A method for processing production data of a full life cycle of a bent cap

Abstract

The application provides a cap beam full life cycle production data processing method, which comprises the following steps: dividing the operation process of each production station of the cap beam to obtain a plurality of subdivided operation processes; collecting the personnel information and the station attribute information of each station in real time, and collecting the personnel information and the equipment operation information corresponding to each subdivided operation process; recording and storing the personnel information and the station attribute information of each station, and the personnel information and the equipment operation information corresponding to each subdivided operation process in a database of a cloud service platform; and the cloud service platform displays the data of the personnel information and the station attribute information of each station, and the personnel information and the equipment operation information corresponding to each subdivided operation process in real time, and simultaneously backs up the storage information in the database to a disaster recovery server. The cap beam full life cycle production data processing method can complete the information collection, information analysis and processing, and identification and evaluation of the whole cap beam production full life cycle, and finally realizes full informatization processing.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of intelligent processing of production data of bent cap, and particularly relates to a production data processing method for the whole life cycle of bent cap. BACKGROUND

[0002] Bent cap refers to a beam arranged on the top of a bent pile of a rowed frame for supporting, distributing and transmitting the load of an upper structure. The main function of the bent cap is to support the upper structure of the bridge and transmit all the load to the lower structure. At present, in the prior art, the production process of the bent cap is relatively complex, and the production quality is difficult to control. A key factor is that the current bent cap process is complicated, and the sub-operation steps between the nodes of each work station may affect the quality of the subsequent product. However, the current production process of each work station and work node does not adopt the whole life cycle data monitoring for the production of the bent cap.

[0003] Due to the long-term existence of the above-mentioned deficiencies in the traditional bent cap production technology, it is difficult to control the production quality of the bent cap, the production quality is difficult to trace, and the intelligent transformation of the modern bent cap production is seriously restricted. SUMMARY

[0004] In order to solve the above-mentioned technical problems in the prior art, the application provides a production data processing method for the whole life cycle of bent cap.

[0005] In order to achieve the above-mentioned purpose, the technical scheme of the application is as follows:

[0006] The application provides a production data processing method for the whole life cycle of bent cap, comprising the following operation steps:

[0007] The operation process of each production station of the bent cap is divided to obtain a plurality of sub-operation processes;

[0008] The personnel information and the station attribute information of each station are collected in real time, and the personnel information and the equipment operation information corresponding to each sub-operation process are also collected. Then, the personnel information and the station attribute information of each station and the personnel information and the equipment operation information corresponding to each sub-operation process are recorded and stored in the database of the cloud service platform.

[0009] The cloud service platform displays the data of the personnel information and the station attribute information of each station and the personnel information and the equipment operation information corresponding to each sub-operation process in real time, and simultaneously backs up the storage information in the above-mentioned database to the disaster recovery server.

[0010] Preferably, as an implementable solution; before the cloud service platform displays the real-time personnel information on the workstation, the workstation attribute information, and the personnel information corresponding to each sub-work process and the equipment operation information, the cloud service platform further includes the following operations:

[0011] Establish a production data quality detection model;

[0012] Detect the current real-time equipment operation information using the production data quality detection model to generate a corresponding model detection result;

[0013] When the current detection result is determined to be abnormal, perform reverse tracing on the production data.

[0014] Preferably, as an implementable solution; the establishment of the production data quality detection model specifically includes the following operations:

[0015] Obtain the equipment operation parameters of the single device in the abnormal running state in each sub-work process, and then establish a first equipment operation parameter abnormality database based on the data set composed of the equipment operation parameters of the single device in the abnormal running state. At the same time, in the case of two or more equipment abnormalities, first determine the target multiple sets of equipment associated in the multiple devices, and then continuously monitor the equipment operation parameters output by the multiple devices involved in the sub-work process based on the Apriori algorithm to obtain the associated target multiple sets of equipment operation parameters. A second equipment operation parameter abnormality database is established based on the data set of the target multiple sets of equipment operation parameters;

[0016] Using the equipment operation parameters of the single device in the abnormal running state under the same sub-work process as the training data for machine learning includes: determining whether the equipment operation parameters of the single device exceed a threshold value, and in the case where the threshold value is exceeded, determining that the equipment operation parameters of the single device are used as the first training data for machine learning. At the same time, in the case of two or more equipment abnormalities, using the target multiple sets of equipment operation parameters as the training data for machine learning includes: collecting the target multiple sets of equipment operation parameters to construct the second training data for machine learning;

[0017] Based on the first training data for machine learning, a first production data quality detection model is constructed, based on the second training data for machine learning, a second production data quality detection model is constructed, and the two are combined to form a production data quality detection model.

