Assembly optimization method and system based on industrial Internet of Things, and medium

Abstract

The invention provides an assembly optimization method and system based on an industrial internet of things, and a medium. The method comprises the steps of obtaining and uploading link assembly data of a production line; obtaining an initial assembly scheme; determining an optimized assembly scheme based on the link assembly data and the initial assembly scheme; determining a target assembly scheme based on the optimized assembly scheme; based on the target assembly scheme, an assembly regulation and control instruction is generated and stored; in response to the arrival of the adjustment time, issuing an assembly regulation and control instruction, and regulating and controlling the equipment operation parameters of the production equipment and the conveying belt parameters of the conveying belt; obtaining reference assembly data of the production line after the assembly regulation and control instruction is executed; determining a correction optimization link of the current assembly scheme in response to the condition that the reference assembly data meets a correction condition; based on the correction optimization link, generating a corrected assembly scheme; the corrected assembly scheme is stored, and a correction regulation and control instruction is generated; and a correction regulation and control instruction is issued, and the equipment operation parameters and the conveying belt parameters of the production equipment are corrected.

Description

Technical Field

[0001] This specification relates to the field of industrial Internet of Things assembly, and in particular to an assembly optimization method, system and medium based on the industrial Internet of Things. Background Art

[0002] With the advancement of industrial intelligence, intelligent manufacturing has become an important direction for the digital transformation of the manufacturing industry. As one of the key technologies of intelligent manufacturing, the Industrial Internet of Things (IIoT) integrates multiple technologies such as sensor technology, network communication, and big data analysis into all aspects of industrial production. The Industrial Internet of Things can collect a large amount of data in the industrial production process in real time, conduct data interaction between equipment and equipment, and between equipment and personnel, and help production managers to perform intelligent management.

[0003] Product assembly and manufacturing is an important part of product quality assurance. In the traditional manufacturing process, data collection mainly relies on manual collection or simple equipment monitoring, which often leads to problems such as untimely and inaccurate data acquisition, lack of effective synchronization mechanism between the assembly site and the information system, and lack of data analysis and application methods. As a result, the quality data in the assembly process is often scattered in various links and departments and cannot be transmitted to the information system in time. The fragmentation and unsystematic nature of data make the analysis and improvement of quality problems difficult. The Industrial Internet of Things can monitor the production process and equipment status in real time. Through the analysis and processing of the production process and equipment data, it can diagnose equipment failures and implement intelligent maintenance. At the same time, it can also discover and solve potential production problems, thereby significantly improving production efficiency and product quality.

[0004] Therefore, it is hoped to propose an assembly optimization method based on industrial Internet of Things to improve the level of intelligent management in the manufacturing industry. Summary of the invention

[0005] The invention content includes an assembly optimization method based on the industrial Internet of Things. The method includes: obtaining the link assembly data of the production line through the industrial Internet of Things perception control platform, and uploading the link assembly data to the industrial Internet of Things management platform via the industrial Internet of Things sensor network platform; obtaining the initial assembly plan from the plan database, and the plan database is stored in the data center on the industrial Internet of Things management platform; determining the optimized assembly plan based on the link assembly data and the initial assembly plan; determining the target assembly plan based on the optimized assembly plan; generating assembly control instructions based on the target assembly plan; determining the target assembly plan as the current assembly plan and storing it in the data center; in response to the arrival adjustment time, issuing the assembly control instructions to the industrial An Internet of Things perception control platform is used to control the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt; based on the industrial Internet of Things perception control platform, the reference assembly data of the production line after the assembly control instruction is executed is obtained; in response to the reference assembly data satisfying the correction condition: determining the correction optimization link of the current assembly plan; based on the correction optimization link, correcting the current assembly plan to generate a corrected assembly plan; storing the corrected assembly plan in the data center, and generating a correction control instruction based on the corrected assembly plan; issuing the correction control instruction to the industrial Internet of Things perception control platform to correct the equipment operating parameters of the production equipment and the conveyor belt parameters.

[0006] The invention includes an assembly optimization system based on the Industrial Internet of Things. The system includes: an Industrial Internet of Things user platform, an Industrial Internet of Things service platform, an Industrial Internet of Things management platform, an Industrial Internet of Things sensor network platform and an Industrial Internet of Things perception control platform. The Industrial Internet of Things management platform is communicatively connected with the Industrial Internet of Things sensor network platform and the Industrial Internet of Things service platform; the Industrial Internet of Things perception control platform includes production equipment of a production line and a data acquisition device deployed on the production line, and the Industrial Internet of Things perception control platform realizes data interaction with the Industrial Internet of Things management platform through the Industrial Internet of Things sensor network platform.

[0007] The invention includes a computer-readable storage medium storing computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the assembly optimization method.

[0008] The present invention improves the adaptability and flexibility of the system by dynamically adjusting the assembly plan, timely discovers and corrects problems in actual production, and reduces assembly errors and failures; by continuously optimizing the assembly plan, the overall performance and efficiency of the production line can be effectively improved to ensure the feasibility and reliability of the optimized assembly plan and reduce production risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 It is a schematic diagram of the platform structure of the industrial Internet of Things according to some embodiments of this specification; Figure 2 is an exemplary flow chart of an assembly optimization method based on the industrial Internet of Things according to some embodiments of this specification; Figure 3 is an exemplary flow chart for determining an optimized assembly solution according to some embodiments of this specification; Figure 4 is an exemplary schematic diagram of a parameter generation model according to some embodiments of this specification; Figure 5 is an exemplary flow chart for determining a target assembly solution according to some embodiments of the present specification. DETAILED DESCRIPTION

[0010] Combined with the industrial Internet of Things technology, it is possible to formulate reasonable industrial assembly plans and optimize industrial production processes. Therefore, it is hoped that an assembly optimization method based on the industrial Internet of Things can be proposed, which can optimize and simulate the assembly plan based on the data collected by the industrial Internet of Things and intelligent tools to ensure the feasibility of the plan, thereby further improving the level of intelligent management in the manufacturing industry.

[0011] Figure 1 It is a schematic diagram of the platform structure of the industrial Internet of Things shown in some embodiments of this specification.

[0012] In some embodiments, Figure 1 As shown, the industrial Internet of Things 100 includes an industrial Internet of Things user platform 110, an industrial Internet of Things service platform 120, an industrial Internet of Things management platform 130, an industrial Internet of Things sensor network platform 140 and an industrial Internet of Things perception control platform 150.

[0013] The IIoT user platform 110 refers to a platform for interacting with IIoT users. In some embodiments, the IIoT user platform 110 may be configured as a terminal device.

[0014] In some embodiments, the IIoT user platform 110 may perform data exchange with the IIoT service platform 120 .

[0015] The IIoT service platform 120 is a platform for data interaction with the IIoT user platform 110 and / or the IIoT management platform 130 .

[0016] In some embodiments, the industrial Internet of Things service platform 120 is communicatively connected with the industrial Internet of Things management platform 130 .

[0017] The IIoT management platform 130 is a platform for overall management of the entire IIoT. In some embodiments, the IIoT management platform 130 is configured with a processor. The processor can interact with one or more platforms in the IIoT 100 (e.g., the IIoT service platform 120, the IIoT sensor network platform 140, etc.), perform data processing, and generate and / or control the execution of one or more program instructions in this specification.

[0018] In some embodiments, the IIoT management platform 130 can achieve data interaction with the IIoT perception control platform 150 through the IIoT sensor network platform 140. For example, the IIoT management platform 130 can generate assembly control instructions and correction control instructions, and send the assembly control instructions and correction control instructions to the IIoT perception control platform 150 through the IIoT sensor network platform 140. For a description of assembly control instructions and correction control instructions, see Figure 2 and related descriptions.

[0019] In some embodiments, the industrial Internet of Things management platform 130 is communicatively connected with the industrial Internet of Things sensor network platform 140 and the industrial Internet of Things service platform 120 .

