Glass integrity detection system and use method

Abstract

The present invention belongs to the field of glass detection technology, and specifically relates to a glass integrity detection system and a method of use, wherein the glass integrity detection system comprises an intelligent management and control platform, a glass transmission module, a brush cleaning module, an image detection module, an automatic plate dropping module and a display warning module; the present invention cleans broken glass and dust on the surface of photovoltaic glass through the brush cleaning module, and after the cleaning is completed, the image detection module collects the surface image of the photovoltaic glass and determines the integrity status of the photovoltaic glass, thereby realizing automatic identification and detection, and when it is determined that the corresponding photovoltaic glass is damaged glass, the automatic plate dropping module is controlled to open to allow the damaged glass to fall, thereby realizing automatic plate dropping of glass with poor integrity, avoiding defective products from entering the next process and affecting production, reducing the workload of operators, improving the degree of automation and intelligence of the glass integrity detection process, significantly reducing the missed detection rate and improving the detection efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of glass detection, and in particular to a glass integrity detection system and a use method thereof. Background Art

[0002] Photovoltaic glass is an important packaging material for photovoltaic modules and has high quality requirements. If glass breakage and edge collapse are not discovered in time, it is easy to block the cleaning machine in the subsequent process, seriously affecting the normal production of the photovoltaic glass production line and not conducive to ensuring the quality of the produced photovoltaic glass;

[0003] At present, the integrity inspection of photovoltaic glass after tempering is manual inspection, that is, employees use their naked eyes to identify defects such as broken edges and corners of the glass, as well as scratches and cracks on the outer surface of the glass. This inspection method has a high missed detection rate and low efficiency. It cannot realize the automated inspection of the integrity of photovoltaic glass and has a low degree of intelligence.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a glass integrity detection system and a method of use, which solves the problems that the existing technology judges the integrity of photovoltaic glass through manual detection, has a high missed detection rate and low efficiency, cannot realize automated detection of photovoltaic glass integrity, and has a low level of intelligence.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A glass integrity detection system includes an intelligent management and control platform, a glass transmission module, a brush cleaning module, an image detection module, an automatic plate dropping module and a display warning module; the intelligent management and control platform controls the glass transmission module, the brush cleaning module, the image detection module and the automatic plate dropping module, the intelligent management and control platform controls the glass transmission module to transport photovoltaic glass that needs to be inspected for integrity, the intelligent management and control platform controls the brush cleaning module to clean broken glass and dust on the surface of the photovoltaic glass, and after the cleaning is completed, the photovoltaic glass is continuously transported through the glass transmission module to enter the imaging area of ​​the image detection module;

[0008] The intelligent management and control platform controls the image detection module to collect the surface image of the photovoltaic glass through a high-definition camera, identifies the surface defects of the photovoltaic glass based on the surface image, and judges the integrity of the photovoltaic glass. Based on this, the corresponding photovoltaic glass is marked as intact glass or damaged glass, and the marking information of the corresponding photovoltaic glass is sent to the intelligent management and control platform. The intelligent management and control platform sends the marking information of the corresponding photovoltaic glass to the display warning module for display;

[0009] After the inspection is completed, the photovoltaic glass will continue to be transported through the glass transmission module. If the corresponding photovoltaic glass is damaged glass, when the photovoltaic glass is transported to the top of the automatic drop module, the corresponding control command will be issued through the intelligent management and control platform to open the automatic drop module, and the automatic drop module will make the damaged glass fall down.

[0010] Furthermore, the intelligent management and control platform is communicatively connected with the detection accuracy impact assessment module. The detection accuracy impact assessment module is used to generate a high impact signal of detection accuracy or a low impact signal of detection accuracy through analysis before the image detection module collects images of the photovoltaic glass, and send the high impact signal of detection accuracy or the low impact signal of detection accuracy to the display and early warning module via the intelligent management and control platform. When the display and early warning module receives the high impact signal of detection accuracy, it issues a corresponding early warning.

[0011] Furthermore, the specific operation process of the detection accuracy impact assessment module includes:

[0012] During the image acquisition process of the image detection module, the deviation value of the light brightness of the environment in which the photovoltaic glass is located compared to the standard light brightness is marked as the light detection value, and the deviation value of the light color of the environment in which the photovoltaic glass is located compared to the preset standard light color is marked as the light color detection value, and the gas pollution value of the environment in which the photovoltaic glass is located is collected, and the phosgene detection value is obtained by numerically calculating the light detection value, the light color detection value and the gas pollution value, and the phosgene detection value is numerically compared with the preset phosgene detection threshold value. If the phosgene detection value exceeds the preset phosgene detection threshold value, a high-impact signal for detection accuracy is generated;

[0013] If the phosgene inspection value does not exceed the preset phosgene inspection threshold, the vibration frequency and vibration amplitude of the position of the photovoltaic glass are collected, and the actual placement posture image of the photovoltaic glass is overlapped with the standard placement posture image to obtain the posture non-overlap value; the placement inspection value is obtained by numerically calculating the vibration frequency, vibration amplitude and posture non-overlap value, and the placement inspection value is numerically compared with the preset placement inspection threshold. If the placement inspection value exceeds the preset placement inspection threshold, a high-impact signal for detection accuracy is generated; if the placement inspection value does not exceed the preset placement inspection threshold, a low-impact signal for detection accuracy is generated.

