CN117590817A - An end-to-end production line equipment optimization system based on digital twins - Google Patents

Abstract

The present invention provides an end-to-end production line equipment optimization system based on digital twins, including: a factory end, which is used by the factory, a supplier side, which is used by suppliers who provide products to the factory, and an intermediate platform, which serves as intermediate data The interactive processing platform is connected to the factory-side and supplier-side data respectively. Based on the data uploaded by the factory side, a digital twin of the factory is established on the intermediate platform. Based on the target equipment and/or equipment that need to be optimized defined in the digital twin factory by the factory side, In the target area, the intermediate platform generates a query program about the target device and/or all devices in the target area. After the expected performance parameters for at least one target device are obtained from the factory side, the intermediate platform is based on the parameters between the selected devices. The process association automatically shares some of the expected performance parameters with the associated equipment. When the expected performance parameters are determined, the intermediate platform performs the matching operation of the product twin plug-in from the database.

Description

An end-to-end production line equipment optimization system based on digital twins

Technical field

The invention relates to the field of factory production line equipment optimization and upgrading, and in particular to an end-to-end production line equipment optimization system based on digital twins. Belongs to G06F classification.

Background technique

The production line in the factory has a large number of component equipment, and each component equipment assumes certain responsibilities and implements certain functions in the production line. According to the degree of importance, some equipment plays a relatively important role in the production line, while some equipment plays a relatively low role. However, no matter what kind of equipment it is, when the equipment is aging, approaching the end of its service life, and needs to be replaced, factory technicians are required to conduct equipment optimization analysis to provide corresponding equipment optimization plans. According to the current equipment optimization analysis model of most factories, technicians first need to investigate the basic information of the equipment that needs to be optimized, such as the current working parameters of the equipment; secondly, technicians need to design the optimized equipment based on the goals expected to be achieved by the optimized equipment. It has the expected performance parameters; then the technicians find suitable product equipment based on the expected performance parameters of the design. Since most of the optimized replacement of equipment is not repair or modification based on the original equipment, but directly replaces other models of existing products, another supplier is introduced as the source of these existing products. The supplier manufactures product equipment that can be upgraded and replaced for the production line, and the factory inquires and purchases suitable product equipment from the supplier. This is an end-to-end transaction. Suppliers generally provide product catalogs and product parameters by distributing product specifications to factories. Factory technicians check product specifications based on the expected performance parameters of the design and select the most suitable product for optimized replacement.

Using product specifications to query equipment is an inefficient way, so there are existing technologies that establish networked data connections between suppliers and factories. For example, CN112100965A discloses a collaborative innovation platform for electronic manufacturing and its use method. In this technical solution, designers design the products required for simulation through the design tool cloud and IP service cloud of the client. The design files are encrypted by the collaborative innovation platform and then transmitted. to the backend management end; the intelligent identification module receives and decrypts the encrypted design file, automatically matches each supplier's equipment that conforms to the product process and process parameters through the deep mapping algorithm based on the decrypted design file, and generates corresponding encrypted orders and sends them to the supplier; supply After decryption by the merchant, based on the product process flow, process parameters and corresponding supplier equipment recorded in the order, the production scene, supplier equipment and production process are simulated through the digital twin cloud, and a 3D panoramic display platform is generated for online visualization. This technical solution uses the backend management platform to establish a visual model that matches the client's needs, thereby providing intuitive feedback on the client's expected effects so that adaptive adjustments can be made based on the model results. However, the above-mentioned existing technology is limited to the client's demand adjustment and cannot reflect the supplier's adjustment needs.

On the one hand, manufacturers will add some parameter items that they are concerned about when designing expected performance parameters for optimizing equipment. However, these parameter items are not necessarily in the product data list provided by the supplier. In fact, the supplier may not have included them during design or delivery. The detection noticed the corresponding parameters and did not measure them, which led to certain difficulties in the work of "finding existing equipment products based on the expected performance parameters designed by the factory." For example, it was difficult for the system to match equipment products with these additional added parameters. Or the system ignores these additional parameters, making the matching results difficult to meet the needs of the factory.

In addition, on the one hand, there are differences in the understanding of those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making the present invention, but due to space limitations, all details and contents are not listed in detail. However, this is by no means The present invention does not have these features of the prior art. On the contrary, the present invention already has all the features of the prior art, and the applicant reserves the right to add relevant prior art to the background art.