[0018] Preferably, as an implementable solution; the first determining of the target multiple sets of equipment associated in the multiple devices specifically includes the following operations:

[0019] In the initial state, the device running parameter association of two-by-two devices is performed in two sequentially executed subdivision job processes, and it is determined that the current two-by-two devices are associated devices if the variable of the single device running parameter of a single device in the two sequentially executed subdivision job processes affects the change rate of the single device running parameter of another single device in the unit time greater than a standard threshold value.

[0020] The associated devices of the current two-by-two devices are iteratively queried, and finally a plurality of groups of associated devices of two-by-two devices are obtained, and the plurality of groups of associated devices of two-by-two devices constitute the target multi-group devices;

[0021] By determining the target multi-group devices, a target multi-group device running parameter is constructed; the target multi-group device running parameter is a set of device running parameters formed by the target multi-group devices.

[0022] Preferably, as an implementable scheme; further comprising: in the case that the number of groups of devices in the target multi-group devices does not exceed a standard threshold value, saving the target multi-group device running parameter locally; if the number of groups of devices in the current target multi-group devices exceeds the standard threshold value, saving the target multi-group device running parameter on a disaster recovery server of a cloud service platform.

[0023] Preferably, as an implementable scheme; the production data quality detection model is used to detect the current real-time device running information to generate a corresponding model detection result, which specifically includes the following operations:

[0024] According to the device running information of the current device, the actual device running parameter of the current device is extracted, the similarity between the actual device running parameter of the current device and the data set composed of the single device running parameter in the first device running parameter anomaly database is calculated based on the first production data quality detection model in the production data quality detection model, and it is determined whether the monitored actual device running parameter is an abnormal device running parameter; if it is determined to be an abnormal device running parameter, the abnormal device running parameter is saved; the actual device running parameter is a single device running parameter;

[0025] The current device where the abnormality occurs is determined as a target reference device, and a target group of devices associated with the current device is determined based on the target reference device. Then, a set of actual device operation parameters of the target group of devices associated with the current device is called. The similarity between the set of actual device operation parameters of the target group of devices associated with the current device and a set of device operation parameters formed by the target group of devices associated with the current device stored in a second production data quality detection model in the production data quality detection model is calculated based on the second production data quality detection model. It is determined whether the actual device operation parameters of the target group of devices associated with the current device have an associated abnormality. If the similarity is greater than a standard similarity, it is determined that a device other than the current device in the target group of devices associated with the current device is a pseudo-abnormal device affected by association, rather than a real abnormal device.

[0026] Then, the pseudo-abnormal device is removed, and a real abnormal device is obtained as a final target abnormality identification device. The abnormal device operation parameters in the device operation information of the final target abnormality identification device and the operation state of the device are identified again, and the personnel information corresponding to the final target abnormality identification device is locked. The personnel information and the station attribute information corresponding to the current sub-process flow are determined.

[0027] Preferably, as an implementable scheme; when the current detection result is determined to be abnormal, the production data is traced in reverse, specifically including the following operation steps:

[0028] If the abnormal device operation parameter is determined, the abnormal device operation parameter is saved and traced; the abnormal start time and the abnormal end time of the abnormal device operation parameter are locked; if the abnormal start time is not terminated, the time after the preset time after the abnormal start time is taken as the abnormal end time;

[0029] All device operation parameter samples of all devices from the abnormal start time to the abnormal end time are obtained;

[0030] The obtained all device operation parameter samples are traced.

[0031] Preferably, as an implementable scheme; after determining that the current device has an abnormal device operation parameter, a complete evidence chain information collection operation is performed on the device at the current station;

[0032] The abnormal start time of the abnormal device operation parameter is locked as a preset time period, and the abnormal end time is locked as the time when the preset time period expires;

[0033] The personnel behavior image and the device operation image at the station from the abnormal start time to the abnormal end time are called;

[0034] Meanwhile, the corresponding maintenance personnel information, personnel behavior image, equipment operation image and equipment operation parameter of the current abnormal equipment are collected to form a life cycle abnormal work log.

[0035] Preferably, as an implementable solution, the life cycle abnormal work log of the current equipment at the current work station from the abnormal start time to the abnormal end time is further included, and the life cycle abnormal work log is retrieved and information is traced back; when the abnormal equipment operation state of the current equipment is determined, remote alarm control is realized for the current abnormal equipment.

[0036] Preferably, as an implementable solution, after the production data is traced back reversely when the current detection result is abnormal, the product image information of the final target abnormal identification equipment is acquired, a defect sample database is established, and the abnormal batch product is identified according to the defect sample database.