[0014] Furthermore, the intelligent management and control platform is communicatively connected with the management and control abnormality comprehensive evaluation module. The management and control abnormality comprehensive evaluation module is used to set a comprehensive evaluation period, analyze the management and control performance of the intelligent management and control platform for glass integrity detection during the comprehensive evaluation period, and generate a management and control abnormality signal or a management and control qualified signal through the analysis. The management and control abnormality signal or the management and control qualified signal is sent to the display and early warning module via the intelligent management and control platform. The display and early warning module issues a corresponding early warning when receiving the management and control abnormality signal.

[0015] Furthermore, the specific operation process of the control abnormality comprehensive evaluation module includes:

[0016] In the surface cleaning process of the corresponding photovoltaic glass, the cleaning speed fluctuation value and the cleaning pressure average value of the brush cleaning module in the cleaning process are collected, and the cleaning speed fluctuation value and the cleaning pressure average value are numerically compared with the preset cleaning speed fluctuation threshold value and the preset cleaning pressure average value range, respectively. If the cleaning speed fluctuation value exceeds the preset cleaning speed fluctuation threshold value or the cleaning pressure average value is not within the preset cleaning pressure average value range, the cleaning symbol QP-1 is assigned to the corresponding cleaning process;

[0017] In the image acquisition process of the corresponding photovoltaic glass, the number of times the detection accuracy high-impact signal is generated is collected and marked as the high-impact frequency, and the total time consumed for taking corresponding processing measures after the detection accuracy high-impact signal is generated is marked as the processing time value, and the high-impact frequency and the processing time value are numerically compared with the preset high-impact frequency threshold and the preset processing time threshold, respectively. If the high-impact frequency or the processing time value exceeds the corresponding preset threshold, the corresponding image acquisition process is assigned the abnormal symbol QW-1;

[0018] If there is a clearing symbol QP-1 or a sampling symbol QW-1 in the integrity detection process of the corresponding photovoltaic glass, the integrity detection process of the corresponding photovoltaic glass will be marked as a non-optimal process, and the number of non-optimal processes in the comprehensive evaluation period will be obtained and its ratio to the total number of photovoltaic glasses detected will be marked as the non-optimal number ratio;

[0019] By analyzing the conveying pipe analysis value of the glass transmission module and the falling plate pipe analysis value of the automatic falling plate module during the comprehensive evaluation period, the non-excellent number value, the conveying pipe analysis value and the falling plate pipe analysis value are numerically calculated to obtain the control abnormal evaluation value, and the control abnormal evaluation value is numerically compared with the preset control abnormal evaluation threshold. If the control abnormal evaluation value exceeds the preset control abnormal evaluation threshold, a control abnormal signal is generated; if the control abnormal evaluation value does not exceed the preset control abnormal evaluation threshold, a control qualified signal is generated.

[0020] Furthermore, the analysis and acquisition method of the transmission pipe analysis value is as follows:

[0021] During the operation of the glass transmission module in the comprehensive evaluation period, several detection time points are set, and the conveying speed of the glass transmission module at the corresponding time points is collected. The deviation value of the conveying speed compared with the corresponding preset standard conveying speed is marked as the conveying speed detection value, and the fluctuation amplitude generated by the glass transmission module during conveying is marked as the conveying unstable value. The conveying speed detection value and the conveying unstable value are numerically compared with the preset conveying speed detection threshold value and the preset conveying unstable threshold value respectively. If the conveying speed detection value or the conveying unstable value exceeds the corresponding preset threshold value, the corresponding detection time point is marked as the conveying abnormal condition time point;

[0022] The number of abnormal transport conditions during the comprehensive evaluation period is obtained and the ratio is calculated with the number of detection time points to obtain the transport abnormal condition table analysis value, and the average of all transport speed detection values ​​during the comprehensive evaluation period is calculated to obtain the transport speed table value, and the average of all transport unstable values ​​during the comprehensive evaluation period is calculated to obtain the transport stable abnormal value, and the transport pipeline analysis value is obtained by numerically calculating the transport abnormal condition table analysis value, the transport speed table value and the transport stable abnormal value.

[0023] Further, the method for analyzing and obtaining the falling plate tube analysis value is as follows:

[0024] When it is determined that the corresponding photovoltaic glass is damaged glass, the moment when the corresponding photovoltaic glass enters the automatic drop panel module is collected and marked as the entry moment, and the moment when the automatic drop panel module is opened to allow the corresponding photovoltaic glass to fall is collected and marked as the falling moment. The time difference between the falling moment and the entry moment is calculated to obtain the startup duration, and the startup efficiency inspection value is obtained by averaging all startup durations within the comprehensive evaluation period. The startup duration is numerically compared with the preset startup duration threshold, and the number of startup durations that exceed the preset startup duration threshold is marked as the startup abnormality value. The drop panel analysis value is obtained by numerically calculating the startup efficiency inspection value and the startup abnormality value.