Contents of the invention

For the optimization of factory production line equipment, technicians usually design the expected performance parameters of the optimized equipment based on the equipment that needs to be replaced, and then search for products on the market based on the expected performance parameters. In the early days, factory craftsmen would read paper books such as product catalogs or product specifications sent by suppliers to find products that met the requirements. However, this method relies on the timely updating of paper books, but the inability to update in time is one of the disadvantages of paper books, which prevents craftsmen from conducting timely and extensive product searches. With the popularization of electronics and networking, existing technology has also begun to propose using the Internet to provide a product query platform to facilitate craftsmen to query the latest and most suitable products. The prior art CN112100965A discloses an electronic manufacturing collaborative innovation platform and its usage method. This prior art is to conduct 3D modeling of customer needs to facilitate suppliers to clearly understand customer needs from a three-dimensional perspective. , to facilitate suppliers to provide products that are more in line with customer needs. However, this existing technology lacks technical design to determine whether customer needs are unreservedly accepted and reflected in the 3D modeling and supplier product comparison processes. Some customers have customized needs beyond conventional design methods. These needs are initially The program presets that do not exist in the intermediate system do not exist in the supplier's factory inspection standards. In this case, how to match and the problems that need to be solved during digital twin simulation.

In view of the shortcomings of the existing technology, the present invention provides an end-to-end production line equipment optimization system based on digital twins, including: the factory side, which is used by the factory, the supplier side, which is used by the suppliers that provide products to the factory, The intermediate platform serves as an intermediate data interactive processing platform and is connected to factory-side and supplier-side data respectively. Based on the data uploaded by the factory side, a digital twin of the factory is established on the intermediate platform based on the needs of the factory side in the digital twin factory. The optimized target device and/or target area, the intermediate platform generates a query program about the target device and/or all devices in the target area, and after obtaining the expected performance parameters for at least one target device from the factory side, the intermediate platform is based on The process association between the selected equipment automatically shares some of the expected performance parameters with the associated equipment. When the expected performance parameters are determined, the intermediate platform performs the matching operation of the product twin plug-in from the database.

The present invention pays attention to the situation that some factories have customized requirements when optimizing equipment. For example, according to conventional design methods, only conventional expected parameters need to be provided for the optimization and replacement of a certain piece of equipment. However, based on its own process considerations, the factory has put forward new and additional requirements for conventional equipment optimization, which are custom requirements (custom parameters). This requirement did not exist in the middle platform at the beginning (when the factory side uploaded the expected performance parameters to the middle platform, there was no window for input for this custom requirement, and there was no relevant preset key value/digital twin simulation algorithm in the system. The simulation algorithm for this custom requirement has not been prepared) nor does it exist in the supplier's field inspection standards (the supplier does not know the specific values related to the custom parameters). Based on this problem, this solution provides a solution for uploading custom requirements (custom parameters) from the factory to the intermediate platform, and based on the upload of custom requirements, according to the target area selected by the user that needs to be optimized, the system can be based on The process association of the equipment in the area automatically shares some of the expected performance parameters, especially those containing custom parameters, to the relevant equipment. As a result, the factory only needs to provide the expected performance parameters including custom parameters to one or a few devices, and the system can automatically fill in the expected performance parameters (including custom parameters) of the remaining devices, which can significantly reduce the labor on the factory side and also provide The supplier provides timely and complete updated parameter data to provide guarantee.

Based on the above, after the supplier receives an inquiry containing custom parameters, it can update the product twin plug-in it releases to the intermediate platform. Therefore, this solution realizes a factory production line based on the joint construction of the supplier and the factory. Equipment optimization plan. In other words, the matching mode of this solution is not of the "results are available as soon as the requirement is submitted" type, but a retrieval solution that can continuously search according to the factory's needs, and can even reversely increase the number of searchable objects based on the factory's custom needs. The simulation process is not able to perfectly simulate all the working performance of the equipment in the production line from the beginning. When some parameters are missing, even the realistic calculation program will only calculate and predict the equipment's performance according to conventional simulation methods. working performance. Take a pump as an example. A pump is a tool used to pump substances, especially fluid substances. If you want to simulate the working performance of a pump, the required material data are the pump's head, flow rate, medium properties, working environment, etc. These data can be prepared in advance. Provided by the factory or supplier. According to these data, normal simulation can also be carried out. However, if the production line of the factory is relatively special, for example, if the medium is magnetic or has variable fluid properties, although simulation results can be obtained using conventional methods, the simulation results will be different from the actual working conditions. Therefore, the present invention provides a customized parameter uploading solution to the factory. This not only enables suppliers to optimize the inspection project standards of their own products when they leave the factory, but also upgrades the simulation algorithm in digital twin simulation to enable calculations by adding conditions. The results are more convergent and better matched.

