Photovoltaic module production anomaly detection method and system, and computer program product

Abstract

The invention provides a photovoltaic module production anomaly detection method and system and a computer program product, and relates to the technical field of detection, and the detection method comprises the steps: obtaining the monitoring data of at least part of equipment on a photovoltaic module production line; acquiring abnormal data of at least part of equipment according to the monitoring data; querying an abnormal phenomenon name corresponding to the abnormal data in a first database according to the abnormal data, wherein the first database at least stores historical abnormal data of at least part of equipment and the corresponding abnormal phenomenon name; and querying an association reason and / or an association processing method of the abnormal phenomenon name in a second database according to the abnormal phenomenon name, wherein association factors and / or association processing methods of a plurality of abnormal phenomenon names are stored in the second database.

Description

Technical Field

[0001] This specification relates to the field of detection technology, and particularly to a method, a system and a computer program product for detecting production anomalies of photovoltaic modules. Background Art

[0002] A photovoltaic module production line is a complex production system that encapsulates raw materials or semi-finished products of photovoltaic modules such as solar cells into solar photovoltaic modules with power generation functions through a systematic process flow. When an abnormal situation occurs on the photovoltaic module production line, operators need to rely on the professional knowledge and historical maintenance experience accumulated through long-term work to determine and eliminate the cause of the quality problem or the cause of the failure corresponding to the abnormal situation. This places high requirements on the work experience of the operators, and it is often difficult to directly judge the quality problems or equipment failure situations of photovoltaic products that have not occurred before, resulting in low efficiency in handling abnormal events.

[0003] In view of this, the present disclosure provides a method, a system and a computer program product for detecting production anomalies of photovoltaic modules, which can quickly provide direct and accurate anomaly elimination suggestions to operators when an abnormal situation occurs on the photovoltaic module production line, and can effectively shorten the response time of abnormal events and improve the handling efficiency of abnormal events. Summary of the Invention

[0004] One or more embodiments of this specification provide a method for detecting production anomalies of photovoltaic modules, the method including: obtaining monitoring data of at least part of the equipment on the photovoltaic module production line; obtaining abnormal data of at least part of the equipment according to the monitoring data; querying the name of the abnormal phenomenon corresponding to the abnormal data in the first database, where the first database stores at least the historical abnormal data of at least part of the equipment and the corresponding names of abnormal phenomena; querying the associated cause and / or associated processing method of the abnormal phenomenon name in the second database, where the second database stores the associated factors and / or associated processing methods of multiple abnormal phenomenon names.

[0005] According to the method provided by one or more embodiments of this specification, obtaining abnormal data of at least part of the equipment according to the monitoring data includes: virtually replicating at least part of the equipment according to the simulation technology and the monitoring data of at least part of the equipment to generate a digital twin system of at least part of the equipment; monitoring at least part of the equipment through the digital twin system of at least part of the equipment, and when an abnormality is monitored, obtaining the abnormal data.

[0006] According to the method provided in one or more embodiments of the present specification, the digital twin system includes a data acquisition layer, a data processing layer, a twin model layer and an application layer; the data acquisition layer is used to obtain monitoring data of at least part of the equipment; the data processing layer is used to process and analyze the monitoring data to obtain data processing results; the twin model layer is used to construct a virtual model corresponding to at least part of the equipment according to the data processing results; the application layer is used to monitor whether the virtual model has any abnormalities during operation, and obtain abnormal data when an abnormality is detected.

[0007] According to the method provided in one or more embodiments of the present specification, at least part of the equipment is monitored through the digital twin system of at least part of the equipment, including at least one of the following: monitoring the data of at least part of the equipment in at least one state through the digital twin system of at least part of the equipment, and the at least one state includes at least one of waiting for material, material blockage, material replacement, failure, maintenance, and manual operation; monitoring at least one key parameter of at least part of the equipment through the digital twin system of at least part of the equipment; collecting and storing data of sensors of at least part of the equipment through the digital twin system of at least part of the equipment; monitoring the usage and / or remaining life of consumables of at least part of the equipment through the digital twin system of at least part of the equipment; collecting and storing alarm information of at least part of the equipment through the digital twin system of at least part of the equipment; monitoring at least one indicator of the yield, rework rate, and first-time pass rate of at least part of the processes through the digital twin system of at least part of the equipment; monitoring and storing the final inspection data of the final products of the photovoltaic module production line through the digital twin system of at least part of the equipment.

[0008] According to the method provided by one or more embodiments of the present specification, obtaining abnormal data when an abnormality is monitored includes at least one of the following: when data in the status of at least part of the monitored equipment meets abnormal conditions, determining that the data is abnormal data; when key parameters of at least part of the monitored equipment deviate from a preset range, determining that the key parameters are abnormal data; when data from sensors of at least part of the monitored equipment meets abnormal conditions, determining that the data is abnormal data; when a device alarm is monitored, determining that the alarm location and / or alarm content is abnormal data.

[0009] The method provided according to one or more embodiments of this specification further includes: performing at least one of the following through the digital twin system of at least part of the devices: analyzing the relationship between the key parameters of at least part of the devices and the production efficiency or production quality based on historical data and machine learning algorithms to obtain an analysis result, and generating parameter optimization suggestions according to the analysis result; when it is monitored that the consumables of at least part of the devices are shorter than a first preset duration from the replacement cycle or will be out of stock after a second preset duration, reminding the user to replace or supplement materials through a preset method; analyzing at least one of the yield rate, repair rate, and first-pass rate of at least part of the devices to generate quality influencing factors and / or quality improvement suggestions.

[0010] For the method provided according to one or more embodiments of this specification, the monitoring data includes the final inspection data of the final product of the photovoltaic module production line; this method further includes: when the final inspection data indicates that the final product has a downgraded situation, querying the defective code and defective location corresponding to the downgraded situation according to the historical final inspection database of the photovoltaic module production line; generating an abnormal phenomenon name according to the defective code and defective location.

[0011] For the method provided according to one or more embodiments of this specification, before querying the abnormal phenomenon name corresponding to the abnormal data in the first database according to the abnormal data, it further includes: obtaining the historical abnormal data of at least part of the devices; training the historical abnormal data according to machine learning algorithms to obtain the abnormal location and abnormal type corresponding to the historical abnormal data; obtaining a preset naming rule; generating an abnormal phenomenon name according to the preset naming rule and the abnormal location and abnormal type corresponding to the historical abnormal data; storing the abnormal phenomenon name and the corresponding abnormal location and abnormal type in the first database.

[0012] For the method provided according to one or more embodiments of this specification, querying the associated cause and / or associated processing method of the abnormal phenomenon name in the second database includes: parsing the abnormal phenomenon name through natural language processing technology to obtain a parsing result; screening relevant materials from the second database according to the parsing result; extracting the associated cause and / or associated processing method of the abnormal phenomenon name from the relevant materials.

[0013] The method provided according to one or more embodiments of this specification further includes: constructing an abnormal attribution model according to the data in the second database by using machine learning algorithms; extracting the associated cause and / or associated processing method of the abnormal phenomenon name from the relevant materials includes: inputting the relevant materials into the abnormal attribution model to obtain at least one associated cause of the abnormal phenomenon name and the likelihood score of each associated cause.

[0014] The method provided according to one or more embodiments of this specification, the monitoring data includes at least one of the following data: key parameters of at least some devices, device readings of at least some devices, consumable life data of at least some devices; querying the associated cause and / or associated handling method of the abnormal phenomenon name in the second database according to the abnormal phenomenon name, including: querying the associated cause and / or associated handling method of the abnormal phenomenon name in the second database according to the abnormal phenomenon name and the monitoring data.

[0015] The method provided according to one or more embodiments of this specification further includes: generating an abnormal troubleshooting work order according to the associated cause and / or associated handling method of the abnormal phenomenon name and pushing it to the user, and the abnormal troubleshooting work order includes the abnormal phenomenon name and the associated cause and / or associated handling method of the abnormal phenomenon name.

[0016] The method provided according to one or more embodiments of this specification, in the abnormal troubleshooting work order, sort at least one associated cause according to the likelihood score of the associated cause of the abnormal phenomenon name; this method further includes: providing at least one troubleshooting guidance to the user in sequence according to the likelihood score of at least one associated cause through the abnormal troubleshooting work order; receiving the feedback from the user on at least one troubleshooting guidance respectively; determining the abnormal cause of the abnormal phenomenon name according to the feedback.

[0017] The method provided according to one or more embodiments of this specification further includes: obtaining the troubleshooting process and materials used by the user according to the abnormal troubleshooting work order; storing the repair process and materials used in the second database.

[0018] The method provided according to one or more embodiments of this specification further includes: retrieving the repair plan of the abnormal phenomenon name in the third database according to the abnormal cause of the abnormal phenomenon name; generating a repair work order according to the abnormal cause and the repair plan and pushing it to the user.

[0019] The method provided according to one or more embodiments of this specification further includes: obtaining the repair process and materials used by the user according to the repair work order; storing the repair process and materials used in the third database.