[0025] Furthermore, the intelligent management and control platform is connected to the warning effect detection module in communication, and the warning effect detection module detects and analyzes the warning effect of the warning display module, generates a warning abnormality signal or a warning normal signal through analysis, and sends the warning abnormality signal or the warning normal signal to the mobile terminal of the manager via the intelligent management and control platform; the specific analysis process of the warning effect detection module is as follows:

[0026] A number of detection positions are set on the display surface of the display warning module, and the real-time display brightness of the corresponding detection positions is collected. The real-time display brightness of all detection positions is averaged to obtain the brightness actual display value, and the deviation value between the brightness actual display value and the set standard display brightness is marked as the brightness actual deviation value. The real-time display brightness of all detection positions is variance calculated to obtain the brightness fluctuation value, and the brightness actual deviation value and the brightness fluctuation value are numerically compared with the preset brightness actual deviation threshold and the preset brightness fluctuation threshold, respectively. If the brightness actual deviation value or the brightness fluctuation value exceeds the corresponding preset threshold, it is judged that the display warning module is in a display-poor state;

[0027] The proportion of the time that the display warning module is in the display poor meter state within a unit time is obtained and marked as the poor meter display time proportion value, and the real-time warning volume is collected when the display warning module issues a warning, and the deviation value of the actual warning volume compared with the set standard warning volume is marked as the warning sound detection value, and the warning sound detection value is numerically compared with the preset warning sound detection threshold. If the warning sound detection value exceeds the preset warning sound detection threshold, the corresponding warning sound detection value is marked as the warning sound abnormal value, and the number of warning sound abnormal values ​​per unit time is calculated by ratio to the number of warning sound detection values ​​to obtain the poor meter warning frequency proportion value;

[0028] The display warning evaluation value is obtained by numerically calculating the display time proportion value of the poor meter and the poor meter warning frequency proportion value, and the display warning evaluation value is numerically compared with the preset display warning evaluation threshold. If the display warning evaluation value exceeds the preset display warning evaluation threshold, an abnormal warning signal is generated; if the display warning evaluation value does not exceed the preset display warning evaluation threshold, a normal warning signal is generated.

[0029] Furthermore, the present invention also proposes a glass integrity detection method, comprising the following steps:

[0030] Step 1: The glass transmission module transports the photovoltaic glass that needs to be inspected for integrity;

[0031] Step 2: The brush cleaning module cleans the broken glass and dust on the surface of the photovoltaic glass;

[0032] Step 3: After cleaning, the photovoltaic glass is placed in the image collection area of ​​the image detection module;

[0033] Step 4: The image detection module collects the surface image of the photovoltaic glass, identifies the surface defects of the photovoltaic glass based on the surface image of the photovoltaic glass, and determines the integrity of the photovoltaic glass;

[0034] Step 5: If the corresponding photovoltaic glass is determined to be damaged glass, when the photovoltaic glass is transported to the top of the automatic plate dropping module, the automatic plate dropping module is opened and the damaged glass falls down.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. In the present invention, broken glass and dust on the surface of photovoltaic glass are cleaned by a brush cleaning module. After the cleaning is completed, the image detection module collects the surface image of the photovoltaic glass and determines the integrity of the photovoltaic glass, thereby realizing automatic identification and detection. When the corresponding photovoltaic glass is determined to be damaged glass, the automatic plate dropping module is controlled to open so that the damaged glass falls, thereby realizing automatic plate dropping of glass with poor integrity, reducing the workload of operators, improving the degree of automation and intelligence of the glass integrity detection process, significantly reducing the missed detection rate and improving the detection efficiency;

[0037] 2. In the present invention, before the image detection module collects the image of the photovoltaic glass, the detection accuracy impact assessment module generates a detection accuracy high impact signal or a detection accuracy low impact signal through analysis, so as to ensure the quality of the collected image, improve the accuracy of the photovoltaic glass integrity detection results and the defect recognition efficiency, and analyze the control performance of the intelligent control platform for glass integrity detection during the comprehensive evaluation period through the control abnormality comprehensive evaluation module, strengthen the supervision of the glass integrity detection process when generating the control abnormality signal, improve the detection efficiency of the glass integrity detection and the accuracy of the detection results, and have a high degree of intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0039] Figure 1 This is a first system block diagram of Embodiment 1 of the present invention;

[0040] Figure 2 is a second system block diagram of Embodiment 1 of the present invention;

[0041] Figure 3 This is a system block diagram of Embodiment 2 of the present invention;

[0042] Figure 4 This is a flow chart of the method of Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0043] 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.

