A conveying equipment for automobile parts production

Abstract

The present invention relates to the technical field of conveyor belt speed control, and specifically to a conveying equipment for automobile parts production. The present invention evaluates the sliding risk of automobile parts on a conveyor belt by combining image frame analysis with pressure data monitoring. First, image frames of automobile parts on the conveyor belt at different times and pressure data at each measuring point on the conveyor belt are obtained, and the position mutation index is calculated based on the change in the position relationship of automobile parts in the image frame, and the pressure mutation index is calculated according to the change in pressure data at each measuring point at different times, and the sliding risk index is determined by combining the two. When the preset conditions are not met, the conveyor belt speed is adjusted until the corresponding critical speed is found. The method determines the critical speed for each automobile part, and controls the conveyor belt speed through PID during the conveying process of automobile parts, effectively avoiding the sliding and collision of automobile parts, and ensuring the conveying efficiency and automobile parts quality.

Description

Technical Field

[0001] The invention relates to the technical field of conveyor belt speed control, and in particular to a conveying device for automobile parts production. Background Art

[0002] In the production and processing of auto parts, different types of auto parts have unique shapes, sizes, materials and processing requirements, so the production processes and processing steps they need to go through are also different. In order to ensure the smooth and efficient production process, different types of auto parts need to be transported to different production workshops. Conveyor belts, as an efficient and reliable material transportation equipment, have been widely used in the field of auto parts production and transportation.

[0003] Since conveyor belts usually run at a fixed speed when transporting auto parts, if the fixed speed of the conveyor belt is too slow, the transportation efficiency of auto parts will be low; and since auto parts are of different types, shapes, sizes and weights, if the fixed speed of the conveyor belt is too fast, some types of auto parts may slip on the conveyor belt, which may cause collisions between auto parts and cause damage to the auto parts. Summary of the invention

[0004] In order to solve the technical problem that when the conveyor belt uses a fixed speed, too low a speed easily causes low efficiency in conveying automobile parts, while too high a speed causes the automobile parts on the conveyor belt to slip and collide with each other, thereby affecting the quality of automobile parts, the purpose of the present invention is to provide a conveying device for automobile parts production, and the technical solution adopted is as follows:

[0005] The present invention proposes a conveying device for automobile parts production, wherein an image acquisition device is installed on a support of a conveyor belt, and a regional pressure sensor and a motor are installed on the conveyor belt; the image acquisition device is used to obtain image frames of automobile parts on the conveyor belt, and the regional pressure sensor is used to obtain pressure data at each measuring point on the belt of the conveyor belt, and the motor is regulated according to the image frame and the pressure data to realize a speed control method of the conveyor belt, wherein the speed control method of the conveyor belt includes:

[0006] Select any automobile component as the component to be tested, and when the component to be tested is being conveyed, obtain image frames of the component to be tested at different times and pressure data of each measuring point on the conveyor belt at different times at the conveyor belt speed to be tested;

[0007] In different image frames, based on the position change relationship between the components to be tested, the position mutation index of the components to be tested is determined; according to the change of pressure data at all measuring points on the conveyor belt at different times, the pressure mutation index corresponding to the components to be tested is obtained;

[0008] According to the position mutation index of the tested component and the corresponding pressure mutation index, the sliding risk index of the tested component at the tested speed of the conveyor belt is determined; when the sliding risk index does not meet the preset condition, the tested speed of the conveyor belt is increased to the next tested speed, and a new sliding risk index is determined until the preset condition is met, and the final tested speed is used as the critical speed of the conveyor belt when the tested component slips;

[0009] During the transportation of automobile parts, the types of automobile parts on the conveyor belt are identified, and the current speed of the conveyor belt is PID controlled based on the critical speeds corresponding to all types of automobile parts on the conveyor belt.

[0010] Furthermore, the method for obtaining the position mutation index includes:

[0011] In each image frame, based on the Euclidean distance between the centers of gravity of all the components to be tested, determining the position characteristic value of the components to be tested in each image frame;

[0012] The absolute value of the difference between the position characteristic values ​​of the component to be tested in the image frames at two adjacent moments is used as the position mutation factor between the image frames at two adjacent moments of the component to be tested;

[0013] The position mutation index of the tested component is obtained by integrating the position mutation factors of the image frames at all two adjacent moments.

[0014] Furthermore, the method for obtaining the position feature value includes:

[0015] In each image frame, the pre-trained neural network is used to segment the image frame to obtain all the components to be tested;

[0016] Determine the preset range based on the center of gravity of each component to be tested;

[0017] In the preset range corresponding to each component to be tested, the average of the Euclidean distances between the center of the component to be tested and the centers of gravity of all the remaining components to be tested is used as the position factor of the center component to be tested;

[0018] The position factors of all the components to be measured in each image frame are fused to obtain the position characteristic value.

[0019] Furthermore, the method for obtaining the pressure mutation index includes:

[0020] At each measuring point on the conveyor belt, the absolute value of the difference between the pressure data values ​​at two adjacent moments is taken as the pressure mutation factor, and the average of the pressure mutation factors at all adjacent moments at each measuring point is taken as the pressure mutation characteristic value;

[0021] The pressure mutation characteristic values ​​at all measuring points are integrated to determine the pressure mutation index corresponding to the component to be tested.