[0020] One or more embodiments of this specification also provide a photovoltaic module production anomaly detection system, which includes: a first acquisition module for acquiring monitoring data of at least some devices on a photovoltaic module production line; a second acquisition module for acquiring anomaly data of at least some devices according to the monitoring data; a first query module for querying the anomaly phenomenon name corresponding to the anomaly data in a first database, where the first database stores at least the historical anomaly data of at least some devices and their corresponding anomaly phenomenon names; a second query module for querying the associated cause and / or associated processing method of the anomaly phenomenon name in a second database, where the second database stores the associated factors and / or associated processing methods of multiple anomaly phenomenon names.

[0021] One or more embodiments of this specification also provide a computer program product, including computer instructions or a computer program. When at least a part of the computer instructions or the computer program is executed by a processor, it can implement the map generation method provided by the embodiments of this specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] This specification will be further described by way of exemplary embodiments, which will be described in detail through the drawings. The same numbers in the drawings represent the same structures or steps.

[0023] Figure 1 is an example diagram of the operation process of a photovoltaic module production anomaly detection system shown in some embodiments of this specification.

[0024] Figure 2 is an exemplary flowchart of a photovoltaic module production anomaly detection method shown in some embodiments of this specification.

[0025] Figure 3 is an exemplary flowchart of acquiring anomaly data of at least some devices according to monitoring data shown in some embodiments of this specification.

[0026] Figure 4 is an exemplary structural diagram of a digital twin system shown in some embodiments of this specification.

[0027] Figure 5 is a functional module structural diagram of a digital twin system shown in some embodiments of this specification.

[0028] Figure 6 is a functional module structural diagram of an anomaly phenomenon summary subsystem shown in some embodiments of this specification.

[0029] Figure 7 is an exemplary flowchart of configuring a first database shown in some embodiments of this specification.

[0030] Figure 8 It is an exemplary structural diagram of an anomaly attribution subsystem shown in some embodiments of this specification.

[0031] Figure 9 It is an exemplary structural diagram of a troubleshooting work order subsystem shown in some embodiments of this specification.

[0032] Figure 10 It is an exemplary flowchart of another photovoltaic module production anomaly detection method shown in some embodiments of this specification.

[0033] Figure 11 It is an exemplary structural diagram of a maintenance work order subsystem shown in some embodiments of this specification.

[0034] Figure 12 It is an exemplary diagram of the operation process of another photovoltaic module production anomaly detection system shown in some embodiments of this specification.

[0035] Figure 13 It is an exemplary structural diagram of the functional modules of a photovoltaic module production anomaly detection system shown in some embodiments of this specification. Detailed implementation manners

[0036] To more clearly illustrate the technical solutions of the embodiments of this specification, the embodiments will be introduced in detail below with reference to the accompanying drawings. Obviously, the content described below is some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, the technical solutions or means disclosed in this specification can also be applied to other scenarios based on these technical contents.

[0037] It should be understood that the "system", "device", "unit" and / or "module" used in this specification is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, they can be replaced by other expressions.

[0038] Unless otherwise specified, the technical terms describing components, elements, etc. in this specification do not specifically refer to the singular, but may also include the plural. Generally speaking, terms such as "including" and "comprising" only indicate the inclusion of the clearly identified steps, elements or components, and these steps, elements and components do not constitute an exclusive list. For example, the described method or device may also include other steps or components.

[0039] In this specification, flowcharts are used to illustrate the operation steps performed by the devices or systems of the relevant embodiments. However, without special instructions, the order in which these steps are described should not be construed as a limitation on the order of step execution. Those of ordinary skill in the art can adjust the order of execution of these steps based on the knowledge and information conveyed by the embodiments of this specification. The adjustments include, but are not limited to, swapping the order of precedence, merging multiple steps, and splitting a certain step.

[0040] A photovoltaic module production line is a production system that encapsulates raw materials or semi-finished products of photovoltaic modules, such as solar cells, into solar photovoltaic modules with power generation functions through a systematic process flow. Its core processes include key processes such as solar cell sorting, welding, lamination, encapsulation, framing, and testing. Combining automated equipment, such as string welders and laminators, and intelligent control systems, such as Programmable Logic Controllers (PLCs) and Manufacturing Execution Systems (MESs), ensures high-efficiency and high-quality mass production.

[0041] In a photovoltaic module production line, there are multiple components and devices for producing, assembling, and testing photovoltaic modules. When a certain device in the photovoltaic module production line is sensed to have a fault, the cause of the fault may be that the device itself that is sensed to have a fault has a fault, or it may be that other associated devices on the photovoltaic module production line have faults, which in turn cause the device that is sensed to have a fault to malfunction. In one example, when a device fault warning message appears for the string welder device on the photovoltaic module production line, it may be caused by a fault in the previous device of the string welder device, resulting in continuous empty feeding in the feeding area of the string welder device, or it may be caused by the occlusion of a grating in a certain welding area of the string welder device. The above fault causes may all result in a device fault warning message for the string welder device. When a quality problem is sensed during the inspection of the photovoltaic modules produced by the photovoltaic module production line, the cause of the quality problem may be that some production devices in the photovoltaic module production line have faults, or it may be due to operator errors during the operation of the production devices. In one example, when an abnormal situation such as broken solar cells occurs in the photovoltaic cells produced on the photovoltaic module production line, it may be caused by the accumulation of solder tapes during the welding process, causing the solder beads to pierce the photovoltaic cells during the welding process, or it may be caused by incorrect operation of the operator during the film feeding process in the lamination process. The above abnormal causes may all result in the quality problem of broken solar cells in the photovoltaic cells.

[0042] As can be seen from the above description, for the quality problems of photovoltaic products or equipment failures that occur on the photovoltaic module production line, there may be more than one cause for the resulting quality problems or failure reasons; since different quality problem causes or failure reasons require different maintenance or response methods to be eliminated, it often depends on the professional knowledge and historical maintenance experience accumulated by the operators themselves over a long period of time to determine the quality problem causes or failure reasons, and it is often difficult to directly judge the quality problem causes or failure reasons for newly emerging quality problems of photovoltaic products or equipment failure situations. Therefore, when abnormal situations including quality problems of photovoltaic products and equipment failures occur on the photovoltaic module production line, it is often impossible to obtain the corresponding abnormal reasons for the abnormal situations in a timely manner due to the lack of immediately available professional knowledge and historical maintenance experience, and thus it is impossible to respond to and eliminate the abnormal situations in a timely manner based on the abnormal reasons, resulting in low processing efficiency of the abnormal situations and easily affecting the stable operation of the photovoltaic module production line and the quality of the produced photovoltaic module products.

[0043] In view of this, the embodiments of this specification provide a method, system, storage medium, and computer device for detecting production anomalies of photovoltaic modules, which can quickly match the abnormal situation with the historical maintenance event records stored in the database when an abnormal situation occurs on the photovoltaic module production line, obtain historical maintenance cases similar to or the same as the current abnormal situation, and thus quickly provide direct and accurate abnormal elimination suggestions to the operators, which can effectively shorten the response time of abnormal events and improve the handling efficiency of abnormal events.

[0044] Figure 1 It is a flowchart example of the operation of a photovoltaic module production anomaly detection system according to some embodiments of this specification. In some embodiments, as Figure 1 shown, the photovoltaic module production anomaly detection system 100 may include an abnormal data acquisition subsystem 110, an abnormal phenomenon summarization subsystem 120, and an abnormal phenomenon classification subsystem 130.

[0045] In some embodiments, the abnormal data acquisition subsystem 110 is configured to acquire abnormal data that occurs during the daily use or commissioning and preparation process of the photovoltaic module production line. In some embodiments, the abnormal data may be generated by equipment failures of at least some of the equipment on the photovoltaic module production line, or may be caused by operational mistakes or incorrect operations of the operators on at least some of the equipment on the photovoltaic module production line. In some embodiments, the abnormal data may be the equipment parameters corresponding to various types of equipment on the photovoltaic module production line, the consumable material parameters corresponding to various types of equipment on the photovoltaic module production line, or the product parameters corresponding to the finished photovoltaic modules or semi-finished photovoltaic modules produced or assembled on the photovoltaic module production line. In some embodiments, the abnormal data acquisition subsystem 110 may acquire the monitoring data of at least some of the equipment on the photovoltaic module production line regularly or irregularly, and compare the acquired monitoring data with the preset standard data values or standard data ranges. When there is a difference between the monitoring data and the preset standard data values or standard data ranges, it is determined that the monitoring data belongs to the abnormal data that needs to be acquired. In some embodiments, some of the equipment on the photovoltaic module production line has an autonomous alarm function, and the abnormal data acquisition subsystem 110 may acquire the autonomous alarm information of these equipment and use the autonomous alarm information of these equipment as the abnormal data of the photovoltaic module production line.

[0046] In some embodiments, the abnormal phenomenon summarization subsystem 120 is configured to summarize abnormal phenomena based on the abnormal data obtained by the abnormal data acquisition subsystem 110. Based on the relevant descriptions of the foregoing embodiments, it can be understood that when abnormal data appears during the daily use or commissioning and preparation of the photovoltaic module production line, it is often difficult for the operator to determine which component of the photovoltaic module production line the abnormal data is caused by, whether it is a fault or an operator's misoperation, thereby making it difficult to respond to the occurrence of abnormal data in a timely manner. In the above embodiments, in order to respond to the occurrence of abnormal data in a timely manner, the abnormal phenomenon summarization subsystem 120 may summarize and judge the acquired abnormal data. Specifically, in a certain example, as Figure 1 shown, the abnormal phenomenon summarization subsystem 120 may be connected to the first database 121, where the first database 121 stores abnormal phenomenon information such as the abnormal occurrence location and abnormal type corresponding to the abnormal data. The abnormal phenomenon summarization subsystem 120 may retrieve the acquired abnormal data through the abnormal phenomenon information stored in the first database 121 to obtain abnormal phenomenon information such as the abnormal occurrence location and abnormal type corresponding to the abnormal data, and summarize the abnormal phenomenon corresponding to the abnormal data based on the above abnormal phenomenon information.