[0044] Embodiment 1: Figure 1-2 As shown, a glass integrity detection system proposed by the present invention includes an intelligent management and control platform, a glass transmission module, a brush cleaning module, an image detection module, an automatic plate dropping module and a display warning module; the intelligent management and control platform controls the glass transmission module, the brush cleaning module, the image detection module and the automatic plate dropping module, the intelligent management and control platform controls the glass transmission module to transport the photovoltaic glass that needs to be inspected for integrity, the intelligent management and control platform controls the brush cleaning module to clean the broken glass and dust on the surface of the photovoltaic glass, and after the cleaning is completed, the photovoltaic glass is continuously transported through the glass transmission module to enter the imaging area of ​​the image detection module;

[0045] The intelligent management and control platform controls the image detection module to collect the surface image of the photovoltaic glass through a high-definition camera, identifies its surface defects (such as broken edges, missing corners, scratches and cracks on the outer surface of the glass) based on the surface image of the photovoltaic glass, and judges the integrity of the photovoltaic glass. The corresponding photovoltaic glass is marked as intact glass or damaged glass (if the photovoltaic glass has no surface defects, it is marked as intact glass, otherwise it is marked as damaged glass), and the marking information of the corresponding photovoltaic glass is sent to the intelligent management and control platform. The intelligent management and control platform sends the marking information of the corresponding photovoltaic glass to the display and warning module for display, so that the operator can grasp the integrity status of each photovoltaic glass detected in detail, improve the accuracy of the detection results and the detection efficiency, and realize automatic identification and detection, with a high degree of intelligence;

[0046] After the inspection is completed, the photovoltaic glass will continue to be transported through the glass transmission module. If the corresponding photovoltaic glass is damaged glass, when the photovoltaic glass is transported to the top of the automatic drop module, the corresponding control command will be issued through the intelligent management and control platform to open the automatic drop module. The automatic drop module will make the damaged glass fall down, realizing the automatic drop of glass with poor integrity, avoiding defective products from entering the next process and affecting production, improving production efficiency and reducing the workload of operators.

[0047] In addition, the intelligent management and control platform is connected in communication with the detection accuracy impact assessment module. The detection accuracy impact assessment module is used to generate a detection accuracy high impact signal or a detection accuracy low impact signal through analysis before the image detection module collects images of the photovoltaic glass, and sends the detection accuracy high impact signal or the detection accuracy low impact signal to the display warning module via the intelligent management and control platform. When the display warning module receives the detection accuracy high impact signal, it issues a corresponding warning so as to make corresponding control measures in time, thereby ensuring the quality of the collected images, which is conducive to improving the accuracy of the photovoltaic glass integrity detection results and the defect recognition efficiency; the specific operation process of the detection accuracy impact assessment module is as follows:

[0048] During the image collection process of the image detection module, the deviation value of the light brightness of the environment where the photovoltaic glass is located compared to the standard light brightness is marked as the light detection value, and the deviation value of the light color of the environment where the photovoltaic glass is located compared to the preset standard light color is marked as the light color detection value, and the gas pollution value of the environment where the photovoltaic glass is located is collected, wherein the gas pollution value is a data value indicating the concentration of dust particles in the gaseous environment;

[0049] By formula The illumination detection value GW, the light color detection value GK and the gas pollution value GN are numerically calculated to obtain the phosgene detection evaluation value GX, wherein ew1, ew2 and ew3 are preset proportional coefficients, and the values ​​of ew1, ew2 and ew3 are all greater than zero; and the larger the value of the phosgene detection evaluation value GX, the worse the imaging environment is, and the less conducive it is to ensure the quality of the collected image; the phosgene detection evaluation value GX is numerically compared with the preset phosgene detection evaluation threshold value, and if the phosgene detection evaluation value GX exceeds the preset phosgene detection evaluation threshold value, indicating that the imaging environment is poor, a high-impact signal for detection accuracy is generated;

[0050] If the phosgene evaluation value GX does not exceed the preset phosgene evaluation threshold, indicating that the imaging environment is poor, the vibration frequency and vibration amplitude of the photovoltaic glass position are collected, and the actual placement posture image of the photovoltaic glass is overlapped with the standard placement posture image to obtain the posture non-overlap value; wherein, the larger the value of the posture non-overlap value, the less the actual placement posture of the photovoltaic glass meets the requirements, which is not conducive to ensuring the quality of the collected image and needs to be adjusted;

[0051] The vibration frequency FY, the vibration amplitude FM and the attitude non-overlap value FW are numerically calculated by the formula FX=(ey1*FY+ey2*FM) / 2+ey3*FW to obtain the placement inspection value FX, wherein ey1, ey2 and ey3 are preset proportional coefficients, and the values ​​of ey1, ey2 and ey3 are all positive numbers; and the larger the value of the placement inspection value FX is, the worse the position condition of the corresponding photovoltaic glass is, and the less conducive it is to ensure the quality of the collected image;

[0052] The placement inspection value FX is numerically compared with the preset placement inspection threshold. If the placement inspection value FX exceeds the preset placement inspection threshold, it indicates that the position condition of the corresponding photovoltaic glass is poor, and a high-impact signal for detection accuracy is generated; if the placement inspection value FX does not exceed the preset placement inspection threshold, it indicates that the position condition of the corresponding photovoltaic glass is good, and a low-impact signal for detection accuracy is generated.