[0037] The product image information of the final target abnormal identification equipment is acquired, the defect sample database is established, and the abnormal batch product is identified according to the defect sample database, and the method specifically includes the following operations:

[0038] The current abnormal equipment operation parameter of the final target abnormal identification equipment is monitored in real time, if the abnormal equipment operation parameter of the final target abnormal identification equipment is found, the acquisition time period under the influence of the abnormal equipment operation parameter is locked, and then the defect product samples of the corresponding cap beam products in multiple time periods are acquired during the acquisition time period.

[0039] The acquisition time period is from the start time of the current abnormal equipment operation parameter of the final target abnormal identification equipment to the time when the current abnormal equipment operation parameter of the final target abnormal identification equipment is normal.

[0040] Meanwhile, in the cap beam production process, whether the equipment operation parameter of the current equipment is abnormal or not, the image feature recognition is performed on the current product sample and the defect product sample to determine whether the product defect accident occurs in the current product sample.

[0041] Compared with the prior art, the above technical solution provided by the embodiment of the present application has the following advantages: in the execution process of the above cap beam full life cycle production data processing method, firstly, the operation process of each production station of the cap beam is divided to obtain a plurality of subdivided operation processes; the personnel information and station attribute information of each station are collected in real time, and the personnel information and equipment operation information corresponding to each subdivided operation process are also collected, then the personnel information and station attribute information of each station and the personnel information and equipment operation information corresponding to each subdivided operation process are recorded and stored in the database of the cloud service platform; the cloud service platform performs data display on the personnel information and station attribute information of the station and the personnel information and equipment operation information corresponding to each subdivided operation process in real time, and simultaneously backs up the storage information in the above database to the disaster recovery server.

[0042] In the execution process of the above cap beam full life cycle production data processing method, the full life cycle production data of the cap beam can be comprehensively and systematically monitored, especially the subdivision of the multiple subdivided operation processes, the production monitoring data (the cap beam full operation process production data is acquired) is divided, and finally the production data quality detection model is established through systematic collection of production data, the production data is recognized and analyzed (especially single device anomaly and associated device anomaly), the corresponding model test result is generated by using the model, data recognition is performed, and reverse tracing is implemented; at the same time, the information database of the equipment running condition and the running state is established, and the data management and digital management level is realized. BRIEF DESCRIPTION OF DRAWINGS

[0043] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application, the illustrative embodiments of the present application and the description thereof serve to explain the present application, and do not constitute improper limitation on the present application. In the drawings:

[0044] Figure 1 is a main operation step flowchart of a cap beam full life cycle production data processing method of the embodiment of the present application;

[0045] Figure 2 is a specific implementation flowchart of step S30 execution process of the cap beam full life cycle production data processing method of the embodiment of the present application;

[0046] Figure 3 is another specific implementation flowchart of step S301 execution process of the cap beam full life cycle production data processing method of the embodiment of the present application;

[0047] Figure 4 is a specific implementation flowchart of the execution process of determining the associated target multiple groups of equipment in the plurality of equipment of the cap beam full life cycle production data processing method of the embodiment of the present application;

[0048] Figure 5 This is a schematic diagram of a specific implementation flow of step S302 of the method for processing production data of a cap beam throughout its life cycle according to an embodiment of the present application;

[0049] Figure 6 This is a schematic diagram of a specific implementation flow of step S303 of the method for processing production data of a cap beam throughout its life cycle according to an embodiment of the present application;

[0050] Figure 7 It is a schematic diagram of the implementation process of identifying abnormal batches of products based on a defect sample database in the method for processing production data of the entire life cycle of a cap beam in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] In the description of the present invention, it should be noted that certain words indicating orientation or positional relationships are only for the purpose of facilitating the description of the present invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they should not be understood as limitations on the present invention.

[0053] In the description of the present invention, it should be noted that the term "connection" should be understood in a broad sense. For example, it can mean a fixed connection, a detachable connection, or an integral connection; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; and it can mean internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention in specific circumstances.

[0054] The present invention will be further described in detail below through specific implementation examples in conjunction with the accompanying drawings.

[0055] Example 1

[0056] like Figure 1 As shown, the first embodiment of the present invention provides a method for processing production data of a cap beam throughout its life cycle, the method comprising:

[0057] The present invention provides a method for processing production data of a cap beam throughout its entire life cycle, comprising the following steps:

[0058] Step S10, dividing the process operation flow of each production station of the cap beam into multiple subdivided operation flows;

[0059] In step S20, personnel information and work station attribute information at each work station are collected in real time, and personnel information and equipment operation information corresponding to each sub-work process are also collected, and then the personnel information and work station attribute information at each work station and the personnel information and equipment operation information corresponding to each sub-work process are recorded and stored in a database at the cloud service platform;

[0060] In step S30, the cloud service platform displays data of personnel information and work station attribute information at each work station and personnel information and equipment operation information corresponding to each sub-work process in real time, and simultaneously backs up the stored information in the database to a disaster recovery server.