[0020] In some embodiments, the industrial Internet of Things management platform 130 can determine an optimized assembly plan based on the link assembly data and the initial assembly plan; determine a target assembly plan based on the optimized assembly plan; generate assembly control instructions based on the target assembly plan; generate a revised assembly plan based on the revised optimized link; and generate a revised control instruction based on the revised assembly plan. For more information, see Figure 2 and related descriptions.

[0021] In some embodiments, the industrial Internet of Things management platform 130 includes a data center 131 .

[0022] The data center 131 is used to store, manage, and transmit large amounts of data.

[0023] In some embodiments, the data center 131 may store a solution database and various solutions. For example, the initial assembly solution, the revised assembly solution, and the target assembly solution obtained during the assembly optimization method. For a description of the solution database, the initial assembly solution, the revised assembly solution, and the target assembly solution, see Figure 2 and related descriptions.

[0024] The industrial Internet of Things sensor network platform 140 refers to a connection platform for realizing data interaction between the industrial Internet of Things management platform 130 and the industrial Internet of Things perception control platform 150.

[0025] In some embodiments, the IIoT sensor network platform 140 is communicatively connected to the IIoT management platform 130 .

[0026] In some embodiments, the industrial Internet of Things sensor network platform 140 can upload the link assembly data obtained by the industrial Internet of Things perception control platform 150 to the industrial Internet of Things management platform 130.

[0027] The industrial Internet of Things perception control platform 150 is a platform for generating perception information and executing control information.

[0028] In some embodiments, the industrial Internet of Things perception control platform 150 can obtain assembly data of the production line links and adjust the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt.

[0029] In some embodiments, the IIoT perception control platform 150 can achieve data interaction with the IIoT management platform 130 through the IIoT sensor network platform 140. For example, the IIoT perception control platform 150 can obtain the assembly data of the production line in real time, and upload the assembly data to the IIoT management platform 130 via the IIoT sensor network platform 140.

[0030] In some embodiments, the industrial Internet of Things perception control platform 150 may include production equipment 151 and a data acquisition device 152 .

[0031] The production equipment 151 is equipment used for production in each production link of the production line, for example, assembly equipment, processing equipment, testing equipment, etc.

[0032] The data collection device 152 is a device for collecting assembly data of links on the production line.

[0033] In some embodiments, the data acquisition device may be at least one of a sensor, an RFID technology device, a barcode scanner, a camera, and the like.

[0034] In some embodiments, the data collection device can collect a variety of data, such as temperature data, humidity data, voltage and current data, pressure data, radio frequency signals, location and status of items, environmental data, audio data, video data, etc.

[0035] In some embodiments of the present specification, through the various data collected by the data acquisition device, production management personnel can not only monitor the environmental status and equipment operating status in real time, discover potential faults and risks in time, and prevent the occurrence of safety accidents in production; they can also control various parameters of production equipment (such as operating speed, current, voltage, temperature, etc.) to ensure product quality while effectively allocating resources.

[0036] Figure 2 It is an exemplary flow chart of an assembly optimization method based on industrial Internet of Things according to some embodiments of this specification.

[0037] In some embodiments, the assembly optimization method based on the industrial Internet of Things includes: obtaining the link assembly data of the production line through the industrial Internet of Things perception control platform, and uploading the link assembly data to the industrial Internet of Things management platform via the industrial Internet of Things sensor network platform; obtaining the initial assembly plan from the plan database, and the plan database is stored in the data center on the industrial Internet of Things management platform; determining the optimized assembly plan based on the link assembly data and the initial assembly plan; determining the target assembly plan based on the optimized assembly plan; generating assembly control instructions based on the target assembly plan; determining the target assembly plan as the current assembly plan and storing it in the data center; and in response to the arrival adjustment time, The assembly control instructions are sent to the industrial Internet of Things perception control platform to control the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt; based on the industrial Internet of Things perception control platform, the reference assembly data of the production line after the execution of the assembly control instructions is obtained; in response to the reference assembly data satisfying the correction conditions: determining the correction optimization link of the current assembly plan; based on the correction optimization link, correcting the current assembly plan to generate a corrected assembly plan; storing the corrected assembly plan in the data center, and generating a corrected control instruction based on the corrected assembly plan; sending the corrected control instruction to the industrial Internet of Things perception control platform to correct the equipment operating parameters of the production equipment and the conveyor belt parameters.

[0038] like Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the process 200 can be executed by an industrial Internet of Things management platform.

[0039] Step 210, obtaining the assembly data of the production line links through the industrial Internet of Things perception control platform, and uploading the assembly data of the links to the industrial Internet of Things management platform via the industrial Internet of Things sensor network platform.

[0040] For more information about the Industrial Internet of Things perception control platform, Industrial Internet of Things sensor network platform and Industrial Internet of Things management platform, please refer to Figure 1 The corresponding description.

[0041] A production line refers to the collection of all assembly units required to assemble a product. In some embodiments, a production line includes multiple workstations. A workstation refers to the equipment configuration for assembling components or assemblies in each link of a production line.

[0042] Link assembly data refers to the data related to assembly in each link of the production line. For example, link assembly data may include the model, quantity, location, assembly time, manual / machine assembly action sequence, characteristic data of stage assembly components, and temporary storage quantity of temporary storage locations, etc.

[0043] Manual / machine assembly action sequence refers to the sequence of assembly operations when assembling by manual or machine.

[0044] The stage assembly components refer to the components assembled in each link of the production process. The feature data of the stage assembly components refer to the data related to the components assembled in each link of the production process. For example, the image data, weight data, quality inspection data, etc. of the stage assembly components.

[0045] In some embodiments, the link assembly data can be obtained through the data acquisition device of the industrial Internet of Things perception control platform. For more information about the data acquisition device, see Figure 1 The corresponding description.

[0046] Step 220, obtaining an initial assembly plan from a plan database.

[0047] In some embodiments, the solution database is stored in a data center on the industrial Internet of Things management platform.

[0048] For more information about data centers, see Figure 1 The corresponding description.

[0049] The scheme database refers to a database for storing assembly schemes. An assembly scheme refers to a scheme for assembling parts or components into a complete product according to a certain process and method. In some embodiments, the assembly scheme may include assembly auxiliary parameters, assembly sequence, assembly time sequence, assembly method sequence, assembly component object sequence, component list sequence, etc.

[0050] Assembly auxiliary parameters refer to the relevant parameters that assist the assembly process. For example, assembly auxiliary parameters may include but are not limited to conveyor belt speed, robot arm movement speed, torque setting, etc. Torque setting refers to the relevant settings of torque force and lever arm.

[0051] The assembly sequence refers to the order in which parts or components are assembled.

[0052] The assembly time series refers to the sequence of time required to assemble the components of each link.

[0053] The assembly method sequence refers to the sequence of methods for assembling components in each link. For example, the assembly method sequence may include manual or automatic tools, assembly techniques, etc. used in each link of the assembly.

[0054] The assembly component object sequence refers to the sequence of objects that need to be assembled at each workstation.

[0055] The parts list sequence refers to a detailed list of all parts that need to be assembled in each link. For example, the parts list sequence can include the model, specification, quantity, etc. of the assembled parts.

[0056] The initial assembly plan can be preset by the staff based on experience.

[0057] Step 230, determining an optimized assembly plan based on the link assembly data and the initial assembly plan.

[0058] The optimized assembly plan refers to the assembly plan obtained by optimizing the initial assembly plan.

[0059] In some embodiments, the industrial Internet of Things management platform can determine the optimized assembly plan in a variety of ways based on the link assembly data and the initial assembly plan. For example, the industrial Internet of Things management platform can adjust one or more of the assembly auxiliary parameters, assembly sequence, assembly component object sequence, assembly time, etc. of the initial assembly plan based on the link assembly data to obtain multiple optimized assembly plans.

[0060] The adjustment of assembly time can be achieved by adjusting the working parameters of each workstation. The working parameters of each workstation may include the transmission speed of the conveyor belt, the assembly speed of each workstation, the combination of components, etc. The combination of components refers to the combination of components assembled at the same workstation. For example, the initial assembly plan of a workstation is set to assemble component A and component B, which can be changed to assemble component B and component C.

