas meter production line conveying equipment reaches the calculated operating speed Vm. .

[0028] The beneficial effects of the present invention are as follows:

[0029] (1) The present invention collects and analyzes images to determine the operating conditions of the gas meter components on the conveying equipment. The control parameters obtained through analysis are used as the basis for controlling the operation of the conveying equipment. The operating speed of the conveying equipment is appropriately changed to meet production requirements, thereby ensuring the smooth operation of the conveying equipment.

[0030] (2) The present invention collects images to identify the abnormal arrangement of gas meter components on the conveying equipment and the aging of the conveying equipment, and issues early warnings for the failure rate and aging rate. If the failure rate exceeds the preset allowable range, an alarm message is issued in a timely manner to reduce the probability of production accidents and to perform maintenance on the equipment in a timely manner. If the failure rate and aging rate do not exceed the preset allowable range, the failure rate and aging rate are introduced as influencing factors of the control parameters. While ensuring the normal operation of the equipment, the loss of the equipment is reduced, thereby reducing the probability of problems such as setbacks and stops in the conveying equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention will be further described below in conjunction with the accompanying drawings;

[0032] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Embodiment 1:

[0035] The present invention is a gas meter production line transportation equipment operation detection system based on image recognition, a gas meter production line transportation equipment operation detection system based on image recognition, including a position information acquisition unit, an electronic signal acquisition unit, an equipment image acquisition unit, an aging assessment unit, a fault assessment unit, an operation warning unit and an operation analysis processing unit; wherein the position information acquisition unit is unidirectionally communicated with the operation warning unit and the fault assessment unit respectively; the operation warning unit is unidirectionally communicated with the operation analysis processing unit; the electronic signal acquisition unit is unidirectionally communicated with the operation analysis processing unit; the equipment image acquisition unit is unidirectionally communicated with the aging assessment unit, the aging assessment unit is unidirectionally communicated with the operation analysis processing unit; the fault assessment unit is bidirectionally communicated with the operation analysis processing unit.

[0036] When the gas meter production line conveying equipment is running, the robotic arm needs to grab the gas meter components and place them on the conveying device for transmission. The conveying device used in this embodiment is a conveyor belt; in actual operation, the robotic arm only grabs after recognizing the grabbing signal. Therefore, the time when the robotic arm grabs the gas meter components and places them is not necessarily consistent. Therefore, there are also differences in the distance between two adjacent gas meter components on the conveying device. In order to prevent the gas meter components on the conveying device from being too concentrated or too dispersed, timely detection and early warning are required to timely adjust the speed of the conveying device to ensure the stable operation of the gas meter production line.

[0037] In this embodiment, the position information acquisition unit is used to collect the gas meter position image data and gas meter status image data on the gas meter production line conveying equipment, and then send the gas meter position image data to the operation warning unit, and send the gas meter status image data to the fault assessment unit; the operation warning unit calculates the gas meter position image data to obtain the gas meter position angle E, and compares the gas meter position angle E with the set angle threshold L to obtain the operation risk coefficient ZS, and sends the operation risk coefficient ZS to the operation analysis processing unit; wherein, the gas meter position angle E is the plane angle formed by the line connecting the coordinates of two adjacent gas meter components and the coordinates of the first image acquisition camera;

[0038] The method for obtaining the gas meter position angle E is as follows:

[0039] S1. Calibrate the first image acquisition camera to obtain the camera's intrinsic and extrinsic parameters, specifically: shoot a set of images including a calibration plate, where the calibration plate used in this embodiment is a checkerboard pattern plate; use OpenCV tools to calibrate the camera to obtain the camera's intrinsic and extrinsic parameters;

[0040] S2. Target object detection and positioning: Identify and locate the target object in the image, specifically by using the convolutional neural network YOLO model to detect the target object, then obtain the pixel coordinates of the target object in the image, extract the bounding box coordinates of the target object, and calculate the center point coordinates (x, y) of the target object. , ; where x min , x max are the two extreme values ​​of the horizontal coordinate of the target object, x min ,y max are the two extreme values ​​of the ordinate of the target object;