[0047] In some embodiments, the anomaly attribution subsystem 130 is configured to attribute anomalies based on the anomalies obtained by the anomaly summarization subsystem 120. Based on the relevant descriptions of the foregoing embodiments, it can be understood that for the anomalies occurring in the photovoltaic module production line, there may be multiple anomaly causes, and each anomaly cause may have different response and elimination methods. It is often difficult for operators to determine which response means to take to handle the anomalies, thereby reducing the processing efficiency of the anomalies occurring in the photovoltaic module production line. In the above embodiments, in order to improve the processing efficiency of anomalies, the anomaly attribution subsystem 130 can obtain associated causes and / or associated processing methods for the obtained anomalies. Specifically, in one example, as Figure 1 shown, the anomaly attribution subsystem 130 can be connected to the second database 131, where the second database 131 stores the associated causes and / or associated processing methods corresponding to the anomalies. The anomaly attribution subsystem 130 can retrieve the obtained anomalies through the data stored in the second database 131 to obtain the associated causes and / or associated processing methods corresponding to the anomalies, so that operators can respond to and handle the anomalies occurring in the photovoltaic module production line in a timely manner.

[0048] Figure 2 FIG. is an exemplary flowchart of a method for detecting anomalies in photovoltaic module production according to some embodiments of the present specification. In some embodiments, process 200 can be implemented by a photovoltaic module production anomaly detection system 100 deployed on a processing device or a terminal device, where the processing device or the terminal device can be a client device, or a server device or a cloud device. In some embodiments, process 200 can be implemented by a photovoltaic module production anomaly detection system 1300 deployed on a processing device or a terminal device, where the processing device or the terminal device can be a client device, or a server device or a cloud device.

[0049] In some embodiments, as Figure 2 shown, process 200 may include the following steps.

[0050] Step 210: Obtain monitoring data of at least some devices on the photovoltaic module production line. In some embodiments, step 210 can be implemented by a first acquisition module 1310.

[0051] In some embodiments, the monitoring data refers to various data information collected during the process of irregular or continuous observation, detection, and monitoring of a photovoltaic module production line, and can be used to evaluate the overall operating status of the photovoltaic module production line. In some embodiments, the monitoring data may include key parameters of at least some of the equipment on the photovoltaic module production line, equipment readings of at least some of the equipment, consumable life data of at least some of the equipment, and detection data of at least some of the photovoltaic modules, which are not limited herein.

[0052] Step 220: Obtain abnormal data of at least some of the equipment according to the monitoring data. In some embodiments, step 220 can be implemented by the second acquisition module 1320. In some embodiments, step 220 and the foregoing step 210 can be jointly implemented by the abnormal data acquisition subsystem 110.

[0053] In some embodiments, the abnormal data refers to a set of monitoring data that deviates from the normal working state or normal debugging state of the photovoltaic module production line. In some embodiments, the abnormal data may include key parameters of some of the equipment on the photovoltaic module production line that deviate from the preset range. For example, the preset welding temperature fluctuation of the string welding machine equipment on the photovoltaic module production line needs to be less than 2°C under normal working conditions. When the monitoring data reflects that the welding temperature fluctuation of the string welding machine equipment during operation is greater than 2 degrees Celsius, it indicates that abnormal data affecting welding consistency has occurred. In some embodiments, the abnormal data may include monitoring data obtained by the sensing equipment on the photovoltaic module production line that meets the abnormal conditions. For example, if the vision sensing equipment detects an abnormal condition that there are debris in the feeding area of the string welding machine equipment, it indicates that abnormal data of debris in the feeding area has occurred. In some embodiments, the abnormal data may include alarm information of the equipment on the photovoltaic module production line. For example, when the string welding machine equipment has abnormal accumulation of soldering tapes, corresponding alarm information will be generated, and the above alarm information can be obtained as abnormal data.

[0054] Step 230: Query the name of the abnormal phenomenon corresponding to the abnormal data in the first database, where the first database stores at least historical abnormal data of at least some of the equipment and their corresponding abnormal phenomenon names. In some embodiments, step 230 can be implemented by the abnormal phenomenon summary subsystem 120. In some embodiments, step 230 can be implemented by the first query module 1330.

[0055] In some embodiments, the first database may store abnormal data and their corresponding abnormal phenomenon names, where the generation of the abnormal phenomenon names is unique and can accurately describe the actually occurring abnormal phenomena. For example, for the string welding machine equipment in a photovoltaic module production line, when the abnormal data simultaneously includes the situation that the welding tapes of the string welding machine are crossed and there are failure areas in the EL (Electroluminescence Testing) images of the photovoltaic cells (i.e., the images of the photovoltaic cells appear completely black), the corresponding abnormal phenomenon name is "welding tape overlapping short circuit". The abnormal phenomenon name "welding tape overlapping short circuit" describes the actual abnormal situation, facilitating the operators to intuitively understand the abnormal failure situation, and at the same time being unique. Only when the above two conditions are simultaneously included in the abnormal data does it point to the abnormal phenomenon name "welding tape overlapping short circuit". Through the first database, the complex abnormal phenomena monitored on the photovoltaic module production line can be transformed into a description with a clear direction, i.e., the abnormal phenomenon name. In some embodiments, the abnormal data stored in the first database and their corresponding abnormal phenomenon names can be preset by the operators based on historical abnormal troubleshooting experience.

[0056] In some embodiments, to ensure that the data in the first database can be continuously updated to adapt to the continuous update and iteration of the photovoltaic module production line, the first database may store the historical abnormal data of at least some of the equipment on the photovoltaic module production line and their corresponding abnormal phenomenon names, that is, the historical abnormal data that has occurred on the photovoltaic module production line and their corresponding abnormal phenomenon names can be stored in the first database to achieve continuous update of the first database.

[0057] Step 240: Query the associated causes and / or associated handling methods of the abnormal phenomenon name in the second database, where the second database stores the associated factors and / or associated handling methods of multiple abnormal phenomenon names. In some embodiments, step 240 can be implemented by the abnormal phenomenon classification subsystem 130. In some embodiments, step 240 can be implemented by the second query module 1340.

[0058] In some embodiments, the second database may store correlation factors and / or correlation processing methods for multiple abnormal phenomenon names. In some embodiments, the correlation factors may be the abnormal causes that lead to the occurrence of the abnormal phenomenon names. For example, for the abnormal phenomenon name of "battery calendering fragmentation", the corresponding correlation factors may include that the string welding machine magazine scratches the edge of the battery chip and causes damage, and then the chip is fragmented under stress during the lamination process, or the chip is fragmented due to excessive pressure during the calendering process. Operators can check and repair the positions and equipment that may be abnormal in the photovoltaic module production line according to the above correlation factors. In some embodiments, the correlation processing methods may be methods that can process or eliminate the abnormal conditions corresponding to the abnormal phenomenon names. Operators can execute the correlation processing methods to restore the normal production condition or normal debugging condition of the photovoltaic module production line. In some embodiments, when the abnormal phenomenon name corresponding to the abnormal data is obtained in the foregoing step 230, the correlation factors and / or correlation processing methods of the abnormal phenomenon name can be queried and obtained through the second database, so that operators can respond in a timely manner to the abnormal conditions generated in the photovoltaic module production line according to the obtained correlation factors and / or correlation processing methods.

[0059] In some embodiments, for the obtained abnormal phenomenon name, multiple correlation factors and / or correlation processing methods may be queried. The possibility of the multiple queried correlation factors can be further evaluated for the reference of operators to help them quickly determine the true abnormal cause of the abnormal condition. The specific implementation of querying in the second database to obtain the correlation cause and / or correlation processing method of the abnormal phenomenon name will be further described later and will not be elaborated here.

[0060] The following will further describe the specific implementation of the above steps 210 to 240 in combination with the embodiments provided in this specification.

[0061] In some embodiments, the monitoring data can be automatically obtained regularly or irregularly through the sensing devices and / or detection equipment set on the photovoltaic module production line. For example, it can be the image monitoring data of the feeding area obtained by the visual sensing equipment that continuously monitors the feeding area of the string welding machine equipment on the photovoltaic module production line. In some embodiments, the monitoring data can be obtained by the equipment on the photovoltaic module production line through its own built-in function modules regularly or irregularly. For example, it can be the soldering station temperature data and welding pressure data of the string welding machine equipment on the photovoltaic module production line. In some embodiments, the monitoring data can be obtained by manually reading the display readings of the equipment on the photovoltaic module production line.

[0062] Figure 3is an exemplary flowchart for obtaining abnormal data of at least some devices according to monitoring data as shown in some embodiments of this specification. In some embodiments, Figure 3 the process 300 shown may include the following steps.