[0061] As just an example, the industrial Internet of Things management platform can adjust one or more of the assembly sequence of the initial assembly plan, the working parameters of each workstation, etc. based on the link assembly data to obtain multiple optimized assembly plans.

[0062] In some embodiments, during the adjustment process, the industrial Internet of Things management platform can randomly adjust one or more of the assembly links to be adjusted to obtain multiple optimized assembly solutions.

[0063] The assembly link to be adjusted refers to the link whose time difference ratio is greater than the preset ratio. The time difference ratio refers to the ratio of the difference between the actual assembly time and the preset assembly time to the preset assembly time.

[0064] The adjustment includes increasing the assembly time of the assembly link to be adjusted. In some embodiments, the amplitude of adjusting the assembly time of the assembly link to be adjusted is positively correlated with the time difference ratio.

[0065] For other embodiments of determining an optimized assembly plan based on link assembly data and an initial assembly plan, see Figure 3 Related instructions.

[0066] Step 240: determine a target assembly solution based on the optimized assembly solution.

[0067] The target assembly scheme refers to the assembly scheme selected based on the optimized assembly scheme.

[0068] In some embodiments, the industrial Internet of Things management platform can determine the target assembly scheme in a variety of ways based on the optimized assembly scheme. For example, the industrial Internet of Things management platform can select the optimized assembly scheme whose adjusted total assembly time is lower than the adjustment threshold and has the least number of adjusted assembly links as the target assembly scheme based on multiple optimized assembly schemes. The adjusted total assembly time refers to the sum of the individual assembly times in the assembly time sequence of the optimized assembly scheme.

[0069] The adjustment threshold refers to a preset minimum value of the total assembly time after adjustment. The adjustment threshold can be preset by the staff based on experience. In some embodiments, the adjustment threshold is lower than the time threshold. The time threshold refers to a preset maximum value of the total assembly time. The time threshold can be preset by the staff based on experience.

[0070] In some embodiments, selecting an optimized assembly solution with the least number of assembly links to be adjusted can reduce interference with the production line and reduce the complexity of the adjustment.

[0071] For other embodiments of determining a target assembly solution based on an optimized assembly solution, see Figure 5 Related instructions.

[0072] Step 250: Generate assembly control instructions based on the target assembly plan.

[0073] Assembly control instructions refer to instructions for controlling related equipment in the assembly process.

[0074] In some embodiments, the IIoT management platform can generate assembly control instructions in a variety of ways based on the target assembly scheme. For example, the IIoT management platform can modify the last used assembly control instructions based on the assembly auxiliary parameters in the target assembly scheme, such as increasing the conveyor belt speed, to generate the current assembly control instructions.

[0075] Step 260: determine the target assembly solution as the current assembly solution and store it in the data center.

[0076] The current assembly plan refers to the assembly plan that can be used later in the current production line.

[0077] Step 270, in response to reaching the adjustment time, the assembly control instruction is sent to the industrial Internet of Things perception control platform to control the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt.

[0078] The adjustment time refers to the preset time for adjusting the assembly process. In some embodiments, when the adjustment needs to wait until the assembly of the components of the current batch is completed, the adjustment time is related to the assembly time of the components of the current batch. The adjustment time can be preset by the staff based on experience.

[0079] Equipment operating parameters refer to parameters related to the operation process of the production equipment. For example, equipment operating parameters may include robot arm movement speed, torque setting, assembly speed, assembly method, assembly accuracy, etc.

[0080] Conveyor belt parameters refer to parameters related to the operation of the conveyor belt. For example, conveyor belt parameters may include conveyor belt speed, etc.

[0081] In some embodiments, the IIoT management platform can regulate the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt in a variety of ways based on the assembly regulation instructions. For example, when the assembly regulation instructions include increasing the assembly speed of the production equipment and reducing the conveyor belt speed, the IIoT management platform can regulate through the IIoT perception control platform so that the assembly speed of the production equipment is increased accordingly and the conveyor belt speed of the conveyor belt is reduced accordingly.

[0082] Step 280, based on the industrial Internet of Things perception control platform, obtain the reference assembly data of the production line after executing the assembly control instruction.

[0083] Reference assembly data refers to the assembly data after the production line executes the assembly control instructions.

[0084] In some embodiments, the industrial Internet of Things management platform can obtain reference assembly data through a data acquisition device.

[0085] In some embodiments, in response to the reference assembly data satisfying the correction condition, the industrial Internet of Things management platform can perform the following steps 291-294.

[0086] The correction condition may include that the quality sampling data in the reference assembly data is unqualified, etc. The correction condition may be preset by the staff based on experience.

[0087] Step 291, determine the correction and optimization link of the current assembly plan.

[0088] The correction and optimization link refers to the link in the assembly plan that needs to be corrected and optimized.

[0089] In some implementations, the IIoT management platform may determine the correction and optimization link in a variety of ways. For example, the IIoT management platform may select the link with low assembly efficiency and / or low assembly quality after the current assembly solution is applied as the correction and optimization link.

[0090] Step 292, based on the correction optimization link, the current assembly plan is corrected to generate a corrected assembly plan.

[0091] The revised assembly plan refers to the assembly plan after being revised.

[0092] In some embodiments, the Industrial Internet of Things management platform can correct the current assembly plan in a variety of ways based on the correction optimization link. For example, for the correction optimization link with low assembly efficiency, the Industrial Internet of Things management platform increases the assembly rate of the current assembly plan. Among them, the magnitude of the increase in assembly rate is negatively correlated with the current assembly rate. For another example, for the correction optimization link with low assembly quality, the Industrial Internet of Things management platform can improve the assembly accuracy of the current assembly plan. For more information on improving the assembly accuracy of the current assembly plan, please refer to Figure 4 The corresponding description.

[0093] Step 293, storing the revised assembly plan in the data center, and generating a revised control instruction based on the revised assembly plan.

[0094] The revised control instructions refer to the control instructions generated based on the revised assembly plan.

[0095] The method of generating the revised control instructions based on the revised assembly plan is similar to the method of generating the assembly control instructions based on the target assembly plan, which will not be described in detail here.

[0096] Step 294, the correction and control instructions are sent to the industrial Internet of Things perception control platform to correct the equipment operating parameters and conveyor belt parameters of the production equipment.

[0097] The method of correcting the equipment operating parameters and conveyor belt parameters of the production equipment based on the correction control instructions is similar to the method of controlling the equipment operating parameters of the production equipment and the conveyor belt parameters of the conveyor belt based on the assembly control instructions, and will not be elaborated here.

[0098] In some embodiments of the present specification, by dynamically adjusting the assembly scheme through feedback after the actual application of the assembly scheme, the adaptability and flexibility of the system can be improved, problems in actual production can be discovered and corrected in time, and assembly errors and failures can be reduced. By continuously optimizing the assembly scheme, the overall performance and efficiency of the production line can be effectively improved; through actual data verification, the feasibility and reliability of the optimized assembly scheme can be ensured, and production risks can be reduced.

[0099] Figure 3 is an exemplary flow chart for determining an optimized assembly solution according to some embodiments of the present specification.

[0100] In some embodiments, determining an optimized assembly plan based on link assembly data and an initial assembly plan includes: obtaining an assembly feature sequence based on the link assembly data; obtaining assembly sequence requirements and workstation rated parameters of the workstation from a data center; generating candidate assembly plans through a generation layer of a parameter generation model based on the assembly feature sequence, the initial assembly plan, the assembly sequence requirements, and the workstation rated parameters; wherein the parameter generation model is a machine learning model; determining production scoring parameters for the candidate assembly plans through an evaluation layer of a parameter generation model based on the candidate assembly plans, environmental data, material data, historical energy consumption data, and equipment parameters; and determining an optimized assembly plan based on the production scoring parameters and condition parameters.

[0101] like Figure 3 As shown, process 300 includes the following steps. In some embodiments, process 300 can be performed by an industrial Internet of Things management platform. For more information about the industrial Internet of Things management platform, please refer to Figure 1 and its corresponding description.

[0102] Step 310, obtaining an assembly feature sequence based on the link assembly data.