[0041] S3, spatial coordinate calculation: determine the three-dimensional spatial coordinates of the target object, specifically: obtain the actual size of the known target object, and use the external parameters of the camera to calculate the three-dimensional coordinates of the object relative to the camera; in this embodiment, the target object is the shell of the gas meter, and the size is relatively fixed, so it is easy to perform spatial coordinate calculation;

[0042] S4, vector calculation: calculate the vector from the camera to the target object: A = (x1, y1, z1), B = (x2, y2, z2);

[0043] S5. Calculate the gas meter position angle E: Use the formula: ,

[0044] in, ;

[0045] ; ;

[0046] Then E=arccos(cos(E)).

[0047] The position information acquisition unit includes a first image acquisition camera, which is located in the middle above the gas meter production line conveying equipment. The specific steps are as follows:

[0048] An angle threshold L is set. In this embodiment, the angle threshold L is set to 30°. The first image acquisition camera acquires image data of the gas meter assembly on the transmission device to form gas meter position image data, and sends the acquired gas meter position image data to the operation warning unit. The operation warning unit divides the gas meter position image data into n independent position coordinate data, where n is a positive integer. In this embodiment, the gas meter position image data includes 4 gas meter assembly images, and the adjacent gas meter assemblies can be divided into 3 independent position coordinate data, where n=3.

[0049] The position coordinate data only contains the three-dimensional coordinate data of adjacent gas meter components, that is, A=(x1, y1, z1), B=(x2, y2, z2); the operation warning unit calculates the position angle E of two adjacent gas meters in the position coordinate data, and then compares the obtained gas meter position angle E with the set angle threshold L to obtain the operation risk coefficient , i=1~n.

[0050] The operation analysis processing unit obtains the operation status level CM according to the operation risk coefficient ZS: Specifically, the operation analysis processing unit compares whether the operation risk coefficient ZS is within the preset risk coefficient threshold value. Within the range, when the operating risk factor ZS does not exceed the risk factor threshold , then the operation analysis processing unit obtains the operation status level CM according to the operation risk coefficient ZS; if the operation risk coefficient ZS is less than 1, then , where Y is a state parameter, Y is greater than or equal to 2; if the operation risk coefficient ZS is greater than 1, CM=1; in this embodiment, the ZB range is 0.75 to 1.42; the operation warning unit sends the operation state level CM to the operation analysis processing unit, and the operation analysis processing unit generates a control signal for adjusting the motor speed Vr to reach the calculated operation speed Vm.

[0051] The fault assessment unit compares the gas meter status with the standard status, calculates the fault occurrence rate Ry, and determines whether the requirements are met. If the requirements are not met, a fault signal Gb is generated and sent to the operation analysis processing unit.

[0052] In order to evaluate the status of the gas meter component on the transmission device, the first image acquisition camera collects gas meter status image data and sends the gas meter status image data to the fault assessment unit; the fault assessment unit annotates the gas meter status image data, adopts the pre-trained convolutional neural network model ResNet, uses the annotated gas meter status image data to train the convolutional neural network model ResNet, and adjusts the hyperparameters; then uses the cross entropy loss function to measure the difference between the model evaluation and the true label; the trained model is integrated into the fault assessment unit to realize real-time image analysis.

[0053] Among them, the collected gas meter status image data is divided into independent image data, and the image data only contains one gas meter component, and then it is annotated according to the collected gas meter status image data, and is respectively annotated as normal, offset, and tilted; the annotated data is divided into a training set and a test set, the training set accounts for 80% of the annotated data, and the rest is a test set; the training set is used to adjust the hyperparameters of the convolutional neural network model ResNet; the adjusted convolutional neural network model ResNet is used for evaluation, the collected gas meter status image data is identified, and the number of normal, offset, and tilt is evaluated, and then the fault incidence rate Ry evaluated by the model is obtained, and the fault incidence rate Ry is the ratio of the sum of the offset and tilt numbers obtained by evaluation to the total evaluation number, and the total evaluation number is the sum of the normal, offset, and tilt numbers.