[0022] Furthermore, the method for obtaining the sliding risk index includes:

[0023] The value after normalizing the sum of the position mutation index of the tested component and the corresponding pressure mutation index is used as the sliding risk index of the tested component at the tested speed of the conveyor belt.

[0024] Furthermore, the preset condition is set as the sliding risk index being greater than or equal to a preset sliding threshold.

[0025] Furthermore, the identifying the types of automobile parts on the conveyor belt includes:

[0026] The image frames at the current moment in the process of conveying automobile parts are obtained, and the image frames at the current moment are used as the input of the pre-trained neural network, and the output result is the type of automobile parts in the image frames at the current moment.

[0027] Furthermore, the PID control of the current speed of the conveyor belt based on the critical speeds corresponding to all types of automobile parts on the conveyor belt includes:

[0028] Obtaining a preset proportional coefficient, a preset differential coefficient, and a preset integral coefficient;

[0029] Among the critical speeds corresponding to all kinds of automobile parts on the conveyor belt, the smallest critical speed is used as the input of the PID controller;

[0030] The current speed of the conveyor belt is PID controlled based on a preset proportional coefficient, a preset differential coefficient, and a preset integral coefficient.

[0031] Furthermore, the neural network is CNN.

[0032] Further, the obtaining of the preset proportional coefficient, the preset differential coefficient and the preset integral coefficient includes:

[0033] The preset proportional coefficient, the preset differential coefficient and the preset integral coefficient are obtained based on the Ziegler-Nichols method.

[0034] The present invention has the following beneficial effects:

[0035] First, for each type of automobile parts, at the speed to be tested of the conveyor belt, the image frames of the automobile parts at different times and the pressure data at each measuring point on the conveyor belt are obtained. The image frames at different times can be used to capture information such as the position change of the automobile parts on the conveyor belt, and the pressure data can be used to capture the stability of the automobile parts during the transportation process. Then, in different image frames, the position change relationship between the automobile parts is analyzed to obtain the position mutation index of the automobile parts, which is used to evaluate the risk of sliding of the automobile parts at the speed to be tested of the conveyor belt in the subsequent evaluation; then, the change of the pressure data at all measuring points on the conveyor belt is analyzed to calculate the corresponding pressure mutation index of the automobile parts. The pressure mutation index can quantify the stability of the automobile parts during the transportation process at the speed to be tested of the conveyor belt, and then help to discover potential sliding risks. Further, the position mutation index of the automobile parts and the corresponding pressure mutation index are combined to determine the sliding risk index of the automobile parts. The sliding risk index quantifies the risk of sliding of the automobile parts at the speed to be tested of the conveyor belt. When the sliding risk index of the automobile parts does not meet the preset conditions, it can be considered that at the measured speed of the conveyor belt at this time, the possibility of sliding of the automobile parts on the conveyor belt is small, so the measured speed of the conveyor belt is increased and adjusted and the new sliding risk index of the automobile parts is determined, until the new sliding risk index meets the preset conditions and stops, and it is considered that the measured speed of the conveyor belt at this time is the critical speed of sliding of the automobile parts. So far, the critical speed of the conveyor belt when each automobile part may slide can be obtained through the above method. Finally, in the process of conveying automobile parts, the types of all automobile parts on the conveyor belt are identified, and the current speed of the conveyor belt is PID controlled based on the critical speed corresponding to all types of automobile parts on the conveyor belt, which can effectively avoid the sliding of automobile parts on the conveyor belt, and avoid collisions between automobile parts while ensuring the efficiency of automobile parts transportation, thereby ensuring the quality of automobile parts. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A schematic diagram of the structure of a conveying device for automobile parts production provided by one embodiment of the present invention;

[0038] Figure 2A method flow chart of a transmission belt speed control method provided by one embodiment of the present invention;

[0039] Figure 3 A flow chart of a method for obtaining a position mutation index provided by an embodiment of the present invention;

[0040] Figure numerals: 1-feeding port, 2-belt, 3-bracket, 4-image acquisition device, 5-auto parts, 6-discharging port, 7-regional pressure sensor, 8-data processing and calculation module, 9-motor. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a conveying device for automobile parts production proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0043] The specific scheme of a conveying device for automobile parts production provided by the present invention is described in detail below with reference to the accompanying drawings.

[0044] See also Figure 1 , which shows a schematic structural diagram of an automobile parts production equipment provided by an embodiment of the present invention. By regulating the motor, a method for controlling the speed of the conveyor belt can be implemented.