[0103] An assembly feature sequence refers to a sequence consisting of multiple assembly features. Assembly features refer to data related to the assembly process. In some embodiments, the assembly feature sequence includes actual assembly features of the stage assembly components of each link. The actual assembly features include assembly component objects, assembly quality, assembly efficiency, front and rear stacking degree, manual / machine action time, workstation utilization rate, etc.

[0104] The front accumulation degree refers to the accumulation of materials or semi-finished products output from the upstream workstation (previous workstation) of the current workstation at the entrance of the current workstation. The rear accumulation degree refers to the accumulation of materials or semi-finished products output from the current workstation at the entrance of the downstream workstation (next workstation). The workstation utilization rate refers to the utilization rate of the production equipment included in the workstation. For more information about workstations, please refer to Figure 2 The corresponding description.

[0105] In some embodiments, the industrial Internet of Things management platform can obtain the assembly feature sequence in a variety of ways based on the link assembly data. For example, the industrial Internet of Things management platform can obtain the actual assembly features of each link based on the link assembly data within a preset time period.

[0106] As an example only, the industrial Internet of Things management platform can count the average quality, average efficiency, average time of manual / machine actions, etc. of the random inspection samples within a preset time period as the assembly quality, assembly efficiency and manual / machine action time respectively.

[0107] Step 320 , obtaining assembly sequence requirements and workstation rated parameters of the workstation from the data center.

[0108] For more information about data centers, see Figure 1 The corresponding description.

[0109] The assembly sequence requirement refers to the sequence requirement for assembling components. The assembly sequence requirement can also be any sequence.

[0110] The workstation rated parameters are parameters used to characterize the equipment characteristics and equipment processing capabilities of each workstation.

[0111] Equipment characteristics refer to the properties of the production equipment itself, such as equipment type, equipment specifications, etc.

[0112] Equipment processing capacity refers to the ability, scope and efficiency of production equipment to process and complete work. For example, the types of parts that production equipment can handle and the assembly speed that can be achieved.

[0113] Step 330 , based on the assembly feature sequence, the initial assembly plan, the assembly sequence requirements, and the workstation rated parameters, a candidate assembly plan is generated through a generation layer of a parameter generation model.

[0114] For more information on the initial assembly plan, see Figure 2 The corresponding description.

[0115] The parameter generation model refers to a model used to generate candidate assembly solutions and their corresponding production scoring parameters. In some embodiments, the parameter generation model is a machine learning model, such as a recurrent neural network (RNN) model, a deep neural network (DNN) model, etc.

[0116] The generation layer refers to a model layer used to generate candidate assembly solutions. In some embodiments, the generation layer is a machine learning model, such as a recurrent neural network model.

[0117] For more information about generating layers, see Figure 4 The corresponding description.

[0118] A candidate assembly solution refers to a solution that serves as a candidate for optimal assembly solution.

[0119] The Industrial Internet of Things Management Platform can generate candidate assembly solutions through the generation layer of the parameter generation model. For more information about this part, please refer to Figure 4 The corresponding description.

[0120] In some embodiments, the industrial Internet of Things management platform can determine the model through the optimization link based on the assembly feature sequence, workstation rated parameters, equipment parameters, and historical link assembly data, and determine the first optimization link; based on the first optimization link, determine the candidate assembly plan.

[0121] Equipment parameters refer to parameters related to equipment on the production line, such as equipment name, equipment model, equipment status (maintenance record, fault record, etc.).

[0122] In some embodiments, the equipment parameters can be obtained by a data acquisition device or uploaded by a staff member.

[0123] The historical link assembly data refers to the assembly data of the link that needs to be optimized in the historical period. The historical period refers to a period of time in the past, for example, the past half hour.

[0124] The optimization link determination model refers to a model used to determine the optimization link. In some embodiments, the optimization link determination model is a machine learning model, such as a recurrent neural network model.

[0125] In some embodiments, the input of the optimization link determination model includes assembly feature sequence, workstation rated parameters, equipment parameters, and historical link assembly data; and the output includes the first optimization link.

[0126] In some embodiments, the optimization link determination model can be obtained by training in a variety of ways, for example, by training a plurality of third training samples with third labels, etc. A set of third training samples may include sample assembly feature sequences, sample workstation rated parameters, sample equipment parameters, and sample historical link assembly data for the first historical period. The third label corresponding to a set of third training samples is the first optimization link actually used in the second historical period.

[0127] In some embodiments, the industrial Internet of Things management platform can select historical link assembly data, historical workstation rated parameters, and historical equipment parameters in the historical first period of the historical assembly process; determine the historical assembly feature sequence based on the historical link assembly data; use the aforementioned historical link assembly data, historical workstation rated parameters, historical equipment parameters, and historical assembly feature sequence as a group of third training samples; and use the links with low actual assembly efficiency, low assembly quality, or large fluctuations in assembly efficiency and assembly quality in the historical second period of the assembly process under the conditions of the group of third training samples as the third labels corresponding to the group of third samples. The historical first period is before the historical second period.

[0128] Among them, low assembly efficiency and low assembly quality refer to that the assembly efficiency and assembly quality are lower than the efficiency threshold and the quality threshold, respectively. Large fluctuations in assembly efficiency and assembly quality refer to that the fluctuations in assembly efficiency and assembly quality are greater than the efficiency fluctuation threshold and the quality fluctuation threshold. In some embodiments, the assembly efficiency fluctuation and assembly quality fluctuation can be represented by the difference between the upper and lower limits of the assembly efficiency and assembly quality. The efficiency threshold, the quality threshold, the efficiency fluctuation threshold and the quality fluctuation threshold can all be preset by the staff based on experience.

[0129] In some embodiments, the industrial Internet of Things management platform can input the third training sample into the initial link optimization model, construct a third loss function based on the output of the initial link optimization model and the third label, and iteratively update the initial link optimization model based on the third loss function. When the preset conditions are met, the training is completed and a trained link optimization model is obtained. The preset conditions may be that the third loss function converges, the number of iterations reaches a threshold, etc.

[0130] The first optimization link refers to the link where the assembly efficiency and assembly quality of the assembly solution at the current or future moment are low or fluctuate greatly.

[0131] In some embodiments, the IIoT management platform can determine candidate assembly solutions in a variety of ways based on the first optimization link. For example, the IIoT management platform can generate multiple candidate assembly solutions by randomly adjusting the first optimization link based on the first optimization link, in combination with various adjustment ranges and using various adjustment means. The adjustment means may include, but are not limited to, increasing the assembly speed, increasing the assembly accuracy, etc. As an example only, in actual production, the change in assembly accuracy can be achieved by changing the assembly process. Changing the assembly process may include: using assembly tools of different accuracy, using more accurate speed and pressure control solutions, introducing more precise positioning systems, etc.

[0132] In some embodiments of the present specification, the accuracy of the first optimization link can be ensured and the time loss of the calculation process can be reduced through a machine learning model; by adjusting the first optimization link to obtain candidate assembly solutions, the efficiency of obtaining candidate assembly solutions can be improved, and the candidate assembly solutions can be made more targeted.

[0133] Step 340, based on the candidate assembly scheme, environmental data, material data, historical energy consumption data, and equipment parameters, the production scoring parameters of the candidate assembly scheme are determined through the evaluation layer of the parameter generation model.

[0134] Environmental data refers to data related to the environment in which the assembly is located. For example, ambient temperature, ambient humidity, etc. In some embodiments, the environmental data can be obtained by a data acquisition device. For more information about the data acquisition device, see Figure 1 The corresponding description.

[0135] Material data refers to data related to the materials used in the assembly process, such as the name, specification, and material of the parts or components used to build the component.

[0136] Historical energy consumption data refers to data related to the energy consumed during the assembly process, such as electrical energy consumption.

[0137] The evaluation layer refers to a model layer for generating production scoring parameters of candidate assembly solutions. In some embodiments, the evaluation layer can be a machine learning model, such as a deep neural network model.

[0138] For more information about the evaluation layer, see Figure 4 The corresponding description.