[0063] Step 310: Virtually replicate at least some devices according to simulation technology and monitoring data of at least some devices to generate a digital twin system of at least some devices. Among them, a digital twin system is a virtual model that uses digital technology to construct a dynamic and real-time mapping for physical entities (such as devices, production lines, buildings, or cities). Its core is to synchronize the state of the physical world with the virtual model in real time through technologies such as the Internet of Things (IoT), big data, artificial intelligence (AI), and cloud computing, supporting full-life cycle management and intelligent decision-making. In some embodiments, applying the digital twin system to the abnormal detection of a photovoltaic module production line can virtually replicate the physical devices in the photovoltaic module production line through the digital twin system, and simulate the operating state of the physical devices through the replicated virtual devices, so as to realize the early prediction of some possible abnormal situations in the photovoltaic module production line and provide decision-making support when abnormal situations occur in the photovoltaic module production line. In some embodiments, the monitoring data of at least some devices on the photovoltaic module production line can be collected, and computer-aided design (such as CAD, etc.) software or simulation software (such as ANSYS, MATLAB, Simulink, etc.) can be used to establish a virtual model of the device, and the collected monitoring data can be simulated and run through the virtual model to realize the virtual replication of at least some devices. Through the setting of the digital twin system, it is possible to respond, identify, and analyze in a timely manner when abnormal situations occur in the photovoltaic module production line, reducing the impact of abnormal situations on the production activities of the photovoltaic module production line. The specific settings of the digital twin system will be described in detail later and will not be elaborated here.

[0064] Step 320: Monitor at least some devices through the digital twin system of at least some devices, and obtain abnormal data when an abnormality is detected. In some embodiments, the digital twin system obtained through the aforementioned step 310 can simulate the operating state of some devices on the photovoltaic module production line through the obtained monitoring data, monitor the operating state of these devices, and obtain corresponding abnormal data when an abnormality is detected. The monitoring of devices by the digital twin system and the acquisition of abnormal data will be described in detail later and will not be elaborated here.

[0065] Figure 4 is an exemplary structural diagram of a digital twin system as shown in some embodiments of this specification. In some embodiments, such as Figure 4As shown, the digital twin system 400 may include a data acquisition layer 410, a data processing layer 420, a twin model layer 430, and an application layer 440.

[0066] In some embodiments, the data acquisition layer 410 is used to obtain the monitoring data of at least some devices. In some embodiments, the data acquisition layer 410 can implement the acquisition of the monitoring data of at least some devices on the photovoltaic module production line through the relevant descriptions provided in the foregoing embodiments, which will not be elaborated here.

[0067] In some embodiments, the data processing layer 420 is used to process and analyze the monitoring data obtained by the data acquisition layer 410, and then obtain a data processing result. In some embodiments, the data processing layer 420 can use various data processing and analysis techniques including statistics, machine learning, and data mining to process and analyze the collected monitoring data, and extract information meaningful for the identification of abnormal conditions and the analysis of abnormal causes of the photovoltaic module production line. In some embodiments, according to the monitoring data of the string welding machine equipment under normal working conditions of the photovoltaic module production line, the monitoring data range of the string welding machine equipment under normal working conditions can be determined by statistical and / or machine learning methods, providing a basis for subsequent monitoring and judgment of whether the string welding machine equipment is in an abnormal state. In other embodiments, according to the monitoring data of the EL detection results of photovoltaic cells in the photovoltaic module production line obtained by the data acquisition layer 410, the frequency data of failures of photovoltaic cells within a certain period can be obtained through statistical and / or machine learning methods, and the frequency data of failures of photovoltaic cells is correlated with other monitoring data of the photovoltaic module production line obtained by the data acquisition layer 410, such as monitoring data of transmission belt breakage, etc., to obtain the correlation between the frequency of failures of photovoltaic cells and specific abnormal conditions in the photovoltaic module production line, providing a basis for subsequent judgment of the abnormal causes of EL detection failures of photovoltaic cells.

[0068] In some embodiments, the twin model layer 430 is used to construct virtual models corresponding to at least some of the devices according to the data processing results. In some embodiments, the twin model layer 430 can construct virtual models corresponding one-to-one to the physical devices on the photovoltaic module production line according to the data processing results. These virtual models can not only accurately reflect the current state of the physical devices on the photovoltaic module production line in combination with the monitoring data, but also simulate the future behavior of the physical devices to achieve the early prediction of some possible abnormal situations of the physical devices. For example, the early prediction when the consumables of the physical device are approaching the need for replacement. In a certain example, according to the data processing results of the data processing layer 420, such as the time between two replacements of the soldering tape consumables in the photovoltaic module production line, etc., the historical usage of the soldering tape consumables of the stringing machine device in the photovoltaic module production line can be obtained. When an abnormal situation related to the soldering tape consumables occurs or the usage time after the replacement of the soldering tape consumables is close to the average service life in the historical usage, it can be predicted that the soldering tape consumables may need to be replaced abnormally.

[0069] In some embodiments, the application layer 440 is used to monitor whether an abnormality occurs during the operation of the virtual model, and obtain abnormal data when an abnormality is monitored. In some embodiments, the application layer 440 can continuously monitor the operating state during the process of simulating the operating state of the virtual model corresponding to the physical device to determine whether an abnormality occurs, and when an abnormality occurs, obtain the monitoring data related to the abnormality as abnormal data. In some embodiments, the application layer 440 can further process and analyze the monitoring data related to the abnormality through the data processing layer 420 when an abnormality occurs to determine the location of the abnormality or the specific content of the abnormality, and use the location of the abnormality and the specific content of the abnormality as at least a part of the obtained abnormal data.

[0070] In some embodiments, the digital twin system can monitor at least some of the devices on the photovoltaic module production line, specifically including monitoring at least one or any combination of multiple items such as the usage and working status of at least some of the devices, the key parameters of at least some of the devices, the sensor readings of at least some of the devices, the consumable usage of at least some of the devices, the alarm information of at least some of the devices, the detection indexes of at least some of the processes, and the final inspection data of the products on the photovoltaic module production line.

[0071] In some embodiments, monitoring the usage and working status of at least some devices may include real-time monitoring of the device status through the digital twin system of the device, where the device status may include material waiting status, material blocking status, material change status, fault status, maintenance status, manual operation status, etc.; the digital twin system can comprehensively evaluate the usage, operation, and maintenance of the device based on the obtained monitoring data to determine the current device status to which the device belongs. In some embodiments, the digital twin system can also intuitively display the monitored device status through the display interface or graphical user interface of the terminal device, so as to facilitate the operator to quickly obtain the device status and respond in a timely manner when the device status is in abnormal statuses such as material waiting status, material blocking status, and fault status. In some embodiments, the digital twin system also supports automatically triggering an early warning mechanism when it monitors that the device status is abnormal or the device is about to require maintenance, and pushing relevant early warning information to the operator through methods such as email, text message, and application push, so as to ensure that the operator can timely know the relevant early warning events and respond in a timely manner. In some embodiments, monitoring the usage and working status of at least some devices can be implemented through the status monitoring module 510 of the digital twin system.

[0072] In some embodiments, monitoring the key parameters of at least some devices may include real-time and continuous monitoring of the key parameters that affect the normal use of the device through the digital twin system of the device, where the key parameters may include the operating power of the device, continuous operating time, etc. In some embodiments, monitoring the key parameters of at least some devices can be implemented through the parameter monitoring module 520 of the digital twin system.

[0073] In some embodiments, monitoring the sensor readings of at least some devices may include collecting and storing the readings of the built-in or external sensing devices of the device through the digital twin system of the device, where the reading content of the sensing device may include reading information such as the welding temperature of the device, lamination pressure, and ambient humidity, and the digital twin system can store the collected reading information through a memory. In some embodiments, monitoring the sensor readings of at least some devices can be implemented through the sensing reading monitoring module 530 of the digital twin system.

[0074] In some embodiments, monitoring the consumable usage of at least some devices may include monitoring and recording the consumable usage of the device through the digital twin system of the device. In some embodiments, it may also include providing the digital twin system to monitor and record the remaining life of the current consumables of the device. In some embodiments, monitoring the consumable usage of at least some devices can be implemented through the consumable monitoring module 540 of the digital twin system.

[0075] In some embodiments, monitoring the alarm information of at least some devices may include collecting and storing the alarm information autonomously feedback by the devices through the digital twin system of the devices. The alarm information may include the alarm time, alarm type, alarm reason, etc. of the devices. For example, for the string welding machine device in a photovoltaic module production line, the alarm time may be the time information corresponding to the triggered alarm information; the alarm type may include prompt alarms, shutdown alarms, etc., and the operators can classify the alarm types according to the actual business; the alarm reason may include that the grating in the welding area is blocked, the safety door in the welding area is not closed, there are debris in the feeding area, etc. In some embodiments, monitoring the alarm information of at least some devices may be implemented through the alarm monitoring module 550 of the digital twin system.

[0076] In some embodiments, monitoring the detection indexes of at least some processes may include continuously monitoring the detection indexes of the processes through the digital twin systems of the devices participating in the processes in a photovoltaic module production line. The detection indexes of the processes include the yield, repair rate, first-pass rate, etc. of the semi-finished or final photovoltaic modules obtained through the processes. In some embodiments, monitoring the detection indexes of at least some processes may be implemented through the index monitoring module 560 of the digital twin system.