[0045] The process is briefly described as follows: when conveying auto parts, since multiple auto parts will be conveyed on the conveyor belt at the same time, and different types of auto parts have different shapes, sizes, and weights, the critical state of sliding on the conveyor belt is also different. Therefore, in order to control the speed of the conveyor belt in real time, so that the speed of the conveyor belt can ensure the efficiency of auto parts conveying while avoiding auto parts sliding and collision, the critical speed of the conveyor belt when each auto part is about to slide on the belt should be determined first. For any auto part, it is put onto the conveyor belt 2 through the conveyor belt feed port 1. In the process of being transported to the discharge port 6 through the belt 2, the auto parts on the belt 2 are photographed by the image acquisition device 4 installed on the bracket 3 (the shooting range of the image acquisition device can cover the entire belt area), and image frames at different times are obtained. At the same time, the pressure data at each measuring point on the belt 2 is obtained by using the regional pressure sensor 7. The image frame data and pressure data are processed by the data processing and calculation module 8, and the critical speed of the conveyor belt when each auto part is about to slide on the belt 2 can be determined. At this point, the critical speed corresponding to each type of automobile parts can be obtained. Finally, in the process of simultaneously conveying multiple automobile parts, the automobile parts on the belt 2 are photographed by the image acquisition device 4 installed on the bracket 3 to obtain the image frame at the current moment, and the image frame data at the current moment is analyzed based on the data processing and calculation module 8 to identify the type of automobile parts on the conveyor belt, and the conveyor belt motor 9 is regulated based on the critical speed corresponding to all types of automobile parts on the conveyor belt to achieve PID control of the conveyor belt speed.

[0046] See also Figure 2 , which shows a method flow chart of a method for controlling the speed of a conveyor belt provided in one embodiment of the present invention, the method comprising the following steps:

[0047] Step S1: select any automobile component as the component to be tested, and when the component to be tested is conveyed, obtain image frames of the component to be tested at different times and pressure data of each measuring point on the conveyor belt at different times at the measured speed of the conveyor belt.

[0048] When using a conveyor belt to transport auto parts, since the types of auto parts on the conveyor belt are different and their shapes and sizes are different, the critical state of sliding of different types of auto parts on the conveyor belt is also inconsistent. Therefore, if the conveyor belt uses a fixed speed to transport auto parts, some auto parts may slip. Therefore, in order to effectively prevent auto parts from sliding on the belt and then colliding with each other, the critical speed of the conveyor belt when each auto part may slip on the belt is first analyzed.

[0049] In order to obtain the critical speed corresponding to each automobile component, a traversal method can be used. For any automobile component, it is taken as the component to be tested, and its sliding condition is analyzed at a certain speed of the conveyor belt to be tested. Then, it is determined whether the speed of the conveyor belt to be tested at this time may cause the component to be tested to slide. If not, the current speed to be tested is increased to the next speed to be tested, and the next speed to be tested is analyzed until the critical speed of the conveyor belt corresponding to the component to be tested that may slide is found.

[0050] Specifically, the acquisition of critical speed requires analysis of the status of automobile parts on the conveyor belt, mainly position change information and stability. The position change information can be determined by image frame data, and the stability can be characterized by the changes in pressure data at each measuring point on the conveyor belt, so it is necessary to obtain image frame data and pressure data.

[0051] For image frame data, high-precision image acquisition equipment, such as high-definition cameras, can be used to continuously capture image frames of the components to be tested at different times during the operation of the conveyor belt. These image frames not only record the real-time position of the components to be tested on the conveyor belt, but also imply the relative motion relationship between the components to be tested, providing key information for the subsequent analysis of the sliding of the components to be tested. At the same time, based on the regional pressure sensors deployed on the conveyor belt, the pressure data at all measuring points on the conveyor belt at different times are obtained. The pressure data reflects the pressure distribution of the components to be tested on the contact surface with the belt during the transportation process. Its changing fluctuation characteristics are directly related to the stability of the components to be tested on the conveyor belt, and are important indicators for evaluating the sliding risk of the components to be tested.

[0052] It should be noted that the image frame and pressure data should be acquired at the same time to ensure time consistency and thus ensure the accuracy of subsequent process calculations; in the embodiment of the present invention, the setting method for different moments is: obtain the total time of the automobile parts from the feed port to the discharge port at the speed to be tested, and divide the total time by setting breakpoints, and use each breakpoint as each moment in the embodiment of the present invention. For example, at a certain speed to be tested, the total time is 12 seconds. If the number of breakpoints is set to 3, then at 4s, 8s and 12s, the image frame and pressure data are collected once. The specific interval can be adjusted according to the implementation scenario and is not limited here; when calculating the critical speed of the conveyor belt corresponding to each automobile part, the number of automobile parts placed on the conveyor belt is at least 2, and the specific value is not limited here.

[0053] At this point, we can obtain the image frames of the tested parts at different times and the pressure data of each measuring point on the conveyor belt at different times under the measured speed of the conveyor belt. By combining the image frame data with the pressure data, we can monitor the movement state of the tested parts on the conveyor belt from all directions and angles in the same time frame, providing data support for the subsequent calculation process.

[0054] Step S2: In different image frames, based on the position change relationship between the components to be tested, determine the position mutation index of the components to be tested; according to the change of pressure data at all measuring points on the conveyor belt at different times, obtain the pressure mutation index corresponding to the components to be tested.

[0055] By analyzing the positional relationship between the components to be tested in different image frames, the motion state and relative positional relationship of the components to be tested on the conveyor belt can be captured, which helps to quantify the position mutation index of the components to be tested. For the components to be tested on the conveyor belt, since their postures are different when sliding, analyzing their motion state only through image data may result in the relative positional relationship remaining unchanged or only undergoing minor changes at different times even if sliding occurs; and since pressure data is an important parameter reflecting the interaction force between the component and the belt contact surface, the changes in pressure data at all measuring points on the conveyor belt can be analyzed to quantify the pressure mutation index of the components to be tested, which is used to characterize the stability of the state of the components to be tested at the conveyor belt's test speed. By combining image data with pressure data, the evaluation accuracy of the motion state analysis of the components to be tested on the conveyor belt can be improved.