[0061] Referring to Figure 2 In the execution of step S30, before the cloud service platform displays data of personnel information and work station attribute information at each work station and personnel information and equipment operation information corresponding to each sub-work process in real time, the cloud service platform further includes the following operations:

[0062] In step S301, a production data quality detection model is established.

[0063] In step S302, the production data quality detection model is used to detect current real-time equipment operation information to generate corresponding model detection results.

[0064] In step S303, when it is determined that the current detection result is abnormal, the production data is reversely traced.

[0065] The above technical solution provided by the embodiments of the present application has the following advantages compared with the prior art: In step S10, the production process operation flow of each production work station of the bent cap is divided to obtain a plurality of sub-work processes; in step S20, personnel information and work station attribute information at each work station are collected in real time, and personnel information and equipment operation information corresponding to each sub-work process are also collected, and then the personnel information and work station attribute information at each work station and the personnel information and equipment operation information corresponding to each sub-work process are recorded and stored in a database at the cloud service platform; in step S30, the cloud service platform displays data of personnel information and work station attribute information at each work station and personnel information and equipment operation information corresponding to each sub-work process in real time, and simultaneously backs up the stored information in the database to a disaster recovery server.

[0066] In the execution process of the above-mentioned cap beam full life cycle production data processing method, the full life cycle production data of the cap beam can be comprehensively and systematically monitored, especially the subdivision of the multi-subdivision process flow, the production monitoring data is divided (the cap beam full process production data is acquired), and finally the production data quality detection model is established by systematically collecting the production data, the production data is recognized and analyzed (especially single device abnormality and associated device abnormality), the corresponding model test results are generated by using the model, data recognition is performed, and reverse tracing is implemented; at the same time, the information database of the equipment running condition and the running state is established, and the data management and digital management levels are realized.

[0067] Referring to Figure 3 In the execution process of step S301, the production data quality detection model is established, which specifically includes the following operations:

[0068] Step S3011, the equipment running parameters in the abnormal running state of each subdivision operation process are acquired, and then a first equipment running parameter abnormality database is established according to the data set composed of the equipment running parameters of the single device in the abnormal running state; meanwhile, in the case of two or more device abnormalities, the associated target multiple sets of devices are first determined in the multiple devices, and then the equipment running parameters output by the multiple devices involved in the subdivision operation process are continuously monitored based on the Apriori algorithm to obtain the associated target multiple sets of equipment running parameters; a second equipment running parameter abnormality database is established according to the data set of the target multiple sets of equipment running parameters;

[0069] Step S3012, using the equipment running parameters of the single device in the abnormal running state under the same subdivision operation process as the training data for machine learning includes: judging whether the equipment running parameters of the single device exceed a threshold value, and in the case where the threshold value is exceeded, determining that the equipment running parameters of the single device are used as the first training data for machine learning; meanwhile, in the case of two or more device abnormalities, using the target multiple sets of equipment running parameters as the training data for machine learning includes: collecting the target multiple sets of equipment running parameters to construct the second training data for machine learning;

[0070] Step S3013, a first production data quality detection model is constructed based on the first training data for machine learning, a second production data quality detection model is constructed based on the second training data for machine learning, and the production data quality detection model is formed by merging.

[0071] In the specific technical scheme of the embodiment of the present application, the production data quality detection model is composed of a first production data quality detection model and a second production data quality detection model. The first production data quality detection model is used to detect the abnormality of the equipment operation parameters caused by the abnormality of a single equipment, and the second production data quality detection model is used to analyze whether the abnormality of the equipment operation parameters caused by the abnormality of multiple single equipments exists, so as to significantly improve the practicability of the production data quality detection model, and further analyze the specific use scene to determine the cause of the equipment abnormality.

[0072] Referring to Figure 4 , the method comprises the following steps of:

[0073] In step S30121, in the initial state, the equipment operation parameters of two-by-two equipment in two sequentially executed subdivision job flows are associated, and it is determined that the current two-by-two equipment is associated equipment when the change rate of the single equipment operation parameter of a single equipment in a unit of time to the change rate of the single equipment operation parameter of another single equipment in a unit of time is greater than a standard threshold value.

[0074] In step S30122, the associated equipment of the current two-by-two equipment is iteratively queried, and finally the associated equipment of multiple sets of two-by-two equipment is obtained. The associated equipment of multiple sets of two-by-two equipment constitutes the associated target multiple sets of equipment.