[0077] In some embodiments, monitoring the final inspection data of a photovoltaic module production line may include monitoring and storing the final inspection data of the final products of the photovoltaic module production line through the digital twin system. The final inspection data may include the inspection results corresponding to the inspection processes for whether the final products have degradation conditions, specifically including the number of degraded products, the reasons for degradation, etc. The digital twin system may store the collected final inspection data through a memory. In some embodiments, monitoring the final inspection data of a photovoltaic module production line may be implemented through the final inspection monitoring module 570 of the digital twin system.

[0078] Figure 5 is a structural diagram of the functional modules of a digital twin system shown in some embodiments of this specification. In some embodiments, as Figure 5 shown, the digital twin system 500 may include a status monitoring module 510, a parameter monitoring module 520, a sensing reading monitoring module 530, a consumable monitoring module 540, an alarm monitoring module 550, an index monitoring module 560, and a final inspection monitoring module 570. For more content about each module, reference can be made to the relevant descriptions in the foregoing embodiments, which will not be elaborated here. In some embodiments, the digital twin system may include at least one functional module or any combination of multiple functional modules among the above-mentioned status monitoring module 510 to final inspection monitoring module 570 according to the actual anomaly detection needs of the photovoltaic module production line.

[0079] In some embodiments, the digital twin system is capable of acquiring relevant abnormal data when an abnormality occurs in monitoring. In some embodiments, as Figure 5 shown, the status monitoring module 510 in the digital twin system 500 can determine the data as abnormal data when it monitors that the data in the status of at least some devices meets the abnormal conditions. For example, when it monitors that the operating status of the string welding machine device is a material blockage status, since the material blockage status is one of the abnormal conditions preset by the operator, the monitored material blockage status data can be determined as the abnormal data of the photovoltaic module production line.

[0080] In some embodiments, as Figure 5 shown, the parameter monitoring module 520 in the digital twin system 500 can determine the above key parameters as abnormal data when it monitors that the key parameters of at least some devices deviate from the preset range. For example, when it monitors that the operating power of the string welding machine device exceeds the preset rated power range, the operating power parameter of the string welding machine device and the monitored operating power parameter value are determined as the abnormal data of the photovoltaic module production line.

[0081] In some embodiments, as Figure 5 shown, the sensing reading monitoring module 530 in the digital twin system 500 can also determine the above reading data as abnormal data when it monitors that the data of the sensors of at least some devices meets the abnormal condition reading data. For example, when it monitors that the welding temperature of the string welding machine device deviates from the preset normal operating temperature range, the welding temperature reading obtained by the string welding machine device through the sensor and the monitored welding temperature reading value are determined as the abnormal data of the photovoltaic module production line.

[0082] In some embodiments, as Figure 5 shown, the alarm monitoring module 550 in the digital twin system 500 can, when it monitors that an alarm message appears on a device, determine the alarm location and / or alarm content based on the alarm message, and use the determined alarm location and / or alarm content as abnormal data. For example, when it monitors that a prompt-type alarm message indicating that the grating in the welding area of the string welding machine device is blocked appears, the string welding machine device can be used as the alarm location, and the blocking of the grating in the welding area can be used as the alarm content, and the above alarm location and alarm content are determined as the abnormal data of the photovoltaic module production line.

[0083] In some embodiments, as Figure 5The index monitoring module 560 in the digital twin system 500 shown can, when it monitors that the detection indexes of some processes deviate from the preset range, determine the abnormal product type and / or abnormal process position corresponding to the above processes as abnormal data. For example, when it monitors that the yield of the products processed by the string welding process deviates from the preset range, the low yield of the products processed by the string welding process can be used as the abnormal product type, and the position of the string welding process can be used as the abnormal process position, and the above abnormal product type and abnormal process position are determined as the abnormal data of the photovoltaic module production line.

[0084] In some embodiments, the digital twin system can give an optimization plan for the use of some equipment and / or the implementation of some processes in the photovoltaic module production line according to the obtained monitoring data. In some embodiments, as Figure 5 The parameter monitoring module 520 in the digital twin system 500 shown can, based on the historical monitoring data collected and stored, analyze the relationship between the key parameters of at least some equipment and the production efficiency or production quality through a machine learning algorithm to obtain an analysis result, and give corresponding parameter optimization suggestions according to the analysis result.

[0085] In some embodiments, as Figure 5 The consumable monitoring module 540 in the digital twin system 500 shown can, based on the monitored consumable usage situation and the remaining life of the consumables, judge whether the consumables are about to be used up and need to remind the operators to replace them in time. For example, when it monitors that the consumables of at least some equipment are less than the first preset duration from the replacement cycle or will be out of stock after the second preset duration, it can remind the operators to replace or replenish the consumables in time through a preset method, such as text message, email, application notification, etc., to avoid the equipment being difficult to work normally due to the lack of consumables.

[0086] In some embodiments, as Figure 5 The index monitoring module 560 in the digital twin system 500 shown can, based on the monitored process detection indexes, analyze at least one of the yield, repair rate, and first-pass rate of at least some processes, find out the key factors affecting the quality of the process products, and generate quality impact factors and / or quality improvement suggestions, so as to improve the product processing quality and production efficiency corresponding to each process in the photovoltaic module production line.

[0087] In some embodiments, the digital twin system can directly generate the name of the abnormal phenomenon that may cause the final product to be downgraded according to the final inspection data of the final product of the photovoltaic module production line in the obtained monitoring data. In some embodiments, as Figure 5The final inspection monitoring module 570 in the digital twin system 500 shown can, when the final inspection data indicates a degradation situation of the final product of the photovoltaic module production line, query the defect code and defect location corresponding to the degradation situation according to the historical final inspection database of the photovoltaic module production line, and then generate the name of the abnormal phenomenon according to the defect code and defect location. The historical final inspection database can store the historical records of the degradation reasons of each component in the final product, which can specifically include information such as the degradation reason and defect location of each component.

[0088] Figure 6 It is a functional module structure diagram of an abnormal phenomenon summary subsystem shown according to some embodiments of this specification. In some embodiments, as Figure 6 shown, the abnormal phenomenon summary subsystem 600 may include an abnormal situation analysis module 610 and an abnormal name query module 620.

[0089] In some embodiments, the abnormal situation analysis module 610 is used to analyze the abnormal data to obtain the abnormal location and / or abnormal type corresponding to the abnormal situation. In some embodiments, the abnormal situation analysis module 610 may include a pre-trained artificial intelligence analysis model, and use machine learning algorithms to train the historical abnormal data to improve the accuracy of identifying and classifying the abnormal situation, and be able to classify the obtained abnormal data into predefined abnormal types, and obtain the corresponding abnormal location and / or abnormal type according to the predefined abnormal type.

[0090] In some embodiments, the abnormal name query module 620 is used to obtain the name of the abnormal phenomenon corresponding to the abnormal data by retrieving and querying in the first database according to the abnormal location and / or abnormal type obtained by the abnormal situation analysis module 610. In some embodiments, the first database stores the names of abnormal phenomena and the corresponding abnormal locations and / or abnormal types, and the operator can intuitively obtain the name of the abnormal phenomenon corresponding to the abnormal situation by retrieving and querying in the first database. The establishment and configuration of the first database will be further described below.

[0091] Figure 7 It is an exemplary flowchart of configuring the first database shown according to some embodiments of this specification. In some embodiments, as Figure 7 shown, the process 700 can be executed before step 230 provided in the foregoing embodiments. In some embodiments, as Figure 7 shown, the process 700 may include the following steps.

[0092] Step 710: Obtain historical abnormal data of at least some devices. In some embodiments, the abnormal data of at least some devices on the photovoltaic module production line can be continuously obtained based on the methods provided in the foregoing embodiments, and the obtained abnormal data can be stored as historical abnormal data, which will not be elaborated here. In some embodiments, important process nodes on the photovoltaic module production line can be covered by means of visual inspection. For example, electroluminescence detection is performed on the photovoltaic module finished product or semi-finished product at the string welding machine blanking position, appearance inspection and electroluminescence detection are performed before the lamination process, appearance inspection is performed after the lamination process, appearance inspection is performed after the potting process, final inspection appearance inspection and electroluminescence detection, etc., to achieve the acquisition of historical abnormal data. In some embodiments, the obtained historical abnormal data can be stored in the first database.

[0093] Step 720: Train the historical abnormal data according to the machine learning algorithm to obtain the abnormal location and abnormal type corresponding to the historical abnormal data. In some embodiments, during the process of obtaining and storing the historical abnormal data, according to the record of the elimination of the abnormal situation corresponding to the historical abnormal data by the operator, the abnormal location and / or abnormal type corresponding to these historical abnormal data can be directly obtained according to the record content input by the operator. In some embodiments, the machine learning algorithm can also be used to train the historical abnormal data to improve the accuracy of identifying and classifying the abnormal situation, and obtain the abnormal location and abnormal type corresponding to the historical abnormal data.

[0094] Step 730: Obtain a preset naming rule. In some embodiments, the execution order between step 730 and the foregoing steps 710 and 720 is not limited. In some embodiments, the preset naming rule can be a set of standardized abnormal phenomenon name generation rules summarized by the operator according to the historical abnormal handling situation of the photovoltaic module production line. This standardized generation rule can generate a unique and descriptive abnormal phenomenon name according to the abnormal location and abnormal type.