Preferably, the product twin plug-in is formed by a digital twin from the data uploaded by the supplier to the intermediate platform, and has at least one three-dimensional model.

Preferably, the expected performance parameters can include custom parameters filled in by the factory side independently, and the project names and values of the custom parameters can be input by the factory side independently.

Preferably, the intermediate platform allocates at least one digital twin factory copy to each matched product twin plug-in, replaces the target device in the digital twin copy with the product twin plug-in, and performs simulation on the replaced digital twin factory copy, Output the simulation results to the factory.

Preferably, when the items of the expected performance parameters do not correspond or do not completely correspond to the product twin plug-in, the intermediate platform selects the non-corresponding items to generate a query list and sends it to the supplier, so that the supplier can provide updates based on the query list detection data to the intermediate platform.

Preferably, after obtaining the updated detection data, the intermediate platform updates the simulation algorithm based on this, so that it can output simulation results to the factory that are more in line with its expected formation parameters.

Preferably, after receiving the expected performance parameters uploaded by the factory side, the intermediate platform continuously queries the matching product twin plug-in. After the supplier side updates or adds a new product twin plug-in, the updated or new product twin plug-in can be added. Database for query matching.

Preferably, when the factory side selects a target device or target area that needs to be optimized, the intermediate platform performs a first search based on all device types in the target device or target area to filter product twin plug-ins that meet the type.

Preferably, when replacing a product twin, the model is replaced as well as the process parameters of the equipment.

Preferably, the intermediate platform performs a second search based on the first search results according to preset rules. The preset rules can include similar usage conditions, similar working performance, and the process parameters of the product twin plug-in are greater than or equal to the expected performance parameters.

Description of drawings

Figure 1 is a schematic diagram of the system flow provided by the present invention;

Figure 2 is a schematic diagram of the input and output of the twin factory stored in the intermediate platform of the present invention;

Figure 3 is a schematic flow chart of the selection method of the present invention;

Figure 4 is a schematic diagram of data interaction at each end of the present invention;

Figure 5 is a schematic diagram of the composition of the intermediate platform of the present invention;

Figure 6 is a schematic diagram of the relationship between various platform modules of the present invention.

Detailed ways

Detailed description will be given below with reference to Figures 1 to 6.

The production line in the factory has a large number of component equipment, and each component equipment assumes certain responsibilities and implements certain functions in the production line. According to the degree of importance, some equipment plays a relatively important role in the production line, while some equipment plays a relatively low role. However, no matter what kind of equipment it is, when the equipment is aging, approaching the end of its service life, and needs to be replaced, factory technicians are required to conduct equipment optimization analysis to provide corresponding equipment optimization plans. According to the current equipment optimization analysis model of most factories, technicians first need to investigate the basic information of the equipment that needs to be optimized, such as the current working parameters of the equipment; secondly, technicians need to design the optimized equipment based on the goals expected to be achieved by the optimized equipment. It has the expected performance parameters; then the technicians find suitable product equipment based on the expected performance parameters of the design. Since most of the optimized replacement of equipment is not repair or modification based on the original equipment, but directly replaces other models of existing products, another supplier is introduced as the source of these existing products. The supplier manufactures product equipment that can be upgraded and replaced for the production line, and the factory inquires and purchases suitable product equipment from the supplier. This is an end-to-end transaction. Suppliers generally provide product catalogs and product parameters by distributing product specifications to factories. Factory technicians check product specifications based on the expected performance parameters of the design and select the most suitable product for optimized replacement.

Using product specifications to query equipment is an inefficient way, so there are existing technologies that establish networked data connections between suppliers and factories. For example, CN112100965A discloses a collaborative innovation platform for electronic manufacturing and its use method. In this technical solution, designers design the products required for simulation through the design tool cloud and IP service cloud of the client. The design files are encrypted by the collaborative innovation platform and then transmitted. to the backend management end; the intelligent identification module receives and decrypts the encrypted design file, automatically matches each supplier's equipment that conforms to the product process and process parameters through the deep mapping algorithm based on the decrypted design file, and generates corresponding encrypted orders and sends them to the supplier; supply After decryption by the merchant, based on the product process flow, process parameters and corresponding supplier equipment recorded in the order, the production scene, supplier equipment and production process are simulated through the digital twin cloud, and a 3D panoramic display platform is generated for online visualization. This technical solution uses the backend management platform to establish a visual model that matches the client's needs, thereby providing intuitive feedback on the client's expected effects so that adaptive adjustments can be made based on the model results. However, the above-mentioned existing technology is limited to the client's demand adjustment and cannot reflect the supplier's adjustment needs.