[0054] The failure rate Ry of the model evaluation is compared with the failure rate Rr of the true label, and the cross entropy loss function is used to measure the difference between the model evaluation and the true label, where the cross entropy loss function is , the smaller L is, the closer the evaluation of the convolutional neural network model ResNet is to the real distribution; when L is less than or equal to the set threshold, the trained model is integrated into the fault evaluation unit; the detection time is set to Tb, and the fault occurrence rate Ry is calculated within the time range of Tb. When the fault occurrence rate Ry is less than the threshold Ra, the fault occurrence rate Ry is sent to the operation analysis processing unit, which is used as one of the bases for the operation analysis processing unit to obtain the control parameter K.

[0055] In this embodiment, the electronic signal acquisition unit is used to collect the running speed Vt data, vibration frequency F data, and motor speed Vr data of the gas meter production line conveying equipment, and send the running speed Vt data, vibration frequency F data, and motor speed Vr data to the operation analysis and processing unit; the analysis and processing unit obtains the frequency modulation parameter CL according to the vibration frequency F, and the frequency modulation parameter ,in is the standard vibration frequency.

[0056] In this embodiment, the equipment image acquisition unit is used to collect image data of the end of the gas meter production line conveying equipment, and send the collected image data of the end of the gas meter production line conveying equipment to the aging assessment unit. The aging assessment unit analyzes the image data of the end of the gas meter production line conveying equipment, calculates the aging rate Rl, and identifies whether there is a defect. If there is a defect, an aging signal Gc is generated, and the aging signal Gc is sent to the operation analysis processing unit.

[0057] Among them, the equipment image acquisition unit includes a second image acquisition camera, which collects data of the end image of the gas meter production line conveying equipment and sends it to the aging assessment unit; the aging assessment unit divides the collected gas meter production line conveying equipment end image data into M sub-areas, and then extracts features of the M sub-area data respectively, establishes a gray level co-occurrence matrix GL, calculates the contrast Dt and entropy value St of the M sub-areas respectively according to the gray level co-occurrence matrix GL, and then calculates the aging number m; calculates the contrast Dt multiplied by the entropy value St to obtain the value SD, calculates the product of the contrast Dt and the entropy value St of the standard image to obtain the value LD, if SD is greater than LD, it is considered that the sub-area has a defect and is included in the aging number m; if SD is less than LD, it is considered that the sub-area is normal and is not included in the aging number m; calculates the aging rate Rl=m / M*100%; the aging assessment unit generates an aging signal Gc, and sends the aging signal Gc and the aging rate Rl to the operation analysis processing unit, and the aging rate Rl is used as one of the bases for the operation analysis processing unit to obtain the control parameter K.

[0058] Establish the gray-level co-occurrence matrix GL and calculate the contrast Dt and entropy value St. The specific steps are as follows:

[0059] The M sub-region data are grayed out separately to create a matrix with a size of N*N, where N is the number of gray levels and the contrast ;

[0060] Entropy ;

[0061] Each element (i, j) represents the number of occurrences of a pixel pair with grayscale values ​​i and j in a specified direction and distance; is a decimal greater than 0. It is 0.236.

[0062] In this embodiment, if the fault occurrence rate Ry, the aging rate Rl, and the operation risk coefficient ZS all exceed the preset threshold, the operation analysis processing unit generates a control signal according to the fault occurrence rate Ry, the aging rate Rl, and the operation state level CM; wherein the control signal is used to adjust the motor speed Vr, and the control signal includes a control parameter K, and the control parameter K is used to calculate the operation speed Vm;

[0063] Control parameters ,in is the current running speed, is the ideal operating speed, Ry is the failure rate, and Rl is the aging rate;

[0064] , , are weight coefficients and are all greater than 0. ;

[0065] The operation analysis processing unit adjusts the motor speed Vr so that the gas meter production line conveying equipment reaches the calculated operating speed Vm. .