[0095] Step 740: Generate an abnormal phenomenon name according to the preset naming rule, and the abnormal location and abnormal type corresponding to the historical abnormal data. In some embodiments, the abnormal phenomenon name can include or reflect the abnormal location and / or abnormal type corresponding to the abnormal situation of the photovoltaic module production line, and has uniqueness and readability, so that the operator can directly learn the specific content of the abnormal situation that occurs in the photovoltaic module production line through the abnormal phenomenon name.

[0096] Step 750: Store the abnormal phenomenon name, the corresponding abnormal location, and the abnormal type in the first database. In some embodiments, the first database can be configured to have at least one data query interface, and the operator or the abnormal phenomenon summary subsystem provided in the foregoing embodiments can access and query the first database according to the data query interface. In some embodiments, as Figure 6 shown, the abnormal phenomenon summary subsystem 600 can use the obtained abnormal location and / or abnormal type as the abnormal location and / or abnormal type through the abnormal name query module 620 to query the corresponding abnormal phenomenon name in the first database and provide it to the operator. The abnormal phenomenon name retrieved through the query of the first database can accurately and intuitively reflect the abnormal situation that occurs in the photovoltaic module production line to the operator.

[0097] Figure 8 is an exemplary structural diagram of an abnormal phenomenon classification subsystem according to some embodiments of this specification. In some embodiments, as Figure 8 shown, the abnormal phenomenon classification subsystem 800 can include a first interface module 810, an abnormal phenomenon analysis module 820, and an abnormal cause analysis module 830.

[0098] In some embodiments, the first interface module 810 can be used to receive the abnormal phenomenon name output by the abnormal phenomenon summary subsystem provided in the foregoing embodiments.

[0099] In some embodiments, the abnormal phenomenon analysis module 820 can parse the abnormal phenomenon name through natural language processing technology to obtain a parsing result. In some embodiments, the abnormal phenomenon name can include descriptive content of the abnormal location and abnormal type corresponding to the abnormal situation, and these descriptive contents of the abnormal situation can be parsed through natural language processing (NLP) technology, and then relevant data can be screened from the second database through methods such as keyword matching and semantic similarity calculation.

[0100] In some embodiments, the abnormal cause analysis module 830 may screen and analyze materials related to the parsing result from the second database according to the parsing result obtained by the abnormal phenomenon analysis module 820, and extract the associated causes and / or associated handling methods of the abnormal phenomenon names from the related materials. In some embodiments, the abnormal cause analysis module 830 may be implemented by an abnormal attribution model obtained through pre-training. The abnormal attribution model may adopt a machine learning algorithm and be constructed based on various data stored in the second database, including knowledge about various abnormal phenomena, associated factors, historical cases, and solutions. In some embodiments, the related materials may be input into the abnormal attribution model to obtain at least one associated cause of the abnormal phenomenon name and the possibility score for each associated cause; due to the high complexity of the photovoltaic module production line, one abnormal phenomenon name may correspond to one or more associated causes. The abnormal attribution model can obtain all the associated causes that may lead to the abnormal phenomenon name and obtain the possibility score corresponding to each associated cause. The higher the possibility score, the more likely the associated cause is the real cause of the abnormal situation corresponding to the abnormal phenomenon name. In some embodiments, the abnormal attribution model may also integrate a data warehouse and a data analysis tool to perform data mining on the abnormal resume, extract information such as abnormal features, occurrence frequencies, and handling processes, and analyze the historical occurrence of abnormal phenomena by combining local abnormal resume data to provide a more accurate reference for attribution analysis.

[0101] In some embodiments, during the process of determining the associated causes and / or associated handling methods of the abnormal phenomenon name, considering that one abnormal phenomenon name may correspond to multiple associated causes, in order to determine the associated causes corresponding to the abnormal phenomenon name, the associated causes and / or associated handling methods of the abnormal phenomenon name may be queried in the second database according to the abnormal phenomenon name and the monitoring data. The second database also stores the abnormal data conditions corresponding to each associated cause, and one or more accurate associated causes corresponding to the abnormal phenomenon name may be determined by comparing the monitoring data and the abnormal data conditions. In some embodiments, the monitoring data may include at least one or any combination of key parameters of at least some devices, device readings of at least some devices, and consumable life data of at least some devices. In some embodiments, the monitoring data of the key parameters of at least some devices may be implemented by the parameter monitoring module 520 provided in the foregoing embodiments; in some embodiments, the monitoring data of the device readings of at least some devices may be implemented by the sensing reading monitoring module 530 provided in the foregoing embodiments; in some embodiments, the consumable life data of at least some devices may be implemented by the consumable monitoring module 540 provided in the foregoing embodiments.

[0102] In some embodiments, when an abnormal situation occurs in the photovoltaic module production line, it is also necessary to generate corresponding troubleshooting work orders based on the obtained abnormal phenomenon names and the associated reasons corresponding to the abnormal phenomenon names, and push these work orders to the operators in a timely manner, so that the operators can respond to the abnormal situation in a timely manner. Figure 9 It is an exemplary structural diagram of a troubleshooting work order subsystem shown in some embodiments of this specification. In some embodiments, as Figure 9 shown, the troubleshooting work order subsystem 900 may include a second interface module 910, a work order generation module 920, a question-and-answer guidance module 930, a work order push module 940, and a tracking and feedback module 950.

[0103] In some embodiments, the second interface module 910 may be used to receive the associated reasons corresponding to the abnormal phenomenon names output by the abnormal phenomenon classification subsystem provided in the foregoing embodiments. In some embodiments, based on the relevant descriptions provided in the foregoing embodiments, an abnormal phenomenon name may correspond to multiple associated reasons. The second interface module 910 may form a list of possible reasons according to the multiple associated reasons corresponding to the abnormal phenomenon name. The list of possible reasons may include the possibility scores corresponding to each associated reason. The higher the possibility score, the more likely the associated reason is the real reason for the abnormal situation corresponding to the abnormal phenomenon name.

[0104] In some embodiments, the work order generation module 920 may automatically generate an abnormal situation troubleshooting work order according to the information provided by the second interface module 910 based on the abnormal situation attribution system. In some embodiments, the abnormal situation troubleshooting work order may specifically include a specific description of the abnormal situation that occurs in the photovoltaic module production line, including the abnormal phenomenon name, the abnormal location, the description of the abnormal phenomenon, etc.; a list of possible reasons, including the possibility scores corresponding to each associated reason, etc.; and a troubleshooting question-and-answer process for each associated reason. In some embodiments, the operator may be guided through the troubleshooting question-and-answer process to check the real-time operation status of the photovoltaic module production line to confirm whether the associated reason is the actual reason for the abnormality of the photovoltaic module production line.

[0105] In some embodiments, the question-and-answer guidance module 930 may be used to design multiple troubleshooting questions for each associated reason corresponding to the abnormal phenomenon name, and gradually guide the operator to troubleshoot and verify each associated reason corresponding to the abnormal phenomenon name in the photovoltaic module production line in the form of questions and answers. In some embodiments, if the abnormal phenomenon name corresponds to only one abnormal reason, there is no need for question-and-answer guidance troubleshooting. In some embodiments, if the abnormal phenomenon name corresponds to multiple abnormal reasons, a series of troubleshooting questions may be designed for each abnormal reason, and the employees may be gradually guided to troubleshoot in the form of questions and answers. The troubleshooting design in the form of questions and answers can effectively help the operator locate the actual abnormal reason.

[0106] In some embodiments, the work order push module 940 is configured to push the generated work order to designated operators through appropriate channels, such as emails, text messages, mobile application notifications, etc., so that the operators can receive in a timely manner the abnormal reasons that need to be investigated.

[0107] In some embodiments, the tracking and feedback module 950 is configured to continuously track the troubleshooting progress recorded by the operator when processing the work order and receive feedback information such as photo evidence and log records feedback by the operator, update the status of the troubleshooting work order in real time and confirm the abnormal reasons according to the feedback information, and can adjust the troubleshooting process when necessary. In some embodiments, the tracking and feedback module 950 can also obtain the troubleshooting process and materials used by the user according to the abnormal troubleshooting work order, and store the repair process and materials used in the second database to update the knowledge data of the second database.

[0108] Figure 10 is an exemplary flowchart of another method for detecting abnormalities in the production of photovoltaic modules according to some embodiments of the present specification. In some embodiments, as Figure 10 shown, the process 1000 may include the following steps.

[0109] Step 1010: Obtain the monitoring data of at least some of the devices on the photovoltaic module production line. In some embodiments, Figure 10 step 1010 in Figure 2 is similar to step 210 in

[0110] and will not be described in detail here. Figure 10 Step 1020: Obtain the abnormal data of at least some of the devices according to the monitoring data. In some embodiments, Figure 2 step 1020 in

[0111] is similar to step 220 in Figure 10 and will not be described in detail here. Figure 2

[0112] Step 1030: Query the name of the abnormal phenomenon corresponding to the abnormal data in the first database according to the abnormal data, where the first database stores at least the historical abnormal data of at least some of the devices and the corresponding names of the abnormal phenomena. In some embodiments, Figure 10 step 1030 in Figure 2 is similar to step 230 in

[0113] In some embodiments, such as Figure 10 the specific implementation of steps 1010 to 1040 shown can refer to the solutions provided in the foregoing embodiments and will not be elaborated herein.