[0075] In step S30123, the target multiple sets of equipment are determined, and the target multiple sets of equipment operation parameters are constructed. The target multiple sets of equipment operation parameters are a set of equipment operation parameters formed by the associated target multiple sets of equipment.

[0076] In the specific technical scheme of the embodiment of the present application, the above steps S30121-S30123 can be used to determine that the current two-by-two equipment is associated equipment when the change rate of the single equipment operation parameter of a single equipment in a unit of time to the change rate of the single equipment operation parameter of another single equipment in a unit of time is greater than a standard threshold value. In this way, the current two-by-two equipment can be determined to be associated equipment, and then the iteration process continues. Through multiple analysis iterations, multiple sets of two-by-two equipment can be determined. In this way, the associated equipment of multiple sets of two-by-two equipment constitutes the associated target multiple sets of equipment.

[0077] Through the above-mentioned associated target multi-group equipment target multi-group equipment operation parameters, a series of abnormal equipment operation parameters of multiple single equipments in a series of subdivided work processes caused by the influence of the equipment operation parameters of a single equipment can be obtained (so that it can be further judged whether the abnormality is of a single equipment or of a series of equipment caused by a single equipment), so that misjudgment of the equipment operation parameters of a single equipment is avoided.

[0078] Preferably, as an implementable solution, the cap beam full life cycle production data processing method provided by the application further comprises: saving the target multi-group equipment operation parameters locally when the number of groups of multiple equipments in the target multi-group equipment does not exceed a standard threshold; and saving the target multi-group equipment operation parameters on a disaster recovery server of a cloud service platform when the number of groups of multiple equipments in the current target multi-group equipment exceeds the standard threshold.

[0079] It should be noted that when the number of groups of multiple equipments in the current target multi-group equipment is not large, the target multi-group equipment operation parameters of the number of groups of multiple equipments in the current target multi-group equipment can be saved locally through the above-mentioned process; when the number of groups of multiple equipments in the current target multi-group equipment is large, the target multi-group equipment operation parameters can be saved on the disaster recovery server of the cloud service platform, which can avoid insufficient use of the use capacity of the local server and also ensure the safety of cloud storage (saved on the disaster recovery server of the cloud service platform).

[0080] Referring to Figure 5 In the process of step S302, the current real-time equipment operation information is detected by using the production data quality detection model to generate a corresponding model detection result, and the process specifically includes the following operations:

[0081] In step S3021, the actual equipment operation parameters of the current equipment are extracted according to the equipment operation information of the current equipment, the similarity of the actual equipment operation parameters of the current equipment and a data set composed of the single equipment operation parameters in the first equipment operation parameter abnormality database is calculated based on the first production data quality detection model in the production data quality detection model, and it is judged whether the monitored actual equipment operation parameters are abnormal equipment operation parameters; if it is judged that the equipment operation parameters are abnormal, the abnormal equipment operation parameters are saved; the actual equipment operation parameters are single equipment operation parameters.

[0082] Step S3022, determine the current device that occurs an abnormality as a target reference device, determine a target multi-group device associated with the current device based on the target reference device; then call an actual device running parameter set of the target multi-group device associated with the current device; calculate, based on a second production data quality detection model in the production data quality detection model, a similarity between the actual device running parameter set of the target multi-group device associated with the current device and a set of device running parameters formed by the target multi-group device associated with the second production data quality detection model stored, to determine whether an actual device running parameter of the target multi-group device associated with the current device has an associated abnormality, and if the similarity is greater than a standard similarity, identify and determine that a device other than the current device in the target multi-group device associated with the current device is a pseudo abnormal device affected by association, rather than a real abnormal device;

[0083] Step S3023, then eliminate the pseudo abnormal device, obtain a real abnormal device as a final target abnormality identification device, perform secondary identification on an abnormal device running parameter in device running information of the final target abnormality identification device and a running state of the device, and lock personnel information corresponding to the final target abnormality identification device, to determine personnel information and station attribute information corresponding to a current subdivided work flow.

[0084] It should be noted that the above technical solution can eliminate a pseudo abnormal device, lock and obtain a real abnormal device as a final target abnormality identification device, and perform secondary identification on an abnormal device running parameter in device running information of the final target abnormality identification device; then perform secondary identification on a running state of the current device, and lock personnel information corresponding to the final target abnormality identification device, to determine personnel information and station attribute information corresponding to a current subdivided work flow.