[0053] Furthermore, the intelligent control platform is connected to the control anomaly comprehensive evaluation module in communication. The control anomaly comprehensive evaluation module is used to set the comprehensive evaluation period, analyze the control performance of the intelligent control platform for glass integrity detection during the comprehensive evaluation period, generate a control anomaly signal or a control qualified signal through the analysis, and send the control anomaly signal or the control qualified signal to the display warning module via the intelligent control platform. When the display warning module receives the control anomaly signal, it issues a corresponding warning, so as to promptly strengthen the supervision of the glass integrity detection process in the future, thereby ensuring the smooth and stable progress of the glass integrity detection process, and improving the detection efficiency and accuracy of the detection results of the glass integrity detection, with a high degree of intelligence; the specific operation process of the control anomaly comprehensive evaluation module is as follows:

[0054] During the surface cleaning process of the corresponding photovoltaic glass, the cleaning speed fluctuation value and the cleaning pressure average value of the brush cleaning module during the cleaning process are collected, wherein the cleaning speed fluctuation value is a data value indicating the magnitude of the speed fluctuation degree during the cleaning process, and the cleaning pressure average value is a data value indicating the magnitude of the average pressure applied to the contact surface of the photovoltaic glass during the cleaning process;

[0055] The cleaning speed fluctuation value and the cleaning pressure average value are numerically compared with the preset cleaning speed fluctuation threshold value and the preset cleaning pressure average value range, respectively. If the cleaning speed fluctuation value exceeds the preset cleaning speed fluctuation threshold value or the cleaning pressure average value is not within the preset cleaning pressure average value range, indicating that the cleaning performance of the corresponding cleaning process is poor, the cleaning symbol QP-1 is assigned to the corresponding cleaning process;

[0056] In the image acquisition process of the corresponding photovoltaic glass, the number of times the detection accuracy high impact signal is generated is collected and marked as the high impact frequency, and the total time consumed for taking corresponding processing measures after the detection accuracy high impact signal is generated is marked as the processing time value, and the high impact frequency and the processing time value are numerically compared with the preset high impact frequency threshold and the preset processing time threshold, respectively. If the high impact frequency or the processing time value exceeds the corresponding preset threshold, indicating that the control performance of the corresponding image acquisition process is poor, the corresponding image acquisition process is assigned the abnormal symbol QW-1;

[0057] If there is a clearing symbol QP-1 or a sampling symbol QW-1 in the integrity detection process of the corresponding photovoltaic glass, it means that the performance of the corresponding integrity detection process is abnormal, and the integrity detection process of the corresponding photovoltaic glass is marked as a non-optimal process, and the number of non-optimal processes in the comprehensive evaluation period is obtained and its ratio to the total number of photovoltaic glasses detected in the comprehensive evaluation period is marked as the non-optimal number ratio RX;

[0058] The conveying pipe analysis value of the glass transmission module and the plate dropping pipe analysis value of the automatic plate dropping module during the comprehensive evaluation period are obtained through analysis. Specifically, several detection time points are set during the operation of the glass transmission module during the comprehensive evaluation period, and the conveying speed of the glass transmission module at the corresponding time points is collected. The deviation value of the conveying speed compared with the corresponding preset standard conveying speed is marked as the conveying speed detection value, and the fluctuation amplitude (i.e., shaking amplitude) generated by the glass transmission module during conveying is marked as the conveying unstable value;

[0059] The transport speed inspection value and the transport instability value are numerically compared with the preset transport speed inspection threshold and the preset transport instability threshold, respectively. If the transport speed inspection value or the transport instability value exceeds the corresponding preset threshold, indicating that the transport performance at the corresponding detection time point is poor, the corresponding detection time point is marked as a transport abnormality time point; the number of transport abnormality time points during the comprehensive evaluation period is obtained and the ratio is calculated with the number of detection time points to obtain the transport abnormality table value, and all the transport speed inspection values ​​during the comprehensive evaluation period are averaged to obtain the transport speed table value, and all the transport instability values ​​during the comprehensive evaluation period are averaged to obtain the transport stability abnormality value;

[0060] By formula The transport abnormality table value TR, the transport speed table value TP and the transport stability abnormality value TM are numerically calculated to obtain the transport pipe analysis value TX; wherein, kp1, kp2, kp3 are preset proportional coefficients, and the values ​​of kp1, kp2, kp3 are all positive numbers; and the larger the value of the transport pipe analysis value TX is, the worse the overall transport performance of the glass transport module is during the comprehensive evaluation period;

[0061] When the corresponding photovoltaic glass is judged to be damaged glass, the time when the corresponding photovoltaic glass enters the automatic drop plate module is collected and marked as the entry time, and the time when the automatic drop plate module is opened to allow the corresponding photovoltaic glass to fall is collected and marked as the falling time, and the time difference between the falling time and the entering time is calculated to obtain the start-up duration;

[0062] The starting efficiency inspection value is obtained by calculating the average of all starting durations in the comprehensive evaluation period, and the starting duration is numerically compared with the preset starting duration threshold, and the number of starting durations exceeding the preset starting duration threshold is marked as the starting abnormal table value. The starting efficiency inspection value LP and the starting abnormal table value LY are numerically calculated by the formula LX=b1*LP+b2*LY to obtain the drop plate analysis value LX; wherein b1 and b2 are preset proportional coefficients, and the values ​​of b1 and b2 are both positive numbers; and the larger the value of the drop plate analysis value LX, the slower the start control response of the automatic drop plate module during the comprehensive evaluation period;