[0114] Step 1050: Generate an exception troubleshooting work order based on the associated cause and / or associated handling method of the exception phenomenon name and push it to the user, where the exception troubleshooting work order includes the exception phenomenon name and the associated cause and / or associated handling method of the exception phenomenon name. In some embodiments, step 1050 can be implemented by, for example, Figure 9 the work order push module 940 in the troubleshooting work order subsystem 900 shown.

[0115] In some embodiments, such as Figure 10 the process 1000 shown may further include step 1060: Provide at least one troubleshooting guidance to the user in sequence according to the possibility scores of at least one associated cause through the exception troubleshooting work order. In some embodiments, the troubleshooting guidance can be presented in the form of questions and answers to guide the operator to cover each troubleshooting point corresponding to the associated cause in the photovoltaic module production line one by one for troubleshooting. In some embodiments, step 1060 can be implemented by, for example, Figure 9 the question-and-answer guidance module 930 in the troubleshooting work order subsystem 900 shown.

[0116] In some embodiments, such as Figure 10 the process 1000 shown may further include step 1070: Receive the feedback from the user on at least one troubleshooting guidance respectively, and determine the cause of the exception phenomenon name according to the feedback. In some embodiments, step 1070 can be implemented by, for example, Figure 9 the tracking feedback module 950 in the troubleshooting work order subsystem 900 shown.

[0117] In some embodiments, such as Figure 10 the process 1000 shown may further include step 1080: Obtain the troubleshooting process and materials used by the user according to the exception troubleshooting work order, and store the repair process and materials used in the second database. In some embodiments, step 1080 can be implemented by, for example, Figure 9 the tracking feedback module 950 in the troubleshooting work order subsystem 900 shown.

[0118] In some embodiments, when an abnormal situation occurs in the photovoltaic module production line, it is also necessary to generate a corresponding repair work order according to the obtained exception phenomenon name and the associated cause corresponding to the exception phenomenon name, and the operator can make timely and effective handling of the abnormal situation according to the guidance of the repair work order. Figure 11 is an exemplary structural diagram of a repair work order subsystem according to some embodiments of this specification. In some embodiments, such asFigure 11 The maintenance work order subsystem 1100 shown may include a third interface module 1110, a maintenance work order push module 1120, and a feedback record module 1130.

[0119] In some embodiments, the third interface module 1110 is configured to receive the cause of the abnormality corresponding to the name of the abnormal phenomenon output by the troubleshooting work order subsystem provided in the foregoing embodiments.

[0120] In some embodiments, the maintenance work order generation module 1120 is configured to retrieve the maintenance plan for the abnormal phenomenon name from the third database according to the cause of the abnormal phenomenon, and generate a maintenance work order based on the cause of the abnormality and the maintenance plan and push it to the user. In some embodiments, the third database may store knowledge information such as solutions to various equipment failures and maintenance operation guides, and the solution and specific maintenance steps can be determined according to the cause of the abnormal phenomenon corresponding to the name of the abnormal phenomenon.

[0121] In some embodiments, the feedback record module 1130 may be used to track the specific execution progress of the maintenance work order, guiding the maintenance personnel to check and repair problems one by one. In some embodiments, the feedback record module 1130 may also determine whether the maintenance work order is completed according to the monitoring data and / or monitoring status of the relevant abnormal equipment or processes on the photovoltaic module production line, so as to ensure that the completion status of the maintenance work by the operator is verified. In some embodiments, the feedback record module 1130 may also obtain the maintenance process and materials used by the user according to the maintenance work order, and store the maintenance process and materials used in the third database to update the solution for eliminating the cause of the abnormality.

[0122] The following will provide an exemplary description of the specific implementation of the photovoltaic module production abnormality detection method provided by the present disclosure in combination with each functional module subsystem provided in the foregoing embodiments. Figure 12 It is a flowchart example of the operation of another photovoltaic module production abnormality detection system shown in some embodiments of this specification. In some embodiments, as Figure 12 The photovoltaic module production abnormality detection system 1200 shown may include a digital twin subsystem 500, an abnormal phenomenon summary subsystem 600, an abnormal phenomenon classification subsystem 800, a troubleshooting work order subsystem 900, and a maintenance work order subsystem 1100. In some embodiments, as Figure 12As shown, the index monitoring module 560 in the digital twin subsystem 500 is used to output the abnormal type and abnormal location monitored when an abnormal situation occurs according to the monitoring situation. The status monitoring module 510 and the alarm monitoring module 550 in the digital twin subsystem 500 can cooperate to monitor the equipment status and process status in the photovoltaic module, and output the corresponding alarm points when an abnormality occurs. The abnormal phenomenon summary subsystem 600 can query the first database 1210 according to the received information such as the abnormal type, abnormal location, and alarm points to output the name of the abnormal phenomenon that occurs in the photovoltaic module production line. In some embodiments, as Figure 12 shown, when the final inspection monitoring module 570 in the digital twin subsystem 500 determines that the final product of the photovoltaic module production line has a downgraded situation based on the final inspection data, it can query the corresponding defective code and defective location, and then directly obtain the corresponding abnormal phenomenon name according to the defective code and defective location. In some embodiments, as Figure 12 shown, the abnormal phenomenon attribution subsystem 800 queries the second database 1220 according to the received abnormal phenomenon name to output the abnormal cause corresponding to the abnormal phenomenon name that occurs in the photovoltaic module production line; during the process of obtaining the abnormal cause, the abnormal phenomenon attribution subsystem 800 simultaneously receives the monitoring data of the photovoltaic module production line provided by the parameter monitoring module 520, the sensing reading monitoring module 530, and the consumable monitoring module 540 in the digital twin subsystem 500 to improve the accuracy of obtaining the abnormal cause. In some embodiments, as Figure 12 shown, the troubleshooting work order subsystem 900 and the maintenance work order subsystem 1100 query the third database 1230 according to the received abnormal cause to troubleshoot the abnormal cause, and provide solutions and specific maintenance operation guidelines for the determined abnormal cause; during the process of processing the work order, the troubleshooting work order subsystem 900 and the maintenance work order subsystem 1100 simultaneously receive the status monitoring data of the photovoltaic module production line provided by the status monitoring module 510 in the digital twin subsystem 500 to realize continuous tracking and monitoring of the troubleshooting and maintenance processes.

[0123] In some embodiments of this specification, a photovoltaic module production abnormal detection system is also provided. Figure 13 FIG. is a functional module structure diagram of a photovoltaic module production abnormal detection system according to some embodiments of this specification. In some embodiments, as Figure 13 shown, the photovoltaic module production abnormal detection system 1300 may include a first acquisition module 1310, a second acquisition module 1320, a first query module 1330, and a second query module 1340.

[0124] In some embodiments, the first acquisition module 1310 is used to acquire the monitoring data of at least some of the equipment on the photovoltaic module production line.

[0125] In some embodiments, the second acquisition module 1320 may be connected to the first acquisition module 1310 and is configured to acquire abnormal data of at least some devices according to the monitoring data.

[0126] In some embodiments, the first query module 1330 may be connected to the second acquisition module 1320 and is configured to query the name of the abnormal phenomenon corresponding to the abnormal data in the first database according to the abnormal data. Wherein, the first database stores at least historical abnormal data of at least some devices and the names of their corresponding abnormal phenomena.

[0127] In some embodiments, the second query module 1340 may be connected to the first query module 1330 and is configured to query the associated cause and / or associated processing method of the abnormal phenomenon name in the second database according to the abnormal phenomenon name. Wherein, the second database stores the associated factors / or associated processing methods of multiple abnormal phenomenon names.

[0128] For more content about each module, reference can be made to Figures 1 to 12 the relevant description, which will not be elaborated here. It should be understood that Figure 13 the system and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented through hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art can understand that the above methods and systems can be implemented using computer-executable instructions and / or control codes included in a processor. For example, such codes are provided in a carrier medium such as a magnetic disk, CD, or DVD-ROM, or in the memory of a programmable device. The system and its modules of this specification can not only be implemented by a hardware circuit of a programmable hardware device such as a very large scale integrated circuit or a gate array, a semiconductor such as a logic chip or a transistor, or a programmable logic device such as a field programmable gate array, but also be implemented by software executed by various types of processors, or be implemented by a combination of the above hardware circuit and software (for example, firmware).

[0129] It should be noted that the above description of the system and its modules is only for convenience of description and does not limit this specification within the scope of the examples given. It can be understood that for those skilled in the art, after understanding the principle of the system, they may, without departing from this principle, arbitrarily combine the various modules to form a subsystem connected to other modules. Or split some modules to obtain more modules or multiple units under that module. Such deformations are all within the scope disclosed in this specification.

[0130] In some embodiments provided in this specification, a computer program product is further provided, including computer instructions or a computer program. When at least a part of the computer instructions or the computer program is executed by a processor, the method provided in the foregoing embodiments of this specification can be implemented.