[0085] Locking personnel information corresponding to the final target abnormality identification device and personnel information and station attribute information corresponding to a current subdivided work flow can further determine a person in charge and a station construction condition, thereby providing effective data support for subsequent construction.

[0086] Referring to Figure 6 In step S303, when determining that a current detection result is abnormal, reverse trace production data, and the specific operation steps include the following steps:

[0087] Step S3031, if it is determined that an abnormal device running parameter, save and trace the abnormal device running parameter; lock an abnormal starting time and an abnormal ending time of the abnormal device running parameter; if the abnormal ending time has not ended, intercept a time after a preset time after the abnormal starting time as the abnormal ending time;

[0088] Step S3032, obtaining the device running parameter samples of all devices from the abnormal start time to the abnormal end time;

[0089] Step S3033, performing a traceback on the obtained device running parameter samples of all devices.

[0090] It should be noted that when the current detection result is determined to be abnormal, the production data is traced back in reverse, and specific reference is made to the operation steps S3031-S3033. In this way, the abnormal start time and the abnormal end time of the abnormal device running parameter are locked, and the production data is traced back in reverse (in specific implementation, the device running parameter samples of all devices from the abnormal start time to the abnormal end time are obtained; and then the obtained device running parameter samples of all devices are traced back).

[0091] Preferably, as an implementable scheme, after determining that the current device has an abnormal device running parameter, a complete evidence chain information collection operation is performed on the device at the current station;

[0092] The abnormal start time of the abnormal device running parameter is set as a preset time period, and the expiration time of the preset time period is set as the abnormal end time.

[0093] The personnel behavior image and the device running image at the station from the abnormal start time to the abnormal end time are called.

[0094] The corresponding maintenance personnel information of the current abnormal device, the personnel behavior image, the device running image, and the device running parameter form a life cycle abnormal work log.

[0095] It should be noted that in the specific technical scheme of the embodiment of the present application, a complete evidence chain information collection operation can be performed on the device at the current station, the personnel behavior image and the device running image at the station from the abnormal start time to the abnormal end time are called (so as to further analyze whether the personnel behavior image and the device running image are in violation, the personnel behavior image and the device running image at the station from the abnormal start time to the abnormal end time are called so as to be identified and analyzed by image recognition, and whether the personnel behavior image and the device running image at the station are abnormal is determined by image recognition means (for example, intelligent image analysis algorithm)); and the corresponding maintenance personnel information of the current abnormal device, the personnel behavior image, the device running image, and the device running parameter form a life cycle abnormal work log, so that the cloud service platform can subsequently call the life cycle abnormal work log and query the corresponding technical data.

[0096] Preferably, as an implementable solution, after the life cycle abnormal operation log of the current equipment at the current station from the abnormal starting moment to the abnormal ending moment is obtained, the life cycle abnormal operation log is further obtained and information tracing is performed, and remote alarm control is realized on the current abnormal equipment when the abnormal equipment running state is determined.

[0097] Preferably, as an implementable solution, after step S303, the product image information of the final target abnormal identification equipment is obtained, a defect sample database is established, and the abnormal batch product is identified according to the defect sample database.

[0098] Referring to Figure 7 , the product image information of the final target abnormal identification equipment is obtained, a defect sample database is established, and the abnormal batch product is identified according to the defect sample database, and the specific operations include the following operations.

[0099] Step S401, real-time monitoring of the current abnormal equipment running parameter of the final target abnormal identification equipment is performed, if the current abnormal equipment running parameter of the final target abnormal identification equipment is found, the acquisition time period under the influence of the abnormal equipment running parameter is quickly locked, and then the defect product samples of the corresponding cap beam product in multiple time periods are continuously obtained during the acquisition time period.

[0100] The acquisition time period is from the starting moment of the current abnormal equipment running parameter of the final target abnormal identification equipment to the moment when the current abnormal equipment running parameter of the final target abnormal identification equipment is normal.

[0101] Step S402, simultaneously in the cap beam production process, whether the equipment running parameter of the current equipment is abnormal or not, image feature recognition is performed on the current product sample and the defect product sample to determine whether the product defect accident occurs in the current product sample.

[0102] It should be noted that in the specific technical solution of the embodiment of the present application, after the above step S303 is executed, it can be identified that the equipment abnormal event has occurred, at this time, the product image information of the final target abnormal identification equipment can be continuously obtained, the defect sample database can be established, and the abnormal batch product can be identified according to the defect sample database.