[0063] The non-optimal value RX, the delivery pipe analysis value TX and the drop plate pipe analysis value LX are numerically calculated by the formula QY=(tq1*RX+tq2*TX+tq3*LX) / 3 to obtain the control abnormality value QY, wherein tq1, tq2 and tq3 are preset proportional coefficients, and the values ​​of tq1, tq2 and tq3 are all positive numbers; and the larger the value of the control abnormality value QY is, the worse the overall control performance is; the control abnormality value QY is numerically compared with the preset control abnormality threshold value, if the control abnormality value exceeds the preset control abnormality threshold value, indicating that the overall control performance is poor, then a control abnormality signal is generated; if the control abnormality value QY does not exceed the preset control abnormality threshold value, indicating that the overall control performance is good, then a control qualified signal is generated.

[0064] Embodiment 2: Figure 3 As shown, the difference between this embodiment and embodiment 1 is that the intelligent management and control platform is connected to the display warning effect inspection module in communication, and the display warning effect inspection module detects and analyzes the display warning effect of the display warning module, generates an abnormal display warning signal or a normal display warning signal through analysis, and sends the abnormal display warning signal or the normal display warning signal to the mobile terminal of the manager through the intelligent management and control platform, so that the corresponding manager can conduct cause investigation and analysis in time, and take corresponding processing measures for the display warning module to ensure its display warning effect and improve its performance; the specific analysis process of the display warning effect inspection module is as follows:

[0065] A number of detection positions are set on the display surface of the display warning module, and the real-time display brightness of the corresponding detection positions is collected. The real-time display brightness of all detection positions is averaged to obtain the brightness actual display value, and the deviation value between the brightness actual display value and the set standard display brightness is marked as the brightness actual deviation value. The real-time display brightness of all detection positions is variance calculated to obtain the brightness fluctuation value, and the brightness actual deviation value and the brightness fluctuation value are numerically compared with the preset brightness actual deviation threshold and the preset brightness fluctuation threshold, respectively. If the brightness actual deviation value or the brightness fluctuation value exceeds the corresponding preset threshold, indicating that the display state of the display warning module at the corresponding moment is not good, then it is judged that the display warning module is in a display poor state;

[0066] The proportion of the time that the display warning module is in the display poor meter state within a unit time (that is, the ratio of the time that the display warning module is in the display poor meter state to its total operation time) is obtained and marked as the poor meter display time proportion value, and the real-time warning volume is collected when the display warning module issues a warning, and the deviation value of the actual warning volume compared with the set standard warning volume is marked as the warning sound detection value, and the warning sound detection value is numerically compared with the preset warning sound detection threshold. If the warning sound detection value exceeds the preset warning sound detection threshold, the corresponding warning sound detection value is marked as the warning sound abnormal value, and the number of warning sound abnormal values ​​per unit time is calculated by ratio to the number of warning sound detection values ​​to obtain the poor meter warning frequency proportion value;

[0067] The display warning evaluation value WX is obtained by numerically calculating the bad meter display time occupation value WF and the bad meter warning frequency occupation value WY through the formula WX=a1*WF+a2*WY, wherein a1 and a2 are preset proportional coefficients, a1>a2>0; and the larger the value of the display warning evaluation value WX is, the worse the operation effect of the display warning module is overall; the display warning evaluation value WX is numerically compared with the preset display warning evaluation threshold value, if the display warning evaluation value WX exceeds the preset display warning evaluation threshold value, it indicates that the operation effect of the display warning module is poor overall, and a display warning abnormal signal is generated; if the display warning evaluation value WX does not exceed the preset display warning evaluation threshold value, it indicates that the operation effect of the display warning module is good overall, and a display warning normal signal is generated.

[0068] Embodiment 3: Figure 4 As shown, the difference between this embodiment and embodiment 1 and embodiment 2 is that a glass integrity detection method proposed in the present invention comprises the following steps:

[0069] Step 1: The glass transmission module transports the photovoltaic glass that needs to be inspected for integrity;

[0070] Step 2: The brush cleaning module cleans the broken glass and dust on the surface of the photovoltaic glass;

[0071] Step 3: After cleaning, the photovoltaic glass is placed in the image collection area of ​​the image detection module;

[0072] Step 4: The image detection module collects the surface image of the photovoltaic glass, identifies the surface defects of the photovoltaic glass based on the surface image of the photovoltaic glass, and determines the integrity of the photovoltaic glass;

[0073] Step 5: If the corresponding photovoltaic glass is determined to be damaged glass, when the photovoltaic glass is transported to the top of the automatic plate dropping module, the automatic plate dropping module is opened and the damaged glass falls down.