[0056] Because the conveyor belt itself has a speed, even if the parts to be tested on the conveyor belt do not slide, when the image acquisition device is fixed, the size of the parts to be tested in the image frames at different times will also change (larger when near and smaller when far). However, the relative position relationship between the parts to be tested is fixed. Therefore, if the relative position relationship between the parts to be tested changes significantly in the image frames at different times, it can be regarded as the parts to be tested sliding. Therefore, in different image frames, based on the position relationship between the parts to be tested, the position mutation index of the parts to be tested is determined, which is used to quantify the risk of sliding of the parts to be tested at the measured speed of the conveyor belt.

[0057] Preferably, in one embodiment of the present invention, the method for obtaining the position mutation index includes:

[0058] See also Figure 3 , which shows a method flow chart of a method for obtaining a position mutation index in one embodiment of the present invention, the method comprising the following steps:

[0059] Step S201: In each image frame, based on the Euclidean distance between the centers of gravity of all the components to be tested, the position characteristic value of the components to be tested in each image frame is determined.

[0060] Since the information contained in each image frame is all the information on the belt, first in each image frame, a pre-trained neural network is used to segment the image frame to determine the areas corresponding to all the components to be tested, and then the center of gravity of each component to be tested is obtained. The position of the center of gravity can be used as a reference position to determine the position information of the component to be tested. Then in each image frame, a two-dimensional coordinate system is constructed with the same position, such as the lower left corner of the image frame as the origin, the horizontal right direction as the horizontal axis, and the vertical upward direction as the vertical axis. At this time, the center of gravity of each component to be tested in each image frame has a two-dimensional coordinate.

[0061] Because the positional relationship can be characterized by distance, in each image frame, a preset range is constructed with the center of gravity of each component to be tested as the center, and the components to be tested within the preset range except the central component to be tested are regarded as the neighborhood components of the central component to be tested. The average of the Euclidean distances between the center of gravity of each central component to be tested and its corresponding neighborhood components is used as the position factor of each central component to be tested, and then the position factors of all components to be tested on the conveyor belt are integrated to obtain the positional characteristic values ​​of all components to be tested on the conveyor belt. At this time, the positional characteristic values ​​can characterize the relative positional relationship between the components to be tested in each image frame. The formula model of the positional characteristic value can be, for example, as follows:

[0062]

[0063] in, Represents the position characteristic values ​​of all components to be tested in each image frame; Indicates the number of parts to be tested on the conveyor belt; Represents the number of neighboring components of each component to be tested on the conveyor belt; Indicates The center of gravity of the component to be tested and its corresponding The Euclidean distance between the centroids of neighboring components.

[0064] In the formula model of the position eigenvalue, the Euclidean distance between the center of gravity of each component to be tested and the center of gravity of each corresponding neighboring component is calculated. The larger the Euclidean distance value, the more dispersed the distribution of components is within the preset range of the components to be tested. The average of the Euclidean distances between each component to be tested and all corresponding neighboring components is taken as the position factor of each component to be tested. The value of the position factor measures the positional relationship between the component to be tested and its surrounding components to be tested. The larger the value, the more dispersed the distribution. Finally, the position factors of all components to be tested in the image frame are fused, and the mean is obtained as the position characteristic value of the component to be tested in each image frame. The position characteristic value measures the position distribution of all components to be tested in the entire image frame. The larger the value, the more dispersed the distribution, and vice versa.

[0065] It should be noted that the training process of the neural network is a well-known technology and will not be elaborated here; the preset range is set to a circular area with the center of gravity of each component to be tested as the center and a preset length as the radius, wherein the preset length is set to 5, and the specific length of the radius can be adjusted according to the implementation scenario, and is not limited here, but it is necessary to ensure that there is at least one component other than the central component to be tested within the preset range; at the same time, the preset range can also be set to other shapes in other embodiments of the present invention, which will not be elaborated or limited here.

[0066] Step S202: Analyze the differences between the position characteristic values ​​of the component to be tested in different image frames to determine the position mutation index of the component to be tested.

[0067] When the parts to be tested slide, the relative position relationship between the parts to be tested will change, that is, the original position distribution will be destroyed. Therefore, the deviation of the relative position relationship between the parts to be tested in the image frames at different times is finally analyzed, that is, in the image frames at two adjacent times, the absolute value of the difference between the position characteristic values ​​of the parts to be tested is calculated as the position mutation factor between the image frames at two adjacent times of the parts to be tested. At this time, the parts to be tested will have a position mutation factor between the image frames at every two adjacent times. Finally, the position mutation factors of the parts to be tested at all two adjacent times are combined to obtain the position mutation index of the parts to be tested. The formula model of the position mutation index can be specifically, for example:

[0068]

[0069] in, represents the position mutation index; represents the number of adjacent moments; Indicates Position characteristic values ​​of all components to be tested in the image frame at time 1 in a group of adjacent time points; Indicates Position characteristic values ​​of all components to be tested in the image frame at time 2 in a group of adjacent time points; Represents the normalization function.