[0103] In the specific implementation process, the current abnormal equipment operation parameters of the final target abnormal identification equipment are monitored in real time, if the current abnormal equipment operation parameters of the final target abnormal identification equipment are found, the acquisition time period under the influence of the abnormal equipment operation parameters is quickly locked, and then the defect product samples of the corresponding cap beam products in multiple time periods are continuously acquired during the acquisition time period; at the same time, during the cap beam production process, whether the current equipment operation parameter is abnormal or not, the current product sample and the defect product sample are subjected to image feature recognition to determine whether the current product sample has a product defect accident. In this way, whether the current product sample has a product defect accident can be identified through the above-mentioned manner.

[0104] In summary, the cap beam full life cycle production data processing method provided by the embodiments of the present application can acquire cap beam full process production data and trace it in reverse, can comprehensively and systematically monitor cap beam full life cycle production data, especially the subdivision of the multi-subdivision process flow, and finally establish a production data quality detection model by systematically collecting production data, identify and analyze the production data to generate corresponding model test results, and simultaneously trace in reverse, so as to comprehensively and informatively manage cap beam full life cycle production data.

[0105] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for processing production data of a cap beam throughout its entire life cycle, characterized in that: The method comprises: The process operation flow of each production station of the cap beam is divided into multiple subdivided operation flows; Collect the personnel information and workstation attribute information of each workstation in real time, as well as the personnel information and equipment operation information corresponding to each subdivided operation process. Then, store the personnel information and workstation attribute information of each workstation and the personnel information and equipment operation information corresponding to each subdivided operation process in the database of the cloud service platform. The cloud service platform displays data on personnel information and workstation attributes at workstations, as well as personnel information and equipment operation information corresponding to each detailed operation process in real time, and simultaneously backs up the stored information in the above database to the disaster recovery server; Before the cloud service platform displays data on personnel information at the workstations, workstation attribute information, and personnel information and equipment operation information corresponding to each subdivided operation process in real time, the cloud service platform further performs the following operations: Establish a production data quality detection model; Utilize the production data quality detection model to detect the current real-time equipment operation information and generate corresponding model detection results; When the current test result is judged to be abnormal, its production data is traced back; The production data quality detection model is used to detect the current real-time equipment operation information and generate corresponding model detection results, which specifically includes the following operations: Extracting actual equipment operating parameters of the current equipment based on the equipment operating information of the current equipment, calculating similarity between the actual equipment operating parameters of the current equipment and a data set consisting of equipment operating parameters of a single equipment in the first equipment operating parameter abnormality database based on the first production data quality detection model in the production data quality detection model, and determining whether the monitored actual equipment operating parameters are abnormal equipment operating parameters; if determined to be abnormal equipment operating parameters, saving the abnormal equipment operating parameters; the actual equipment operating parameters are equipment operating parameters of a single equipment; Determine the current device in which the abnormality occurs as the target reference device, and determine the target multiple groups of devices associated with the current device based on the target reference device; then retrieve the actual device operating parameter set of the target multiple groups of devices associated with the current device; calculate, based on the second production data quality detection model in the production data quality detection model, the similarity between the actual device operating parameter set of the target multiple groups of devices associated with the current device and the set of device operating parameters formed by the target multiple groups of devices associated with the second production data quality detection model, and determine whether the actual device operating parameters of the target multiple groups of devices associated with the current device have an associated abnormality; if the similarity is greater than the standard similarity, identify and determine that the devices other than the current device in the target multiple groups of devices associated with the current device are pseudo-abnormal devices affected by the association, rather than real abnormal devices; Then, the pseudo-abnormal devices are eliminated, and the real abnormal devices are obtained as the final target abnormal identification devices. The abnormal device operating parameters and the operating status of the device in the device operating information of the final target abnormal identification device are re-identified, and the personnel information corresponding to the final target abnormal identification device is locked, and the personnel information and workstation attribute information corresponding to the current subdivided operation process are determined; When the current test result is judged to be abnormal, the production data is traced back, including the following steps: If it is determined to be an abnormal device operating parameter, the abnormal device operating parameter is saved and traced back; the abnormal start time of the abnormal device operating parameter and the abnormal end time of the abnormality are locked; if it is not terminated, the time after the preset time after the abnormal start time is intercepted as the abnormal end time; Acquiring equipment operating parameter samples of all equipment from the start time to the end time of the abnormality; The equipment operating parameter samples of all equipment obtained are traced back.