[0074] The working principle of the present invention is as follows: when in use, the broken glass and dust on the surface of the photovoltaic glass are cleaned by a brush cleaning module. After cleaning, the photovoltaic glass is allowed to enter the imaging area of ​​the image detection module. The image detection module collects the surface image of the photovoltaic glass, identifies its surface defects based on the surface image of the photovoltaic glass, and judges the integrity status of the photovoltaic glass to realize automatic identification and detection. When it is judged that the corresponding photovoltaic glass is damaged glass, the automatic plate dropping module is controlled to open to allow the damaged glass to fall, thereby realizing automatic plate dropping of glass with poor integrity, avoiding defective products from entering the next process and affecting production, reducing the workload of operators, significantly improving the degree of automation and intelligence of the glass integrity detection process, significantly reducing the missed detection rate and improving the detection efficiency.

[0075] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the contents of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A glass integrity detection system, characterized in that: It includes an intelligent management and control platform, a glass transmission module, a brush cleaning module, an image detection module, an automatic plate dropping module and a display warning module; the intelligent management and control platform controls the glass transmission module, the brush cleaning module, the image detection module and the automatic plate dropping module, the intelligent management and control platform controls the glass transmission module to transport the photovoltaic glass that needs to be inspected for integrity, and the intelligent management and control platform controls the brush cleaning module to clean the broken glass and dust on the surface of the photovoltaic glass. After the cleaning is completed, the photovoltaic glass is continuously transported through the glass transmission module to enter the imaging area of ​​the image detection module; The intelligent management and control platform controls the image detection module to collect the surface image of the photovoltaic glass through a high-definition camera, identifies the surface defects of the photovoltaic glass based on the surface image, and judges the integrity of the photovoltaic glass. Based on this, the corresponding photovoltaic glass is marked as intact glass or damaged glass, and the marking information of the corresponding photovoltaic glass is sent to the intelligent management and control platform. The intelligent management and control platform sends the marking information of the corresponding photovoltaic glass to the display warning module for display; After the inspection is completed, the photovoltaic glass is continuously transported through the glass transmission module. If the corresponding photovoltaic glass is damaged glass, when the photovoltaic glass is transported to the top of the automatic plate drop module, a corresponding control instruction is issued through the intelligent management and control platform to open the automatic plate drop module, and the automatic plate drop module causes the damaged glass to fall down; The intelligent control platform is connected to the control abnormality comprehensive evaluation module in communication. The control abnormality comprehensive evaluation module is used to set the comprehensive evaluation period, analyze the control performance of the intelligent control platform for glass integrity detection during the comprehensive evaluation period, generate a control abnormality signal or a control qualified signal through the analysis, and send the control abnormality signal or the control qualified signal to the display warning module via the intelligent control platform. When the display warning module receives the control abnormality signal, it issues a corresponding warning; The specific operation process of the control anomaly comprehensive evaluation module includes: In the surface cleaning process of the corresponding photovoltaic glass, the cleaning speed fluctuation value and the cleaning pressure average value of the brush cleaning module in the cleaning process are collected, and the cleaning speed fluctuation value and the cleaning pressure average value are numerically compared with the preset cleaning speed fluctuation threshold value and the preset cleaning pressure average value range, respectively. If the cleaning speed fluctuation value exceeds the preset cleaning speed fluctuation threshold value or the cleaning pressure average value is not within the preset cleaning pressure average value range, the cleaning symbol QP-1 is assigned to the corresponding cleaning process; In the image acquisition process of the corresponding photovoltaic glass, the number of times the detection accuracy high-impact signal is generated is collected and marked as the high-impact frequency, and the total time consumed for taking corresponding processing measures after the detection accuracy high-impact signal is generated is marked as the processing time value, and the high-impact frequency and the processing time value are numerically compared with the preset high-impact frequency threshold and the preset processing time threshold, respectively. If the high-impact frequency or the processing time value exceeds the corresponding preset threshold, the corresponding image acquisition process is assigned the abnormal symbol QW-1; If there is a clearing symbol QP-1 or a sampling symbol QW-1 in the integrity detection process of the corresponding photovoltaic glass, the integrity detection process of the corresponding photovoltaic glass will be marked as a non-optimal process, and the number of non-optimal processes in the comprehensive evaluation period will be obtained and its ratio to the total number of photovoltaic glasses detected will be marked as the non-optimal number ratio; The conveying pipe analysis value of the glass transmission module and the plate drop pipe analysis value of the automatic plate drop module during the comprehensive evaluation period are obtained through analysis. The non-excellent number RX, the conveying pipe analysis value TX and the plate drop pipe analysis value LX are numerically calculated by the formula QY=(tq1*RX+tq2*TX+tq3*LX) / 3 to obtain the control abnormal evaluation value QY, among which tq1, tq2, and tq3 are preset proportional coefficients, and the values ​​of tq1, tq2, and tq3 are all positive numbers; if the control abnormal evaluation value exceeds the preset control abnormal evaluation threshold, a control abnormal signal is generated; if the control abnormal evaluation value does not exceed the preset control abnormal evaluation threshold, a control qualified signal is generated; The analysis and acquisition method of the transmission pipe analysis value is as follows: During the operation of the glass transmission module during the comprehensive evaluation period, several detection time points are set, and the conveying speed of the glass transmission module at the corresponding time point is collected. The deviation value of the conveying speed compared with the corresponding preset standard conveying speed is marked as the conveying speed detection value, and the fluctuation amplitude generated by the glass transmission module during conveying is marked as the conveying unstable value. If the conveying speed detection value or the conveying unstable value exceeds the corresponding preset threshold value, the corresponding detection time point is marked as the conveying abnormal condition time point; The number of abnormal transport conditions during the comprehensive evaluation period is obtained and the ratio is calculated with the number of detection time points to obtain the transport abnormality table value, and all transport speed detection values ​​during the comprehensive evaluation period are averaged to obtain the transport speed table value, and all transport unstable values ​​during the comprehensive evaluation period are averaged to obtain the transport stable abnormality value; By formula The transport abnormal condition table value TR, the transport speed table value TP and the transport stability abnormal value TM are numerically calculated to obtain the transport pipe analysis value TX; wherein kp1, kp2, kp3 are preset proportional coefficients, and the values ​​of kp1, kp2, kp3 are all positive numbers; The analysis method for obtaining the falling plate tube analysis value is as follows: When the corresponding photovoltaic glass is judged to be damaged glass, the time when the corresponding photovoltaic glass enters the automatic drop plate module is collected and marked as the entry time, and the time when the automatic drop plate module is opened to allow the corresponding photovoltaic glass to fall is collected and marked as the falling time, and the time difference between the falling time and the entering time is calculated to obtain the start-up duration; The starting efficiency inspection value is obtained by calculating the average of all starting durations within the comprehensive evaluation period, and the number of starting durations that exceed the preset starting duration threshold is marked as the starting abnormality value. The starting efficiency inspection value LP and the starting abnormality value LY are numerically calculated using the formula LX=b1*LP+b2*LY to obtain the falling plate analysis value LX; wherein b1 and b2 are preset proportional coefficients, and the values ​​of b1 and b2 are both positive numbers.