[0139] The production scoring parameters refer to the relevant data for scoring the candidate assembly solutions, for example, the production scoring parameters may include the production rate, the defect rate, and the failure rate.

[0140] Step 350, determining an optimized assembly solution based on the production scoring parameters and condition parameters.

[0141] Condition parameters refer to pre-set parameters related to determining the optimal assembly solution. For example, condition parameters may include a preset production rate, a preset scrap rate, and a preset failure rate. The preset production rate, the preset scrap rate, and the preset failure rate may be preset by the staff based on experience.

[0142] In some embodiments, the industrial Internet of Things management platform can select a candidate assembly plan with a production rate greater than a preset production rate, a scrap rate lower than a preset scrap rate, and a failure rate lower than a preset failure rate as an optimized assembly plan based on production scoring parameters and condition parameters.

[0143] In some embodiments of the present specification, the obtained assembly feature sequence can be made consistent with the actual production process through link assembly data; determining candidate assembly schemes and their corresponding production parameter scores based on a machine learning model can make the determination of candidate assembly schemes more accurate and save time in the determination process; based on production scoring parameters and condition parameters, the accuracy of the optimized assembly scheme can be improved.

[0144] Figure 4 is an exemplary schematic diagram of a parameter generation model according to some embodiments of the present specification.

[0145] In some embodiments, the parameter generation model 450 is obtained by training a training sample data set, and the training process of the parameter generation model 450 includes an initial training phase and an enhanced training phase. Figure 4 As shown, the training samples in the training sample data set include a first training sample 410, a second training sample 420, a first label corresponding to the first training sample, and a second label corresponding to the second training sample.

[0146] The first training sample 410 is applied to the training generation layer 451, the first training sample 410 includes a sample assembly feature sequence 411, a sample initial assembly solution 412, a sample assembly sequence requirement 413, and a sample workstation rated parameter 414, and the first label includes an available assembly solution 430 corresponding to the first training sample; The second training sample 420 is applied to the training evaluation layer 452. The second training sample 420 includes a sample assembly solution 421, sample environmental data 422, sample material data 423, sample energy consumption data 424, and sample equipment parameters 425. The second label includes the actual production scoring parameter 440 corresponding to the second training sample. For more information about the parameter generation model, generation layer, and evaluation layer, see Figure 3 The corresponding description.

[0147] The training sample dataset refers to the dataset used to train the parameter generation model.

[0148] In some embodiments, the industrial Internet of Things management platform can obtain the first training sample and the first label corresponding to the first training sample in a variety of ways. For example, the industrial Internet of Things management platform can analyze the historical assembly schemes, select the historical subsequent assembly schemes whose adjusted production rate is higher than the production rate of the historical initial assembly scheme, use the historical subsequent assembly scheme as the first label, use the historical initial assembly scheme corresponding to the historical subsequent assembly scheme as the sample initial assembly scheme, and use the historical assembly feature sequence, historical assembly sequence requirements, and historical workstation rated parameters corresponding to the historical initial assembly scheme as the sample assembly feature sequence, sample assembly sequence requirements, and sample workstation rated parameters, respectively.

[0149] The industrial Internet of Things management platform can obtain multiple groups of first training samples and first labels corresponding to the first training samples according to the above method based on multiple historical assembly plans.

[0150] For another example, for an initial assembly plan, the staff can generate multiple experimental assembly plans based on experience, and use the initial assembly plan and the experimental assembly plan on the production line, and select the experimental assembly plan with a production rate higher than the initial assembly plan as the candidate assembly plan corresponding to the initial assembly plan, i.e., the first label, and use the assembly feature sequence, assembly sequence requirements, and workstation rated parameters corresponding to the initial assembly plan as the sample assembly feature sequence, sample assembly sequence requirements, and sample workstation rated parameters, respectively; repeat this process multiple times to obtain multiple groups of first training samples and the first labels corresponding to the first training samples.

[0151] In some embodiments, the second training sample and the second label may be obtained based on simulation data.

[0152] Simulation data refers to the relevant data obtained by simulating the assembly process.

[0153] In some embodiments, the industrial Internet of Things management platform simulates obtaining the second training sample and the second label corresponding to the second training sample based on the simulation tool. For more information about the simulation tool and how to perform the simulation, please refer to Figure 5 The corresponding description.

[0154] In some embodiments of the present specification, the second training sample and the second label are obtained by simulation, so as to ensure the accuracy of the obtained training data and reduce the time consumption of obtaining the training data.

[0155] The initial training phase refers to the pre-training phase before the parameter generation model is connected to a specific factory. In the initial training phase, the training sample data set is obtained based on the general data set on the cloud platform.

[0156] The intensive training phase refers to the phase in which the parameter generation model is trained based on personalized customization for each factory. In the intensive training phase, the training sample data set is obtained based on historical data actually collected in the factory, and the proportion of samples corresponding to one fault type is not less than the preset threshold.

[0157] Fault type refers to the type of fault that occurred, for example, conveyor belt fault, robot arm fault, etc.

[0158] The preset threshold value can be preset by the staff based on experience.

[0159] In some embodiments, the preset threshold is positively correlated to the loss cost corresponding to a fault type.

[0160] Loss cost refers to the cost of losses caused by failure.

[0161] In some embodiments, the industrial Internet of Things management platform can determine the loss cost corresponding to the fault type by querying a preset table. The preset table refers to a table including the correspondence between fault types and loss costs. Exemplarily, the industrial Internet of Things management platform can count the actual loss cost corresponding to a historical fault type and a historical obstacle type in the historical data, average the actual loss costs corresponding to multiple historical obstacle types, and use the average as the loss cost corresponding to the historical fault type, and enter it into the table; the industrial Internet of Things management platform obtains multiple fault types and their corresponding loss costs through the above method.

[0162] In some embodiments, the industrial Internet of Things management platform can obtain the generation layer and the evaluation layer through separate training, or can obtain the generation layer and the evaluation layer through joint training.

[0163] The process of training the generative layer alone is as follows: In some embodiments, the industrial Internet of Things management platform can input the first training sample into the initial generation layer, construct a first loss function based on the output of the initial generation layer and the first label, and iteratively update the initial generation layer based on the first loss function; when the first preset condition is met, the training is completed and a trained generation layer is obtained. The first preset condition can be that the first loss function converges, the number of iterations reaches a threshold, etc.

[0164] The method of training the evaluation layer separately is similar to the method of training the generation layer separately, so we will not go into details here.

[0165] The process of jointly training the generation layer and the evaluation layer is as follows: In some embodiments, the industrial Internet of Things management platform can input the first training sample into the initial generation layer to obtain the output of the initial generation layer; input the output of the initial generation layer and the parameters of other second training samples, that is, sample environment data, sample material data, sample energy consumption data, and sample equipment parameters, into the initial evaluation layer to obtain the output of the initial evaluation layer; construct a second loss function based on the output of the initial evaluation layer and the second label; iteratively update the initial evaluation layer and the initial generation layer based on the second loss function; when the second preset condition is met, the training is completed to obtain the trained generation layer and evaluation layer. The second preset condition can be that the second loss function converges, the number of iterations reaches a threshold, etc.

[0166] In some embodiments, the input of the generation layer of the parameter generation model also includes a first optimization step.

[0167] In some embodiments, when the input of the generation layer of the parameter generation model also includes a first optimization step, the first training sample also includes a sample first optimization step.

[0168] In some embodiments, the industrial Internet of Things management platform can obtain the first training sample including the first optimization link of the sample in a variety of ways. For example, for the initial assembly plan, the staff can generate the first optimization link of the initial assembly plan based on experience, further adjust the first optimization link to obtain multiple groups of experimental assembly plans, use the initial assembly plan and the experimental assembly plan on the production line, select the experimental assembly plan with a production rate higher than the initial assembly plan as the candidate assembly plan corresponding to the initial assembly plan, that is, the first label, and use the first optimization link, assembly feature sequence, assembly sequence requirements, and workstation rated parameters corresponding to the initial assembly plan as the sample first optimization link, sample assembly feature sequence, sample assembly sequence requirements, and sample workstation rated parameters, respectively; repeat multiple times to obtain multiple groups of first training samples and labels corresponding to the first training samples.