[0131] In some embodiments, the foregoing processor may be a combination of one or more of the following processors: central processing unit (CPU), application specific integrated circuit (ASIC), application specific instruction set processor (ASIP), graphics processing unit (GPU), physics processing unit (PPU), digital signal processor (DSP), field programmable gate array (FPGA), programmable logic device (PLD), programmable logic controller (PLC), reduced instruction set computer (RISC), microprocessor. The beneficial effects that the embodiments of this specification may bring include, but are not limited to: when an abnormal situation occurs in a photovoltaic module production line, the abnormal situation can be quickly matched with the historical maintenance event records stored in the database to obtain historical maintenance cases similar to or the same as the current abnormal situation, so as to quickly provide direct and accurate abnormal troubleshooting suggestions to the operators, which can effectively shorten the response time of abnormal events and improve the handling efficiency of abnormal events. It should be noted that the beneficial effects that different embodiments may produce are different. In different embodiments, the beneficial effects that may be produced may be any one or several combinations of the above, or any other beneficial effects that may be obtained.

[0132] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only an example and does not constitute a limitation to this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are taught in this specification, so such modifications, improvements, and corrections still fall within the spirit and scope of the exemplary embodiments of this specification.

Claims

1. A method for detecting abnormality in photovoltaic module production, characterized in that: include: Acquiring monitoring data of at least part of the equipment on the photovoltaic module production line; Acquire abnormal data of at least part of the equipment according to the monitoring data; According to the abnormal data, query the first database for the name of the abnormal phenomenon corresponding to the abnormal data, wherein the first database at least stores the historical abnormal data of at least part of the equipment and the corresponding abnormal phenomenon name; According to the abnormal phenomenon name, the associated cause and / or associated processing method of the abnormal phenomenon name is searched in the second database, and the second database stores the associated factors and / or associated processing methods of multiple abnormal phenomenon names.

2. The method according to claim 1, characterized in that: The acquiring the abnormal data of at least part of the equipment according to the monitoring data includes: Virtually replicating at least part of the equipment according to the simulation technology and the monitoring data of at least part of the equipment to generate a digital twin system of at least part of the equipment; At least part of the equipment is monitored by the digital twin system of at least part of the equipment, and when an abnormality is detected, abnormal data is obtained.

3. The method according to claim 2, characterized in that The digital twin system includes a data acquisition layer, a data processing layer, a twin model layer and an application layer; The data acquisition layer is used to obtain monitoring data of at least part of the equipment; The data processing layer is used to process and analyze the monitoring data to obtain data processing results; The twin model layer is used to construct a virtual model corresponding to at least part of the equipment in a one-to-one manner according to the data processing result; The application layer is used to monitor whether an abnormality occurs in the operation of the virtual model, and obtain the abnormal data when an abnormality is monitored.

4. The method according to claim 2 or 3, characterized in that: The monitoring of at least part of the equipment by means of the digital twin system of at least part of the equipment comprises at least one of the following: Monitoring data of at least part of the equipment in at least one state through the digital twin system of at least part of the equipment, wherein the at least one state includes at least one of waiting for material, material blockage, material change, failure, maintenance, and manual operation; Monitoring at least one key parameter of at least part of the equipment through a digital twin system of at least part of the equipment; Collecting and storing data of sensors of at least some of the devices through the digital twin system of at least some of the devices; Monitoring the usage and / or remaining life of consumables of at least part of the equipment through the digital twin system of at least part of the equipment; Collecting and storing alarm information of at least part of the equipment through the digital twin system of at least part of the equipment; Monitoring at least one indicator of the yield rate, rework rate, and first pass rate of at least part of the process by using the digital twin system of at least part of the equipment; The final inspection data of the final products of the photovoltaic module production line are monitored and stored through the digital twin system of at least part of the equipment.

5. The method according to claim 4, characterized in that The acquisition of abnormal data when an abnormality is detected includes at least one of the following: When it is monitored that data in the state of at least part of the devices meets an abnormal condition, determining that the data is abnormal data; Monitoring that the key parameter of at least part of the equipment deviates from a preset range, and determining that the key parameter is abnormal data; When monitoring data of the sensors of at least part of the devices to meet an abnormal condition, determining the data to be abnormal data; When a device alarm is detected, the alarm location and / or alarm content is determined to be abnormal data.

6. The method according to claim 4, characterized in that The method further includes: performing at least one of the following by means of the digital twin system of at least part of the device: Analyze the relationship between the key parameters of at least part of the equipment and production efficiency or production quality based on historical data and a machine learning algorithm to obtain analysis results, and generate parameter optimization suggestions based on the analysis results; When it is monitored that the consumables of at least part of the equipment have a replacement period shorter than a first preset time or are about to run out of material after a second preset time, a user is reminded to replace or replenish the material in a preset manner; At least one indicator of the yield rate, the rework rate, and the first-pass rate of at least part of the equipment is analyzed to generate quality influencing factors and / or quality improvement suggestions.

7. The method according to claim 1, characterized in that The monitoring data includes final inspection data of the final products of the photovoltaic module production line; The method further comprises: When the final inspection data indicates that the final product is degraded, querying the defect code and defect position corresponding to the degradation according to the historical final inspection database of the photovoltaic module production line; The abnormal phenomenon name is generated according to the defect code and the defect position.

8. The method according to claim 1, characterized in that The method further comprises searching a first database for a name of an abnormal phenomenon corresponding to the abnormal data according to the abnormal data, and: Acquiring historical abnormal data of at least part of the equipment; Training the historical abnormal data according to a machine learning algorithm to obtain an abnormal location and an abnormal type corresponding to the historical abnormal data; Get the preset naming rules; Generate a name for the abnormal phenomenon according to the preset naming rule, and the abnormal location and abnormal type corresponding to the historical abnormal data; The abnormal phenomenon name and the corresponding abnormal location and abnormal type are stored in the first database.

9. The method according to claim 1, characterized in that: The method of searching the second database for the associated cause and / or associated processing method of the abnormal phenomenon name according to the abnormal phenomenon name includes: Parsing the name of the abnormal phenomenon by natural language processing technology to obtain a parsing result; Filter relevant materials from the second database according to the analysis result; Extract the associated causes and / or associated processing methods of the abnormal phenomenon name from the relevant materials.

10. The method according to claim 9, characterized in that The method further comprises: constructing an abnormal attribution model based on the data in the second database using a machine learning algorithm; The extracting of the associated causes and / or associated processing methods of the abnormal phenomenon names from the relevant materials includes: The relevant materials are input into the abnormal attribution model to obtain at least one associated cause of the abnormal phenomenon name and a possibility score of each associated cause.

11. The method according to claim 1, characterized in that: The monitoring data includes at least one of the following data: key parameters of at least part of the equipment, equipment readings of at least part of the equipment, and life data of consumables of at least part of the equipment; The method of searching the second database for the associated cause and / or associated processing method of the abnormal phenomenon name according to the abnormal phenomenon name includes: According to the abnormal phenomenon name and the monitoring data, the associated cause and / or associated processing method of the abnormal phenomenon name is queried in the second database.

12. The method according to claim 1, 9 or 10, characterized in that The method further comprises: An abnormality troubleshooting work order is generated based on the associated causes and / or associated processing methods of the abnormal phenomenon name and pushed to the user. The abnormality troubleshooting work order includes the abnormal phenomenon name and the associated causes and / or associated processing methods of the abnormal phenomenon name.

13. The method according to claim 12, characterized in that In the abnormality troubleshooting work order, the at least one associated cause is sorted according to the likelihood score of the associated cause of the abnormal phenomenon name; The method further comprises: Providing at least one troubleshooting guide to the user in sequence according to the possibility score of the at least one associated cause through the abnormality troubleshooting worksheet; receiving feedback from users on the at least one troubleshooting guide; The abnormal cause of the abnormal phenomenon name is determined according to the feedback.

14. The method according to claim 13, characterized in that The method further comprises: Obtain the troubleshooting process and materials used by the user according to the abnormality troubleshooting work order; The maintenance process and the used materials are stored in the second database.

15. The method according to claim 13, characterized in that The method further comprises: According to the abnormal cause of the abnormal phenomenon name, a maintenance solution for the abnormal phenomenon name is retrieved from a third database; A maintenance work order is generated based on the abnormal cause and the maintenance plan and pushed to the user.

16. The method according to claim 15, characterized in that The method further comprises: Obtaining the maintenance process and materials used by the user according to the maintenance work order; The maintenance process and the used materials are stored in the third database.

17. A photovoltaic module production anomaly detection system, characterized in that: include: A first acquisition module is used to acquire monitoring data of at least part of the equipment on the photovoltaic module production line; A second acquisition module, configured to acquire abnormal data of at least part of the equipment according to the monitoring data; A first query module, configured to query a first database for a name of an abnormal phenomenon corresponding to the abnormal data according to the abnormal data, wherein the first database at least stores historical abnormal data of at least part of the equipment and the names of the abnormal phenomena corresponding to the abnormal data; The second query module is used to query the associated cause and / or associated processing method of the abnormal phenomenon name in the second database according to the abnormal phenomenon name, and the second database stores the associated factors and / or associated processing methods of multiple abnormal phenomenon names.

18. A computer program product, comprising computer instructions or a computer program, wherein when at least part of the computer instructions or the computer program is executed by a processor, the method according to any one of claims 1 to 16 can be implemented.

Images