2. A glass integrity detection system according to claim 1, characterized in that: The intelligent management and control platform is communicatively connected with the detection accuracy impact assessment module. The detection accuracy impact assessment module is used to generate a high impact signal of detection accuracy or a low impact signal of detection accuracy through analysis before the image detection module collects images of the photovoltaic glass, and send the high impact signal of detection accuracy or the low impact signal of detection accuracy to the display and early warning module via the intelligent management and control platform. When the display and early warning module receives the high impact signal of detection accuracy, it issues a corresponding early warning.

3. A glass integrity detection system according to claim 2, characterized in that: The specific operation process of the detection accuracy impact assessment module includes: During the image collection process of the image detection module, the deviation value of the light brightness of the environment where the photovoltaic glass is located compared to the standard light brightness is marked as the light detection value, and the deviation value of the light color of the environment where the photovoltaic glass is located compared to the preset standard light color is marked as the light color detection value, and the gas pollution value of the environment where the photovoltaic glass is located is collected; By formula The light detection value GW, the light color detection value GK and the gas pollution value GN are numerically calculated to obtain the phosgene detection value GX, wherein ew1, ew2 and ew3 are preset proportional coefficients, and the values ​​of ew1, ew2 and ew3 are all greater than zero; if the phosgene detection value exceeds the preset phosgene detection threshold, a high-impact signal for detection accuracy is generated; If the phosgene evaluation value does not exceed the preset phosgene evaluation threshold, the vibration frequency and vibration amplitude of the photovoltaic glass position are collected, and the actual placement posture image of the photovoltaic glass is overlapped with the standard placement posture image to obtain the posture non-overlap value; The vibration frequency FY, vibration amplitude FM and posture non-overlap value FW are numerically calculated through the formula FX=(ey1*FY+ey2*FM) / 2+ey3*FW to obtain the placement inspection value FX, where ey1, ey2, and ey3 are preset proportional coefficients, and the values ​​of ey1, ey2, and ey3 are all positive numbers; if the placement inspection value exceeds the preset placement inspection threshold, a high-impact signal for detection accuracy is generated; if the placement inspection value does not exceed the preset placement inspection threshold, a low-impact signal for detection accuracy is generated.

4. A glass integrity detection system according to claim 1, characterized in that: The intelligent management and control platform is connected to the display warning effect inspection module. The display warning effect inspection module detects and analyzes the display warning effect of the display warning module, generates abnormal display warning signals or normal display warning signals through analysis, and sends the abnormal display warning signals or normal display warning signals to the mobile terminal of the manager through the intelligent management and control platform. The specific analysis process of the display warning effect inspection module is as follows: The display warning evaluation value WX is obtained by numerically calculating the bad meter display time occupation value WF and the bad meter warning frequency occupation value WY through the formula WX=a1*WF+a2*WY, wherein a1 and a2 are preset proportional coefficients, and a1>a2>0; if the display warning evaluation value exceeds the preset display warning evaluation threshold, an abnormal display warning signal is generated; if the display warning evaluation value does not exceed the preset display warning evaluation threshold, a normal display warning signal is generated.

5. A glass integrity detection method, characterized in that: The detection method adopts the glass integrity detection system as described in any one of claims 1-4.

Images