[0169] For more information about the first optimization step, see Figure 3 The corresponding description.

[0170] In some embodiments of the present specification, by using the first optimization link as the input of the generation layer, the model can focus on the specific problems of this link, and the generated candidate assembly solutions are more targeted rather than randomly generated generalized solutions. This also avoids generating a large number of irrelevant candidate solutions, saving computing resources and time. At the same time, it can also ensure that the generated candidate assembly solutions are effective and feasible, thereby significantly improving the overall performance of the assembly line.

[0171] In some embodiments of the present specification, a large number of general data sets are used for pre-training in the initial training phase, so that the parameter generation model has good generalization ability in a wide range of situations, and further enables the parameter generation model to adapt faster when it first contacts the data of a specific factory; in the intensive training phase, personalized customized training is performed based on the data actually collected from each factory, which further improves the adaptability and accuracy of the model in a specific factory environment. This phased training method ensures that the model has both wide applicability and targeted optimization capabilities. In addition, in the intensive training phase, ensuring that the proportion of samples corresponding to each fault type is not less than the preset threshold helps the model to better learn various fault modes and improve the accuracy of fault detection and processing. Setting a preset threshold is positively correlated to the loss cost corresponding to the fault type, so that the model pays more attention to the fault types that will cause greater losses once they occur, thereby giving priority to optimizing these key links.

[0172] Figure 5 is an exemplary flow chart for determining a target assembly solution according to some embodiments of the present specification.

[0173] In some embodiments, based on the optimized assembly scheme, determining the target assembly scheme includes: simulating the production line based on preset simulation parameters and the optimized assembly scheme, obtaining simulated assembly data and simulated production data corresponding to the optimized assembly scheme, the preset simulation parameters including a simulation intensity sequence; determining a simulation feature sequence based on the simulated assembly data; determining an optimized effectiveness sequence corresponding to the optimized assembly scheme based on the simulation feature sequence and the assembly feature sequence; and determining the target assembly scheme based on the optimized effectiveness sequence, the simulated production data, and the optimized assembly scheme.

[0174] like Figure 5 As shown, process 500 includes the following steps. In some embodiments, process 500 can be performed by an industrial Internet of Things management platform. For more information about the industrial Internet of Things management platform, please refer to Figure 1 and its corresponding description.

[0175] Step 510, based on preset simulation parameters and the optimized assembly plan, simulate the production line to obtain simulated assembly data and simulated production data corresponding to the optimized assembly plan, and the preset simulation parameters include a simulation intensity sequence.

[0176] For instructions on optimizing assembly plans and production lines, see Figure 2 and related descriptions.

[0177] The preset simulation parameters refer to the parameters preset for simulating the production line.

[0178] The simulation intensity sequence refers to the sequence combination of simulation efficiency and simulation intensity.

[0179] In some embodiments, the simulation intensity sequence may include the modeling accuracy of each link, the selection of simulation algorithms, the allocation of computing resources, etc. Among them, the allocation of computing resources may include GPU / CPU and content allocation, etc.

[0180] In some embodiments, the industrial Internet of Things management platform can determine the simulation parameters of each assembly link based on the assembly level and assembly type.

[0181] The assembly level is used to characterize the assembly order of a link. The lower the assembly level, the earlier the assembly order of the link. The assembly type refers to the type of components used for assembly in this link. The assembly level and assembly type can be expressed by numerical values. For example, in the first link, a and b are assembled together to form c; in the second link, d and e are assembled together to form f; in the third link, c, f and h are assembled together to form g. Among them, the first and second links are the basis of the third link. Therefore, the assembly level in the first link is 1, and the assembly types are 2; the assembly level in the second link is 1, and the assembly types are 2; the assembly level in the third link is 2, and the assembly types are 3.

[0182] In some embodiments, the higher the assembly level of the link and the more assembly types there are, the greater the simulation intensity of the link. That is, the higher the modeling accuracy of the link, the stronger the selected simulation algorithm and the more computing resources need to be allocated.

[0183] In some embodiments, the preset simulation parameters are positively correlated to the link complexity parameters, which are determined based on the workstation rated parameters, component interconnection relationships, assembly component objects, and process difficulty coefficients.

[0184] The link complexity parameter can characterize the complexity and difficulty of the link assembly. For example, the more assembly types and assembly steps involved in the link assembly process, the longer the assembly time at the same assembly speed, and the greater the assembly error rate, the greater the link complexity parameter.

[0185] In some embodiments, the link complexity parameter can be represented by a numerical value. The larger the numerical value of the link complexity parameter is, the greater the complexity of the link assembly is.

[0186] For more information on the workstation ratings, see Figure 3 The corresponding description.

[0187] Component interconnection refers to the connection relationship between components assembled at a workstation.

[0188] In some embodiments, the component interconnection relationship may include connection type and connection order.

[0189] The connection type may be at least one of bolt connection, welding, bonding, etc.

[0190] The connection sequence means that certain parts must be assembled first, which serves as the basis and prerequisite for subsequent assembly steps.

[0191] Assembly component objects refer to the components involved in the assembly process.

[0192] The process difficulty coefficient is an indicator for evaluating the difficulty of the process.

[0193] In some embodiments, the process difficulty coefficient may include process accuracy and operation complexity.

[0194] Process accuracy is the accuracy required for component assembly. For example, the assembly of some components requires an accuracy of ±0.01 mm, which requires high process accuracy.

[0195] Operational complexity refers to the complexity of the assembly process. For example, some assembly steps require multi-axis synchronous motion, which has high operational complexity.

[0196] In some embodiments, the higher the process precision and the greater the operation complexity, the greater the process difficulty coefficient.

[0197] In some embodiments, the process difficulty coefficient may be obtained by actually measuring operation data of the target process.

[0198] In some embodiments, the link complexity parameters can be determined by vector matching. For example, the industrial Internet of Things management platform can construct a process feature vector based on the workstation rated parameters, component interconnection relationships, assembly component objects, and process difficulty coefficients; based on the process feature vector, the vector database is retrieved to determine the link complexity parameters.

[0199] The process feature vector is a vector used to characterize the process feature. For example, the process feature vector can be expressed as (x0, y0, z0, t0). Among them, x0, y0, z0, t0 can represent the workstation rated parameters, component interconnection relationship, assembly component object and process difficulty coefficient corresponding to the workstation numbered 0.

[0200] Vector databases refer to databases that support similarity search of high-dimensional vectors, such as Milvus and Faiss.

[0201] In some embodiments, the vector database includes a plurality of reference feature vectors and reference complex parameters corresponding to the reference feature vectors.

[0202] In some embodiments, the vector database can be constructed based on historical data. For example, the reference feature vector can be constructed based on the workstation rated parameters, component interconnection relationship, assembly component object and process difficulty coefficient in the historical production and assembly process. The construction method of the reference feature vector is similar to the construction method of the aforementioned process feature vector, which will not be repeated here. Among them, the more types of parts / components involved in the historical production and assembly process, the more assembly steps, the longer the assembly time at the same assembly speed, and the greater the assembly error rate, the greater the link complexity parameter. The assembly error rate includes the probability of occurrence of errors such as misalignment, non-assembly, and looseness. At the same time, the reference complex parameters corresponding to the reference feature vector can be obtained by calling the historical production data stored in the system. According to the above method, multiple reference feature vectors and reference complex parameters corresponding to the reference feature vectors can be obtained, and a vector database can be constructed based on multiple reference feature vectors and reference complex parameters corresponding to the reference feature vectors.

[0203] In some embodiments, the industrial Internet of Things management platform can determine the link complexity parameters based on the similarity between the process feature vector and multiple reference feature vectors in the vector database. For example, the industrial Internet of Things management platform can use the reference feature vector whose similarity with the process feature vector meets the preset conditions as the target vector, and the reference complexity parameters corresponding to the target vector as the link complexity parameters. The preset conditions can be set according to the situation. For example, the similarity is maximum or the similarity is greater than the similarity threshold, etc. The similarity threshold can be set based on experience.