[0066] Implementation 2:

[0067] The difference between this embodiment and the first embodiment is that:

[0068] The operation analysis processing unit compares the operation risk factor ZS to see if it is within the preset risk factor threshold. If the operating risk factor ZS exceeds the risk factor threshold, , the operation analysis and processing unit generates an alarm message to notify the operation and maintenance personnel to perform timely maintenance, and the operation analysis and processing unit does not calculate Vm.

[0069] Implementation three:

[0070] The difference between this embodiment and the first embodiment is that:

[0071] When the fault occurrence rate Ry is greater than the threshold Ra, a fault signal Gb is generated and sent to the operation analysis processing unit, which generates an alarm message to notify the operation and maintenance personnel to perform timely maintenance. The operation analysis processing unit does not calculate Vm.

[0072] When the collected vibration frequency F data is sent to the operation analysis processing unit, if the vibration frequency F is greater than the set vibration frequency threshold, the operation analysis processing unit generates a feedback signal and sends the feedback signal to the fault assessment unit;

[0073] After receiving the feedback signal, the fault assessment unit sets the detection time again and calculates the fault occurrence rate Ry within the corresponding time. If the recalculated fault occurrence rate Ry is less than the threshold value Ra, the vibration frequency threshold is reset. If the recalculated fault occurrence rate Ry is greater than the threshold value Ra, a fault signal Gb is generated and sent to the operation analysis and processing unit. The operation analysis and processing unit generates an alarm message to notify the maintenance personnel to maintain the equipment. After maintenance, the above steps are repeated until the vibration frequency F is less than the set vibration frequency threshold or the fault occurrence rate Ry is less than the threshold value Ra.

[0074] Implementation Four:

[0075] The difference between this embodiment and the first embodiment is that:

[0076] The aging evaluation unit calculates the aging rate Rl. When Rl is greater than the threshold Rb, the operation analysis processing unit generates an alarm message to notify the operation and maintenance personnel to perform timely maintenance. The operation analysis processing unit does not calculate Vm.

[0077] From the above embodiments, the present invention collects images and analyzes the collected images to determine the operation status of the gas meter component on the conveying equipment, obtains the control parameters through analysis as the basis for controlling the operation of the conveying equipment, appropriately changes the operation speed of the conveying equipment to meet the production requirements, and ensures the smooth operation of the conveying equipment;

[0078] The present invention collects images to identify abnormal arrangements of gas meter components on the conveying equipment and the aging conditions of the conveying equipment, and issues early warnings for the failure rate and the aging rate. If they exceed a preset allowable range, an alarm message is issued in a timely manner to reduce the probability of production accidents, and the equipment can be maintained in a timely manner. If they do not exceed the preset allowable range, the failure rate and the aging rate are introduced as influencing factors of the control parameters, which reduces the loss of the equipment while ensuring the normal operation of the equipment, thereby reducing the probability of problems such as setbacks and stops in the conveying equipment.

[0079] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0080] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula that is close to the actual value. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited to this. Any technical personnel familiar with the technical field within the technical scope disclosed by the present invention, according to the technical solution and the inventive concept of the present invention, make equivalent replacement or change, which should be covered within the protection scope of the present invention.