[0070] In the formula model of the position mutation index, since the position characteristic value can characterize the position distribution characteristics of all the parts to be tested in the image frame at a certain moment, if the position distribution characteristic values ​​of all the parts to be tested in the image frames at different moments have changed significantly, it can be regarded as the parts to be tested on the belt slipping at the measured speed of the conveyor belt. Therefore, the absolute value of the difference between the position characteristic values ​​of all the parts to be tested in the image frames at two adjacent moments is calculated as the position mutation factor. When the position mutation factor is larger, it means that in the image frames at two adjacent moments, the position distribution between the parts to be tested has undergone a large change, which may be caused by the sliding of the parts to be tested. On the contrary, if the position mutation factor is smaller, it means that in the image frames at two adjacent moments, the position distribution between the parts to be tested has not undergone a large change, which can be regarded as the possibility of sliding of the parts to be tested. Therefore, there is a positive correlation between the position mutation factor and the position change of the parts to be tested. At the same time, in order to facilitate the calculation in the subsequent process, the mean value of the position mutation factor of the parts to be tested in all image frames at two adjacent moments can be normalized so that its value is between 0 and 1, and the value obtained after processing is used as the position mutation index of the parts to be tested.

[0071] At this point, by analyzing the image data, the position mutation index of the component to be tested is obtained, which can be used as an important indicator for subsequently measuring the state of the component to be tested on the conveyor belt with the speed to be tested.

[0072] In the process of conveying auto parts by conveyor belt, the motion state of auto parts will directly affect the pressure distribution of the area below or around it. When auto parts do not slide on the belt, they are affected by static friction, which keeps the auto parts and the belt relatively still. Then the pressure data at each measuring point on the conveyor belt will fluctuate slightly within a small range. When auto parts slide, there is relative motion between the auto parts and the belt. Then the pressure data at each measuring point on the conveyor belt will mutate. For example, the pressure data measured in the area where there are no auto parts is 0, and when the auto parts slide to this area, the pressure data will suddenly increase. Therefore, by analyzing the changes in the pressure data at all measuring points on the conveyor belt at different times, the pressure mutation index corresponding to the tested parts can be quantified, which will help to evaluate whether the tested parts have slipped at the tested speed of the conveyor belt in the subsequent process.

[0073] Preferably, in one embodiment of the present invention, the method for obtaining the pressure mutation index includes:

[0074] When the component to be tested slips on the belt, the pressure data at the measuring point it passes through when sliding will cause a mutation. Therefore, at each measuring point on the conveyor belt, based on the difference in the pressure data values ​​at two adjacent moments, where the difference refers to the absolute value of the difference, the pressure mutation factor is calculated. The pressure mutation factor at each measuring point characterizes the pressure change at each measuring point on the conveyor belt at two adjacent moments. Then, for each measuring point, the mean of the pressure mutation factors at all adjacent moments at the measuring point is used as the pressure mutation characteristic value. Finally, the pressure mutation characteristic values ​​at all measuring points are integrated to determine the pressure mutation index corresponding to the component to be tested. The formula model of the pressure mutation index can be, for example, as follows:

[0075]

[0076] in, Indicates the pressure mutation index corresponding to the component to be tested; Indicates the number of measuring points on the conveyor belt; represents the number of adjacent moments; Indicates the first The measuring point is at The pressure data at time 1 in the group of adjacent moments; Indicates the first The measuring point is at The pressure data at time 2 in a group of adjacent moments.

[0077] In the formula model of the pressure mutation index, when the tested component on the belt does not slip, the pressure data at the same measuring point at different times should be roughly consistent (due to external factors such as sensor accuracy or conveyor belt vibration, fluctuations are allowed within a small range). When a large mutation occurs, it can be regarded as the tested component slipping on the belt, resulting in changes in the pressure data of the same measuring point at different times. Therefore, the absolute value of the difference between the pressure data of the same measuring point on the conveyor belt at two adjacent times is calculated to obtain the pressure mutation factor. The larger the pressure mutation factor, the greater the pressure data of the measuring point at two adjacent moments has changed greatly, and the reason for the large change is that the position of the measured component on the belt has changed, which means that the measured component has slipped on the belt. Then, for each measuring point, the mean of the pressure mutation factors of the measuring point at all two adjacent moments is taken as the pressure mutation characteristic value. Since the pressure mutation value integrates the pressure mutation factors of the measuring point at all two adjacent moments, the pressure mutation characteristic value can more accurately characterize the pressure change at the measuring point. Finally, the pressure mutation characteristic values ​​at all measuring points on the conveyor belt are integrated, and the average is used as the pressure mutation index of the parts to be tested on the belt. The larger the pressure mutation index, the more it indicates that the parts to be tested on the belt have slipped.

[0078] At this point, by analyzing the changes in pressure data of all measuring points at different times, the pressure mutation index corresponding to the tested component can be determined, which is also an important indicator for measuring whether the tested component slips on the belt.