[0204] In some embodiments of the present specification, link complexity parameters are determined based on four dimensions: workstation rated parameters, component interconnection relationships, assembly component objects, and process difficulty coefficients. This can avoid the one-sidedness of evaluating process complexity in a single dimension and make the determined link complexity parameters more objective and accurate.

[0205] Simulation assembly data refers to the link assembly data obtained during the simulation process.

[0206] The simulated production data refers to the production data obtained by simulating the production line based on the optimized assembly plan. In some implementations, the simulated production data may include the scrap rate and the failure rate.

[0207] In some embodiments, the industrial Internet of Things management platform can simulate the production line through a simulation tool based on at least one preset simulation parameter and at least one optimized assembly solution. The simulation tool can be a simulation software of virtual reality (VR) or augmented reality (AR) technology. For example, Unity + VRTK, ARKit, etc.

[0208] In some embodiments, the simulation tool may simulate the entire production line, or may simulate one or more production links of the production line separately.

[0209] In some embodiments, the link simulation requires the output results of the upstream link as input data. The simulation tool can determine the simulation order based on the assembly level. For example, the simulation tool prioritizes simulating links with low assembly levels, and simulates links with high assembly levels based on the output results of the links with low assembly levels.

[0210] In some embodiments, the industrial Internet of Things management platform can obtain at least one set of simulated assembly data corresponding to at least one optimized assembly solution.

[0211] Step 520, determining a simulation feature sequence based on the simulation assembly data.

[0212] The simulation feature sequence refers to the feature information sequence of the production line during the simulation process.

[0213] In some embodiments, the industrial Internet of Things management platform 130 can determine at least one set of simulation feature sequences based on the simulated assembly data. For a description of the simulation feature sequence, see Figure 3 Description of the assembly feature sequence in .

[0214] Step 530: Determine the optimized effectiveness sequence corresponding to the optimized assembly solution based on the simulation feature sequence and the assembly feature sequence.

[0215] For more information about assembly feature sequences, see Figure 3 For instructions on optimizing assembly solutions, see Figure 2 and related descriptions.

[0216] The optimization validity sequence is used to evaluate the effectiveness of the modified optimization link. There is at least one optimization validity sequence corresponding to an optimized assembly plan.

[0217] In some embodiments, the industrial Internet of Things management platform can determine the optimized effectiveness sequence by comparing the assembly quality, assembly efficiency, front stacking degree, back stacking degree and workstation utilization rate in the simulation feature sequence and the assembly feature sequence. Exemplarily, the optimized effectiveness sequence of each assembly link can be calculated by the following formula (1): A=B(B1-B2)+C(C1-C2)+D(D1-D2)+E(E1-E2)+F(F1-F2)(1) Among them, A is the optimization effectiveness; B is the assembly quality coefficient; B1 is the simulated assembly quality; B2 is the actual assembly quality; C is the assembly efficiency coefficient; C1 is the simulated assembly efficiency; C2 is the actual assembly efficiency; D is the front stacking degree coefficient; D1 is the actual front stacking degree; D2 is the simulated front stacking degree; E is the rear stacking degree coefficient; E1 is the actual rear stacking degree; E2 is the simulated rear stacking degree; F is the workstation utilization coefficient; F1 is the simulated workstation utilization, and F2 is the actual workstation utilization. Based on the optimization effectiveness of each assembly link, the optimization effectiveness sequence of each assembly link can be further determined.

[0218] Among them, coefficients B, C, D, E, and F can be preset through experience; B1, C1, D2, E1, and F1 can be obtained through simulation feature sequences; and B2, C2, D1, E2, and F2 can be obtained through actual assembly feature sequences.

[0219] Step 540, determining a target assembly solution based on optimizing the effective degree sequence, simulating production data, and optimizing the assembly solution.

[0220] For a description of the target assembly scheme, see Figure 2 and related descriptions.

[0221] In some embodiments, the industrial Internet of Things management platform can select a group of optimized assembly solutions with the highest total optimization effectiveness, a scrap rate less than a scrap threshold, and a failure rate less than a failure threshold as the target assembly solution based on multiple optimized assembly solutions.

[0222] Among them, the total optimization effectiveness is used to evaluate the effectiveness of the optimized assembly solution.

[0223] In some embodiments, the total optimization effectiveness can be determined in a variety of ways. For example, the total optimization effectiveness can be the sum or weighted sum of the optimization effectiveness corresponding to all the modified optimization links. The weight corresponding to each modified optimization link is related to the importance of the modified optimization link. The more important the modified optimization link is, the greater the weight.

[0224] In some embodiments, the importance of the correction and optimization link can be determined in a variety of ways. For example, the importance of the correction and optimization link can be determined by those skilled in the art based on experience. For another example, the degree of influence of each correction and optimization link on the assembly efficiency and assembly quality of the overall assembly process can be determined by statistical methods (such as regression analysis, correlation analysis, etc.), and then the importance of the correction and optimization link can be determined. Exemplarily, when the assembly accuracy of link M decreases or fails, it will cause the scrap rate of the finished product to increase or the quality to decrease, then it can be considered that link M is more important. The greater the increase in scrap rate and / or the decrease in quality, the greater the importance of the link.

[0225] In some embodiments, the industrial Internet of Things management platform can determine a second optimization link based on the optimization effectiveness sequence and the effectiveness threshold; based on the second optimization link, update the target assembly plan to determine an updated target assembly plan.

[0226] The effectiveness threshold is used to evaluate whether the optimized assembly solution has achieved the expected effectiveness.

[0227] In some embodiments, the effectiveness threshold is positively correlated to the link complexity parameter.

[0228] The second optimization link refers to the link that can be optimized again during the simulation process, for example, a link with a high degree of importance and / or a link with poor performance.

[0229] In some embodiments, the industrial Internet of Things management platform may determine, in the optimized assembly solution, an assembly link whose optimization effectiveness is lower than an effectiveness threshold and / or whose assembly rate is lower than a rate threshold as a second optimization link. The rate threshold may be obtained by manually presetting.

[0230] In some embodiments, the industrial Internet of Things management platform can also obtain a second optimization link through an optimization link determination model based on a simulated feature sequence. Figure 3In the description of obtaining the first optimization step based on the optimization step determination model.

[0231] In some embodiments, the industrial Internet of Things management platform can directly update the optimized assembly solution based on the second optimization link. For example, the parameters of the workstation corresponding to the second optimization link can be directly and randomly adjusted. In some embodiments, the industrial Internet of Things management platform can input the second optimization link into the generation layer of the parameter generation model to obtain multiple updated candidate assembly solutions. The relevant instructions can be similarly referred to Figure 3 The corresponding description in .

[0232] In some embodiments, the optimized assembly scheme with the second optimization link can be determined as the updated candidate assembly scheme; based on the evaluation layer of the parameter generation model, multiple updated candidate schemes are evaluated to see whether they meet the preset conditions, and the updated candidate schemes that meet the preset conditions can be used as the updated optimized assembly scheme. The preset condition can be that at least one of the production rate, the defect rate, and the failure rate is better than the corresponding parameter in the target assembly scheme.

[0233] In some embodiments, the simulation tool can optimize the assembly solution based on the update, by Figure 5 The method of steps 510 to 540 in the above embodiment simulates the production line to determine the updated target assembly plan. The updated target assembly plan can determine the second optimization link again, and based on the second optimization link, directly update the optimized assembly plan, or according to Figure 5 The method of step 510-step 540 is iterated and updated multiple times until the preset target is reached, and the updated target assembly scheme is determined. The preset target can be that the number of updates reaches the preset number of updates, or the difference between the total optimization effectiveness determined by the last simulation and the total optimization effectiveness determined by the last simulation is less than the update threshold. The update threshold can be preset based on experience.

[0234] In some embodiments of the present specification, multiple optimization updates are performed based on key links and / or unreasonable links in the optimized assembly scheme to determine a method for an updated target assembly scheme. The optimized assembly scheme can be optimized more specifically, and defects in the optimized assembly scheme can be gradually eliminated, thereby making the updated target assembly scheme more perfect.