Claims

1. A gas meter production line conveying equipment operation detection system based on image recognition, characterized in that: It includes a position information acquisition unit, an electronic signal acquisition unit, an equipment image acquisition unit, an aging assessment unit, a fault assessment unit, an operation early warning unit, and an operation analysis and processing unit; The position information acquisition unit is used to collect gas meter position image data and gas meter status image data on the gas meter production line conveying equipment, and then send the gas meter position image data to the operation warning unit, and send the gas meter status image data to the fault assessment unit; the operation warning unit calculates the gas meter position image data to obtain the gas meter position angle E, and the gas meter position angle E is the plane angle formed by the line connecting the coordinates of two adjacent gas meter components and the coordinates of the first image acquisition camera; the gas meter position angle E is compared with the set angle threshold L to obtain the operation risk coefficient ZS, and the operation risk coefficient ZS is sent to the operation analysis processing unit; The operation analysis processing unit obtains the operation state level CM according to the operation risk coefficient ZS; the fault assessment unit compares the gas meter state with the standard state, calculates the fault occurrence rate Ry, and determines whether the requirements are met. If the requirements are not met, a fault signal Gb is generated and sent to the operation analysis processing unit; The electronic signal acquisition unit is used to collect the running speed Vt data, vibration frequency F data, and motor speed Vr data of the gas meter production line conveying equipment, and send the running speed Vt data, vibration frequency F data, and motor speed Vr data to the operation analysis processing unit; the analysis processing unit obtains the frequency modulation parameter CL according to the vibration frequency F; The device image acquisition unit is used to acquire image data of the end of the gas meter production line conveying device, and send the acquired image data of the end of the gas meter production line conveying device to the aging evaluation unit, which analyzes the image data of the end of the gas meter production line conveying device, calculates the aging rate Rl, and identifies whether there is a defect. If there is a defect, an aging signal Gc is generated, and the aging signal Gc is sent to the operation analysis processing unit; The operation analysis and processing unit generates a control signal and / or alarm information according to the fault occurrence rate Ry, the aging rate Rl, and the operation status level CM; the control signal is used to adjust the motor speed Vr, and the alarm information is displayed through the terminal, wherein the control signal includes a control parameter K, and the control parameter K is used to calculate the operation speed Vm.

2. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 1, it is characterized in that: The position information acquisition unit includes a first image acquisition camera, which is located directly above the gas meter production line conveying equipment. The operation process of the position information acquisition unit is as follows: An angle threshold L is set, the first image acquisition camera acquires image data of the gas meter assembly on the transmission device to form gas meter position image data, and sends the acquired gas meter position image data to the operation warning unit, the operation warning unit divides the gas meter position image data into n independent position coordinate data, n is a positive integer; the position coordinate data includes three-dimensional coordinate data of adjacent gas meter assemblies; the operation warning unit calculates the position angle E of two adjacent gas meters in the position coordinate data, and then compares the obtained gas meter position angle E with the set angle threshold L respectively to obtain the operation risk coefficient , i=1~n.

3. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 2, it is characterized in that: The operation analysis processing unit compares whether the operation risk coefficient ZS is within a preset risk coefficient threshold * Within the ZB range, if the operating risk factor ZS exceeds the risk factor threshold *ZB, the operation analysis and processing unit generates alarm information and notifies the operation and maintenance personnel to carry out timely maintenance; If the operating risk factor ZS does not exceed the risk factor threshold *ZB, the operation analysis processing unit obtains the operation status level CM according to the operation risk coefficient ZS; if the operation risk coefficient ZS is less than 1, then , where Y is a state parameter, Y is greater than or equal to 2; if the operation risk coefficient ZS is greater than 1, CM=1; the operation warning unit sends the operation state level CM to the operation analysis processing unit, and the operation analysis processing unit generates a control signal for adjusting the motor speed Vr to reach the calculated operation speed Vm.

4. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 2, it is characterized in that: The first image acquisition camera acquires gas meter status image data and sends the gas meter status image data to the fault assessment unit; the fault assessment unit annotates the gas meter status image data, adopts a pre-trained convolutional neural network model ResNet, uses the annotated gas meter status image data to train the convolutional neural network model ResNet, and adjusts hyperparameters; then uses a cross entropy loss function to measure the difference between the model evaluation and the true label; integrates the trained model into the fault assessment unit to realize real-time image analysis.

5. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 4, it is characterized in that: According to the collected gas meter status image data, the data are labeled as normal, offset, and tilted respectively; the labeled data are divided into a training set and a test set, the training set accounts for 80% of the labeled data, and the rest is a test set; the training set is used to adjust the hyperparameters of the convolutional neural network model ResNet; then the adjusted convolutional neural network model ResNet is used for evaluation to obtain the model-evaluated fault rate Ry, and the model-evaluated fault rate Ry is compared with the real label fault rate Rr, and the cross entropy loss function is used to measure the difference between the model evaluation and the real label, where the cross entropy loss function is , the smaller L is, the closer the convolutional neural network model ResNet evaluation is to the real distribution; When L is less than or equal to the set threshold, the trained model is integrated into the fault assessment unit; The detection time is set to Tb, and the fault occurrence rate Ry is calculated within the time range of Tb. If the fault occurrence rate Ry is greater than the threshold value Ra, a fault signal Gb is generated and sent to the operation analysis processing unit, and the operation analysis processing unit generates an alarm message; if the fault occurrence rate Ry is less than the threshold value Ra, the fault occurrence rate Ry is sent to the operation analysis processing unit.

6. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 5, it is characterized in that: The electronic signal acquisition unit includes a sensor for collecting running speed Vt data, vibration frequency F data, and motor speed Vr data; when the collected vibration frequency F data is sent to the operation analysis processing unit, if the vibration frequency F is greater than the set vibration frequency threshold, the operation analysis processing unit generates a feedback signal and sends the feedback signal to the fault assessment unit; after receiving the feedback signal, the fault assessment unit sets the detection time again and calculates the fault occurrence rate Ry within the corresponding time. If the recalculated fault occurrence rate Ry is less than the threshold Ra, the vibration frequency threshold is reset. If the recalculated fault occurrence rate Ry is greater than the threshold Ra, a fault signal Gb is generated and sent to the operation analysis processing unit. The operation analysis processing unit generates an alarm message to notify the maintenance personnel to maintain the equipment. After maintenance, the above steps are repeated until the vibration frequency F is less than the set vibration frequency threshold or the fault occurrence rate Ry is less than the threshold Ra.

7. According to the image recognition-based gas meter production line conveying equipment operation detection system of claim 1, it is characterized in that: The device image acquisition unit includes a second image acquisition camera, which acquires data of the end image of the gas meter production line conveying equipment and sends it to the aging assessment unit; the aging assessment unit divides the acquired image data of the end image of the gas meter production line conveying equipment into M sub-regions, and then extracts features of the M sub-region data respectively, establishes a gray level co-occurrence matrix GL, calculates the contrast Dt and entropy value St of the M sub-regions respectively according to the gray level co-occurrence matrix GL, and then calculates the aging number m; The contrast Dt is calculated and multiplied by the entropy value St to obtain the value SD, and the product of the contrast Dt of the standard image and the entropy value St is calculated to obtain the value LD. If SD is greater than LD, it is considered that the sub-region has a defect and is included in the aging number m; if SD is less than LD, it is considered that the sub-region is normal and is not included in the aging number m; the aging rate Rl=m / M*100% is calculated; the aging assessment unit generates an aging signal Gc, and sends the aging signal Gc and the aging rate Rl to the operation analysis processing unit. If Rl is greater than the threshold Rb, the operation analysis processing unit generates an alarm message.

8. A gas meter production line conveying equipment operation detection system based on image recognition according to claim 7, characterized in that: Establish the gray-level co-occurrence matrix GL and calculate the contrast Dt and entropy value St. The specific steps are as follows: The M sub-region data are grayed out separately to create a matrix with a size of N*N, where N is the number of gray levels and the contrast ; Entropy ; Each element (i, j) represents the number of occurrences of a pixel pair with grayscale values ​​i and j in a specified direction and distance; A decimal number greater than 0.

9. The gas meter production line conveying equipment operation detection system based on image recognition according to claim 1 is characterized in that: The frequency modulation parameters ,in is the standard vibration frequency; control parameters ,in is the current running speed, is the ideal operating speed, Ry is the failure rate, and Rl is the aging rate; , , are weight coefficients and are all greater than 0. .

10. A gas meter production line conveying equipment operation detection system based on image recognition according to claim 9, characterized in that: The operation analysis processing unit adjusts the motor speed Vr so that the gas meter production line conveying equipment reaches the calculated operation speed Vm. .

Images