[0079] Step S3: Determine the sliding risk index of the component to be tested at the measured speed of the conveyor belt according to the position mutation index and the corresponding pressure mutation index of the component to be tested; when the sliding risk index does not meet the preset conditions, increase the measured speed of the conveyor belt to the next measured speed, determine a new sliding risk index until the preset conditions are met, and stop, and use the final measured speed as the critical speed of the conveyor belt when the component to be tested slips.

[0080] Based on the process in step S2, the position mutation index and the corresponding pressure mutation index of the tested parts on the conveyor belt can be calculated at the tested speed of the conveyor belt. Both of these indicators can evaluate the sliding risk of the tested parts at the tested speed of the conveyor belt, so the sliding risk index of the tested parts at the tested speed of the conveyor belt can be obtained by combining the two. After obtaining the sliding risk index, it is judged. When it does not meet the preset conditions, it is considered that the speed of the conveyor belt at this time will not cause the tested parts to slide. In order to improve the transportation efficiency of automobile parts, the speed of the conveyor belt can be increased; therefore, the tested speed of the conveyor belt is increased and adjusted, and at the next tested speed, the calculation in the above steps is continued to obtain a new sliding risk index of the tested parts at the new tested speed of the conveyor belt. The above process is repeated until the new sliding risk index meets the preset conditions. It can be considered that at the tested speed of the conveyor belt at this time, if the speed of the conveyor belt continues to increase, the tested parts will slide, so the tested speed of the conveyor belt at this time is used as the critical speed of the conveyor belt when the tested parts slide.

[0081] Preferably, in one embodiment of the present invention, the method for obtaining the sliding risk index includes:

[0082] The position mutation index and the corresponding pressure mutation index of the tested parts can both reflect the movement state of the tested parts on the conveyor belt. The larger the position mutation index, the greater the position distribution of the tested parts in the image frames at different times, which is considered to be caused by the sliding of the tested parts. When the corresponding pressure mutation index of the tested parts is larger, it means that the pressure data of the measuring points on the conveyor belt at different times has changed greatly, and the reason for the large change is that the position of the tested parts on the belt has changed, that is, sliding has occurred. Therefore, the position mutation index of the tested parts can be integrated with the corresponding pressure mutation index to more comprehensively evaluate the sliding risk of the tested parts at the tested speed of the conveyor belt.

[0083] In this embodiment of the present invention, the sum of the position mutation index of the component to be tested and the corresponding pressure mutation index is normalized and used as the sliding risk index of the component to be tested at the test speed of the conveyor belt. At this time, the value range of the sliding risk index is 0-1, and the larger the value, the greater the risk of sliding of the component to be tested at the test speed of the conveyor belt.

[0084] It should be noted that normalization can be done by Normalization function.

[0085] Based on the above process, the sliding risk index of the component to be tested on the conveyor belt at the measured speed of the conveyor belt can be obtained. The sliding risk index represents the possibility of sliding of the component to be tested, and then a preset condition is set. Preferably, in one embodiment of the present invention, the preset condition is set to a sliding risk index greater than or equal to a preset sliding threshold. When the sliding risk index of the tested component does not meet the preset conditions, that is, the sliding risk index is less than the preset sliding threshold, it means that at the current tested speed of the conveyor belt, the friction between the tested component and the belt is sufficient to maintain its stable transportation, and the sliding risk is low. It is believed that the tested component will not slip at this time. Therefore, in order to improve the transportation efficiency of automobile parts, the speed of the conveyor belt can be appropriately increased while ensuring the safety of automobile parts. This increase in speed should ensure that the tested component will not slip. Therefore, repeat the above steps to calculate the new sliding risk index of the tested component at the new tested speed of the conveyor belt until the new sliding risk index meets the preset conditions, that is, the sliding risk index is greater than or equal to the preset sliding threshold, which means that at the current tested speed of the conveyor belt, the risk of sliding of the tested component is relatively high. Therefore, in order to avoid collision between parts and damage to parts, the current tested speed of the conveyor belt is used as the critical speed of the conveyor belt when the tested component slips.

[0086] Here, an example is given to illustrate the process of determining the critical speed: for any part to be tested, at the test speed 1 of the conveyor belt, the sliding risk index of the part to be tested is calculated. When the sliding risk index is less than the preset sliding threshold, the test speed 1 of the conveyor belt is adjusted to the test speed 2 (the test speed 2 is greater than the test speed 1), and then at the test speed 2 of the conveyor belt, the sliding risk index of the part to be tested is calculated. If the sliding risk index at this time is still less than the preset sliding threshold, the test speed 2 of the conveyor belt is further adjusted to the test speed 3 (the test speed 3 is greater than the test speed 2), and the sliding risk index of the part to be tested is calculated at the test speed 3. If the sliding risk index at this time is greater than or equal to the preset sliding threshold, If a sliding threshold is set, the process stops and the measured speed 3 is used as the critical speed of the conveyor belt when the measured part slips; on the contrary, if the sliding risk index at this time is still less than the preset sliding threshold, the measured speed 3 of the conveyor belt is adjusted to the measured speed 4 (the measured speed 4 is greater than the measured speed 3), and the new sliding risk index is continuously obtained and the measured speed of the conveyor belt is updated until the new sliding risk index is greater than or equal to the preset sliding threshold, and the process stops and the current measured speed is used as the critical speed of the conveyor belt when the measured part slips. In an embodiment of the present invention, when the measured speed is increased and updated, one unit speed can be increased each time, and the specific unit speed step size can be adjusted according to the implementation scenario, which is not limited here.