[0235] In some embodiments of the present specification, by simulating the optimized assembly scheme and determining the target assembly scheme, not only can the feasibility of the preset scheme be verified, potential problems in the scheme can be discovered and solved in advance, and the additional costs and losses caused by production errors can be reduced; the production line design can also be optimized, resource allocation can be further optimized, and assembly efficiency can be improved.

[0236] It should be noted that the above descriptions of the processes 200, 300 and 500 are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the processes 200, 300 and 500 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

Claims

1. An assembly optimization method based on industrial Internet of Things, characterized in that: The method is performed by an industrial Internet of Things management platform based on the industrial Internet of Things, wherein the industrial Internet of Things includes an industrial Internet of Things user platform, an industrial Internet of Things service platform, the industrial Internet of Things management platform, an industrial Internet of Things sensor network platform and an industrial Internet of Things perception control platform; The industrial Internet of Things management platform is communicatively connected with the industrial Internet of Things sensor network platform and the industrial Internet of Things service platform; the industrial Internet of Things perception control platform includes production equipment of the production line and a data acquisition device deployed on the production line, and the industrial Internet of Things perception control platform realizes data interaction with the industrial Internet of Things management platform through the industrial Internet of Things sensor network platform; The method comprises: Acquire the link assembly data of the production line through the industrial Internet of Things perception control platform, and upload the link assembly data to the industrial Internet of Things management platform via the industrial Internet of Things sensor network platform; Acquire an initial assembly plan from a plan database, wherein the plan database is stored in a data center on the industrial Internet of Things management platform; Determine an optimized assembly plan based on the link assembly data and the initial assembly plan; Based on the optimized assembly scheme, determining a target assembly scheme; Based on the target assembly plan, generating assembly control instructions; Determine the target assembly solution as the current assembly solution and store it in the data center; In response to reaching the adjustment time, the assembly control instruction is sent to the industrial Internet of Things perception control platform to control the equipment operation parameters of the production equipment and the conveyor belt parameters of the conveyor belt; Based on the industrial Internet of Things perception control platform, obtaining reference assembly data of the production line after executing the assembly control instruction; In response to the reference assembly data satisfying a correction condition: Determining a correction and optimization link of the current assembly plan; Based on the correction optimization link, the current assembly plan is corrected to generate a corrected assembly plan; storing the revised assembly plan in the data center, and generating a revised control instruction based on the revised assembly plan; The correction and control instructions are sent to the industrial Internet of Things perception control platform to correct the equipment operating parameters and the conveyor belt parameters of the production equipment.

2. The method according to claim 1, characterized in that The step of determining the optimized assembly plan based on the link assembly data and the initial assembly plan includes: Based on the link assembly data, obtaining an assembly feature sequence; obtaining assembly sequence requirements and workstation rated parameters of the workstation from the data center; Based on the assembly feature sequence, the initial assembly plan, the assembly sequence requirements, and the workstation rated parameters, a candidate assembly plan is generated through a generation layer of a parameter generation model; wherein the parameter generation model is a machine learning model; Based on the candidate assembly scheme, environmental data, material data, historical energy consumption data, and equipment parameters, determining the production scoring parameters of the candidate assembly scheme through the evaluation layer of the parameter generation model; The optimized assembly plan is determined based on the production scoring parameters and condition parameters.

3. The method according to claim 2, characterized in that The method further comprises: Based on the assembly feature sequence, the workstation rated parameters, the equipment parameters, and historical link assembly data, a first optimization link is determined by determining a model through an optimization link; wherein the optimization link determination model is a machine learning model; Based on the first optimization step, the candidate assembly scheme is determined.

4. The method according to claim 1, characterized in that: Determining a target assembly scheme based on the optimized assembly scheme includes: Based on preset simulation parameters and the optimized assembly scheme, the production line is simulated to obtain simulated assembly data and simulated production data corresponding to the optimized assembly scheme, wherein the preset simulation parameters include a simulation intensity sequence; determining a simulation feature sequence based on the simulated assembly data; Determining an optimized effectiveness sequence corresponding to the optimized assembly scheme based on the simulation feature sequence and the assembly feature sequence; The target assembly plan is determined based on the optimized validity sequence, the simulated production data, and the optimized assembly plan.

5. The method according to claim 4, characterized in that The preset simulation parameters are positively correlated with the link complexity parameters; The link complexity parameters are determined based on workstation rated parameters, component interconnection relationships, assembly component objects, and process difficulty coefficients.

6. The method according to claim 4, characterized in that The method further comprises: Based on the optimization validity sequence and the validity threshold, determining a second optimization link; Based on the second optimization step, the target assembly solution is updated to determine an updated target assembly solution.

7. An assembly optimization system based on industrial Internet of Things, characterized in that: Including Industrial Internet of Things user platform, Industrial Internet of Things service platform, Industrial Internet of Things management platform, Industrial Internet of Things sensor network platform and Industrial Internet of Things perception control platform; The industrial Internet of Things management platform is communicatively connected with the industrial Internet of Things sensor network platform and the industrial Internet of Things service platform; the industrial Internet of Things perception control platform includes production equipment of the production line and a data acquisition device deployed on the production line, and the industrial Internet of Things perception control platform realizes data interaction with the industrial Internet of Things management platform through the industrial Internet of Things sensor network platform; The industrial Internet of Things management platform is configured to: Acquire the link assembly data of the production line through the industrial Internet of Things perception control platform, and upload the link assembly data to the industrial Internet of Things management platform via the industrial Internet of Things sensor network platform; Acquire an initial assembly plan from a plan database, wherein the plan database is stored in a data center on the industrial Internet of Things management platform; Determine an optimized assembly plan based on the link assembly data and the initial assembly plan; Based on the optimized assembly scheme, determining a target assembly scheme; Based on the target assembly plan, generating assembly control instructions; Determine the target assembly solution as the current assembly solution and store it in the data center; In response to reaching the adjustment time, the assembly control instruction is sent to the industrial Internet of Things perception control platform to control the equipment operation parameters of the production equipment and the conveyor belt parameters of the conveyor belt; Based on the industrial Internet of Things perception control platform, obtaining reference assembly data of the production line after executing the assembly control instruction; In response to the reference assembly data satisfying a correction condition: Determining a correction and optimization link of the current assembly plan; Based on the correction optimization link, the current assembly plan is corrected to generate a corrected assembly plan; storing the revised assembly plan in the data center, and generating a revised control instruction based on the revised assembly plan; The correction and control instructions are sent to the industrial Internet of Things perception control platform to correct the equipment operating parameters and the conveyor belt parameters of the production equipment.

8. The system according to claim 7, characterized in that The industrial Internet of Things management platform is further configured to: Based on the link assembly data, obtaining an assembly feature sequence; obtaining assembly sequence requirements and workstation rated parameters of the workstation from the data center; Based on the assembly feature sequence, the initial assembly plan, the assembly sequence requirements, and the workstation rated parameters, a candidate assembly plan is generated through a generation layer of a parameter generation model; wherein the parameter generation model is a machine learning model; Based on the candidate assembly scheme, environmental data, material data, historical energy consumption data, and equipment parameters, determining the production scoring parameters of the candidate assembly scheme through the evaluation layer of the parameter generation model; The optimized assembly plan is determined based on the production scoring parameters and condition parameters.

9. The system according to claim 7, characterized in that The industrial Internet of Things management platform is further configured to: Based on preset simulation parameters and the optimized assembly scheme, the production line is simulated to obtain simulated assembly data and simulated production data corresponding to the optimized assembly scheme, wherein the preset simulation parameters include a simulation intensity sequence; determining a simulation feature sequence based on the simulated assembly data; Determining an optimized effectiveness sequence corresponding to the optimized assembly scheme based on the simulation feature sequence and the assembly feature sequence; The target assembly plan is determined based on the optimized validity sequence, the simulated production data, and the optimized assembly plan.

10. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the method as claimed in claim 1.

Images