2. The method for processing production data of a cap beam throughout its entire life cycle according to claim 1, characterized in that: The establishment of the production data quality detection model specifically includes the following operations: Obtaining the equipment operating parameters under abnormal operating conditions in each subdivided operation process, and then establishing a first equipment operating parameter abnormality database based on a data set consisting of the equipment operating parameters of a single device under abnormal operating conditions; simultaneously, in the event of two or more equipment abnormalities, first determining associated target groups of equipment from multiple devices, and then continuously monitoring the equipment operating parameters output by the multiple devices involved in the subdivided operation process based on the Apriori algorithm to obtain associated target groups of equipment operating parameters; Establishing a second equipment operating parameter anomaly database based on the target multiple sets of equipment operating parameter data sets; Using the equipment operating parameters of a single device in an abnormal operating state within the same subdivided operation process as training data for machine learning includes: determining whether the equipment operating parameters of the single device exceed a threshold value, and if the threshold value is exceeded, determining to use the equipment operating parameters of the single device as first training data for machine learning; and simultaneously using target multiple sets of equipment operating parameters as training data for machine learning when two or more devices are abnormal includes: collecting the target multiple sets of equipment operating parameters to construct second training data for machine learning; A first production data quality detection model is constructed based on the first training data of machine learning, and a second production data quality detection model is constructed based on the second training data of machine learning, and the two are combined to form a production data quality detection model.

3. The method for processing production data of a cap beam throughout its entire life cycle according to claim 2, characterized in that: The step of first determining the associated target groups of devices from the multiple devices specifically includes the following operations: In the initial state, the device operating parameters of each pair of devices are associated in two sequentially executed sub-operation processes. If the rate of change per unit time of the variable of the single device operating parameter of a single device affecting the single device operating parameter of another single device in the two sequentially executed sub-operation processes is greater than a standard threshold, then the two devices are considered to be associated devices. Continuously query the associated devices of the current two-by-two devices iteratively, and finally obtain multiple groups of associated devices of the two-by-two devices. The multiple groups of associated devices of the two-by-two devices constitute the associated target multiple groups of devices; By determining the associated target multiple groups of devices, target multiple groups of device operating parameters are constructed; the target multiple groups of device operating parameters are a collection of device operating parameters formed by the associated target multiple groups of devices.

4. The method for processing production data of a cap beam throughout its entire life cycle according to claim 3, characterized in that: Also includes: When the number of the multiple groups of devices in the target multiple groups of devices does not exceed a standard threshold, the operating parameters of the target multiple groups of devices are stored locally; If the number of multiple groups of devices in the current target multiple groups of devices exceeds a standard threshold, the operating parameters of the target multiple groups of devices are saved on a disaster recovery server of the cloud service platform.

5. The method for processing production data of a cap beam throughout its entire life cycle according to claim 4, characterized in that: After determining that there are abnormal equipment operating parameters on the equipment, it also includes implementing a complete evidence chain information collection operation for the equipment at the current workstation; The occurrence of abnormal equipment operating parameters is used as the abnormal start time of the preset time period, and the expiration time of the preset time period is used as the abnormal end time; Retrieve images of personnel behavior and equipment operation at the workstation from the time the abnormality starts to the time the abnormality ends; At the same time, the corresponding maintenance personnel information of the current abnormal equipment, personnel behavior images, equipment operation images and equipment operation parameters are collected to form a life cycle abnormality work log.

6. The method for processing production data of a cap beam throughout its entire life cycle according to claim 5, characterized in that: After retrieving the personnel behavior images and equipment operation images at the workstation from the start time to the end time of the abnormality; and collecting the corresponding maintenance personnel information of the current abnormal equipment, the personnel behavior images, equipment operation images and equipment operation parameters form a life cycle abnormality work log, it also includes retrieving the life cycle abnormality work log and tracing the information; when it is determined that the current equipment has an abnormal equipment operation status, remote alarm control is implemented for the current abnormal equipment.

7. The method for processing production data of a cap beam throughout its entire life cycle according to claim 6, characterized in that: After reverse tracing its production data when the current test result is determined to be abnormal, the method further includes: obtaining product image information of the final target abnormality identification device, establishing a defect sample database, and identifying abnormal batch products based on the defect sample database; The process of obtaining product image information of the final target abnormality identification device, establishing a defect sample database, and identifying abnormal batch products based on the defect sample database specifically includes the following operations: Monitor the current abnormal equipment operating parameters of the final target abnormality identification equipment in real time. If abnormal equipment operating parameters of the final target abnormality identification equipment are currently found, quickly lock the collection time period under the influence of the abnormal equipment operating parameters, and then continuously obtain defective product samples of the corresponding cap beam products in multiple time periods during the collection time period; The collection time period is from the start time of the abnormal device operating parameters of the final target abnormality identification device to the time when the abnormal device operating parameters of the final target abnormality identification device are monitored to be normal; At the same time, during the cap beam production process, regardless of whether the equipment operating parameters of the current equipment are abnormal, image feature recognition is performed on the current product sample and the defective product sample to determine whether a product defect accident has occurred in the current product sample.

Images