[0087] It should be noted that, since the value range of the sliding risk index is 0-1, the preset sliding threshold is set to 0.8. The specific value can be adjusted according to the implementation scenario and is not limited here.

[0088] Based on the above process, the critical speed of the conveyor belt when each component to be tested slips can be obtained.

[0089] Step S4: During the conveying process of automobile parts, the types of automobile parts on the conveyor belt are identified, and the current speed of the conveyor belt is PID controlled based on the critical speeds corresponding to all types of automobile parts on the conveyor belt.

[0090] Based on the above steps, the critical speed of the conveyor belt when each component to be tested, that is, each automobile component is about to slip on the conveyor belt can be determined. Since different types of automobile components have different corresponding critical speeds, and in the actual transportation process of automobile components, a variety of automobile components will be placed on the conveyor belt for transportation to different production workshops, in order to ensure that all automobile components will not slip during the transportation process to avoid collision damage between automobile components, the types of automobile components on the conveyor belt are first identified, and then the current speed of the conveyor belt is PID controlled based on the critical speeds corresponding to all types of automobile components, so that its speed can prevent all automobile components on the current belt from sliding, while ensuring the transportation efficiency and avoiding damage to automobile components.

[0091] Preferably, in one embodiment of the present invention, identifying the types of automobile parts on the conveyor belt includes:

[0092] First, the image frame at the current moment in the process of automobile parts transportation is obtained, and then the image frame at the current moment is used as the input of a pre-trained neural network. After being processed by the neural network, the type of automobile parts in the image frame at the current moment can be output, wherein the neural network can adopt a CNN neural network.

[0093] It should be noted that the training process of the neural network is a well-known technology and will not be elaborated here. It can be briefly described as follows: collect a large amount of image data of automobile parts, which should include various types of automobile parts and cover different viewing angles, lighting conditions and background environments as much as possible, and pre-process all collected image data, including adjusting the image size, etc. All image data are divided into training set and validation set in a ratio of 7:3, the image data of the training set is input into the neural network, and the gradient descent method is used for training until the loss function converges, and the performance of the neural network is evaluated using the validation set to complete the training process of the neural network; wherein, the loss function adopts the mean square error function.

[0094] After obtaining the types of auto parts on the conveyor belt at the current moment, the speed of the conveyor belt at the current moment can be PID controlled based on the critical speeds of the conveyor belts corresponding to the various auto parts, so that the speed of the conveyor belt can ensure the transportation efficiency of auto parts as much as possible while avoiding quality damage that may be caused by sliding between auto parts.

[0095] Preferably, in one embodiment of the present invention, PID control is performed on the current speed of the conveyor belt based on the critical speeds corresponding to all types of automobile parts, including:

[0096] In production automation and production line control, the PID controller is a commonly used feedback control method for adjusting the output of the system to approach the desired set point. In the aforementioned process, the critical speeds of the conveyor belt corresponding to various automotive parts and the types of automotive parts on the conveyor belt at the current moment are known, so among the critical speeds corresponding to all types of automotive parts in the image frame at the current moment, the minimum critical speed is used as the input of the PID controller, and the preset proportional coefficient, preset differential coefficient and preset integral coefficient are obtained at the same time. The preset proportional coefficient, preset differential coefficient and preset integral coefficient can be obtained by the Ziegler-Nichols method, and it is a well-known technology and will not be described here.

[0097] Finally, based on the preset proportional coefficient, the preset differential coefficient and the preset integral coefficient, the current measured speed of the conveyor belt is PID controlled.

[0098] During the transportation of auto parts, continuous monitoring of the types of auto parts on the conveyor belt facilitates real-time control of the conveyor belt speed.

[0099] In summary, first, for each type of automobile parts, at the speed to be measured of the conveyor belt, the image frames of the automobile parts at different times and the pressure data at each measuring point on the conveyor belt are obtained. The image frames at different times can be used to capture information such as the position change of automobile parts on the conveyor belt, and the pressure data can be used to capture the stability of automobile parts during transportation. Then, in different image frames, the position relationship between automobile parts is analyzed to obtain the position mutation index of automobile parts, which is used to evaluate the risk of sliding of automobile parts at the speed to be measured of the conveyor belt in the subsequent evaluation; then, the change of pressure data at all measuring points on the conveyor belt is analyzed to calculate the pressure mutation index corresponding to the automobile parts. The pressure mutation index can quantify the stability of automobile parts during transportation at the speed to be measured of the conveyor belt, and then help to discover potential sliding risks. Further, the position mutation index and pressure mutation index of automobile parts mentioned above are combined to determine the sliding risk index of automobile parts. The sliding risk index quantifies the risk of sliding of automobile parts at the speed to be measured of the conveyor belt. When the sliding risk index of the automobile parts does not meet the preset conditions, it can be considered that at the measured speed of the conveyor belt at this time, the possibility of sliding of the automobile parts on the conveyor belt is small, so the measured speed of the conveyor belt is adjusted and the new sliding risk index of the automobile parts is determined, until the new sliding risk index meets the preset conditions, and it is stopped. It is considered that the measured speed at this time is the critical speed at which the automobile parts do not slide. So far, the critical speed of the conveyor belt when each automobile part may slide can be obtained through the above method. Finally, in the process of conveying automobile parts, the types of all automobile parts on the conveyor belt are identified, and the current speed of the conveyor belt is PID controlled based on the critical speeds corresponding to all types of automobile parts. It can effectively prevent the automobile parts from sliding on the conveyor belt, and while ensuring the transportation efficiency of automobile parts, avoid collisions between automobile parts and ensure the quality of automobile parts.

[0100] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0101] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts, characterized in that: An image acquisition device is installed on a support of a conveyor belt, and a regional pressure sensor and a motor are installed on the conveyor belt; the image acquisition device is used to acquire image frames of automobile parts on the conveyor belt, and the regional pressure sensor is used to acquire pressure data at each measuring point on the belt of the conveyor belt, and the motor is regulated according to the image frame and the pressure data to realize the speed control of the conveyor belt, wherein the speed control method of the conveyor belt includes: Select any automobile component as the component to be tested, and when the component to be tested is being conveyed, obtain image frames of the component to be tested at different times and pressure data of each measuring point on the conveyor belt at different times at the conveyor belt speed to be tested; In different image frames, based on the position change relationship between the components to be tested, the position mutation index of the components to be tested is determined; according to the change of pressure data at all measuring points on the conveyor belt at different times, the pressure mutation index corresponding to the components to be tested is obtained; According to the position mutation index of the tested component and the corresponding pressure mutation index, the sliding risk index of the tested component at the tested speed of the conveyor belt is determined; when the sliding risk index does not meet the preset condition, the tested speed of the conveyor belt is increased to the next tested speed, and a new sliding risk index is determined until the preset condition is met, and the final tested speed is used as the critical speed of the conveyor belt when the tested component slips; During the transportation of auto parts, the types of auto parts on the conveyor belt are identified, and the current speed of the conveyor belt is PID controlled based on the critical speeds corresponding to all types of auto parts on the conveyor belt; The method for obtaining the position mutation index comprises: in each image frame, based on the Euclidean distance between the centers of gravity of all the components to be tested, determining the position characteristic value of the component to be tested in each image frame; taking the absolute value of the difference between the position characteristic values ​​of the component to be tested in the image frames at two adjacent moments as the position mutation factor of the component to be tested between the image frames at two adjacent moments; and obtaining the position mutation index of the component to be tested by comprehensively considering the position mutation factors of the component to be tested between all the image frames at two adjacent moments; The method for obtaining the position feature value comprises: in each image frame, using a pre-trained neural network to segment the image frame to obtain all the components to be tested; taking the center of gravity of each component to be tested as the center, determining a preset range; within the preset range corresponding to each component to be tested, taking the average of the Euclidean distances between the center of gravity of the central component to be tested and all the remaining components to be tested as the position factor of the central component to be tested; fusing the position factors of all the components to be tested in each image frame to obtain the position feature value; The method for obtaining the pressure mutation index includes: At each measuring point on the conveyor belt, the absolute value of the difference between the pressure data values ​​at two adjacent moments is taken as the pressure mutation factor, and the average of the pressure mutation factors at all adjacent moments at each measuring point is taken as the pressure mutation characteristic value; The pressure mutation characteristic values ​​at all measuring points are integrated to determine the pressure mutation index corresponding to the component to be tested; The method for obtaining the sliding risk index includes: The value after normalizing the sum of the position mutation index of the tested component and the corresponding pressure mutation index is used as the sliding risk index of the tested component at the tested speed of the conveyor belt.

2. A method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts according to claim 1, characterized in that: The preset condition is set as the sliding risk index being greater than or equal to a preset sliding threshold.

3. The method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts according to claim 1, characterized in that: The identification of the types of automobile parts on the conveyor belt includes: The image frames at the current moment in the process of conveying automobile parts are obtained, and the image frames at the current moment are used as the input of the pre-trained neural network, and the output result is the type of automobile parts in the image frames at the current moment.

4. The method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts according to claim 1, characterized in that: The PID control of the current speed of the conveyor belt based on the critical speeds corresponding to all types of automobile parts on the conveyor belt includes: Obtaining a preset proportional coefficient, a preset differential coefficient, and a preset integral coefficient; Among the critical speeds corresponding to all kinds of automobile parts on the conveyor belt, the smallest critical speed is used as the input of the PID controller; The current speed of the conveyor belt is PID controlled based on a preset proportional coefficient, a preset differential coefficient, and a preset integral coefficient.

5. The method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts according to claim 3, characterized in that: The neural network is CNN.

6. A method for controlling the speed of a conveyor belt on a conveying device for producing automobile parts according to claim 4, characterized in that: The obtaining of the preset proportional coefficient, the preset differential coefficient and the preset integral coefficient comprises: The preset proportional coefficient, the preset differential coefficient and the preset integral coefficient are obtained based on the Ziegler-Nichols method.

Images