On the one hand, manufacturers will add some parameter items that they are concerned about when designing expected performance parameters for optimizing equipment. However, these parameter items are not necessarily in the product data list provided by the supplier. In fact, the supplier may not have included them during design or delivery. The detection noticed the corresponding parameters and did not measure them, which led to certain difficulties in the work of "finding existing equipment products based on the expected performance parameters designed by the factory." For example, it was difficult for the system to match equipment products with these additional added parameters. Or the system ignores these additional parameters, making the matching results difficult to meet the needs of the factory. Based on this, the present invention provides an end-to-end equipment optimization system based on digital twins. According to Figures 1 and 4, the system is used to establish the supply and demand relationship between suppliers and factories. The system has a platform for receiving data uploaded by both suppliers and factories, which can be called an intermediate platform. The intermediate platform should be composed of servers or server groups with considerable computing power, and have databases that can store large amounts of data. The computing power of the intermediate platform is mainly used to match appropriate equipment products according to the factory's requirements, and to establish a digital twin factory based on the data uploaded by the factory. Specifically, a factory end is allocated specifically for factory use. The port can be a computer, an intelligent computing device, etc. The factory can input information to and obtain information from the factory end. The factory uploads the data used to create digital twins to the intermediate platform through the factory. The data used to create digital twins can be scanned and collected in the factory by a collection device in advance, such as using laser point cloud scanning equipment. After the digital twin factory is established on the intermediate platform, the information uploaded by the factory can be continuously received to update the data in the digital twin factory. In another embodiment, the intermediate platform can open an interface for operating the digital twin factory to the factory. The factory can share the operation and maintenance data in its own factory to the digital twin factory, so that the digital twin factory can compare the simulated factory in real time. Various process statuses, such as real-time material flow data, chemical reaction data, etc. As shown in Figure 5, the middle platform is configured from the device level with interfaces (which can include various interfaces such as data interfaces and graphical interfaces), computers (or CPUs) and servers (or storage databases). The interfaces are used to communicate with other The terminal forms a communication connection and is the main carrier for transmitting data. By establishing a data interaction platform that is trusted by both the factory and the supplier, this solution can facilitate the automatic matching and optimization of equipment and the transaction process without the private information of the factory or supplier being disclosed to the other party, which is secure. A computer, or a processing component (such as a CPU) with data processing capabilities, is mounted in an intermediate platform to perform a large number of data computing and processing tasks. The intermediate platform provided by this solution is responsible for the calculation and processing of the largest amount of data. The construction and storage of digital twin factories can be completed in the cloud, which can significantly reduce the equipment construction requirements on the factory side. In another embodiment, the factory side can also be responsible for the calculation and construction of the digital twin factory. The server is used to store a large amount of data. Since this platform can be opened to multiple factories (corresponding to multiple users) and multiple suppliers (corresponding to multiple suppliers), there are many digital twin factories and product twin plug-ins. The amount of data will be huge, so the data is stored in the form of a server or multiple server groups to form a storage database. The database will be made available to computers to process the data therein.

When the equipment needs to be optimized and upgraded, the process personnel use the factory side to select the equipment that needs to be optimized and upgraded in the digital twin factory, which can be called the target equipment. At this time, the digital twin factory can provide the parameter data related to the target equipment to the process personnel. . Relevant parameter data can be the basic configuration parameters of the equipment that have been stored when building the digital twin factory, such as dimensional data, weight, etc., or it can also be the current and/or historical working parameters of the equipment, such as flow rate, real-time power, etc. . Process personnel refer to the above information and expected process goals to design expected performance parameters and upload them to the intermediate platform. The expected performance parameters can include the project name and the expected value. The project name can be completely customized by the craftsman. This plan pays attention to that when optimizing the equipment, some factories will pay attention to some parameters that are not part of the regular parameter items of the equipment. These parameters can be called automatic parameters. Define parameters. Taking the pump as an example, the conventional specification sheet items only include the size, head, flow rate, etc. of the pump. However, some factories want to pay attention to the viscosity characteristics of the pump tube, but these are not included in the factory inspection items of pump products, nor are they provided for upload to the system. List. Therefore, this system allows users to completely customize parameter item names and values. In a preferred embodiment, the system is equipped with a correlation model that can automatically correlate parameter names with similar semantics, such as flow rate and flux. This enables user-defined parameter items to be matched using fuzzy retrieval, expanding the search surface.

The process of establishing a digital twin factory can be roughly divided into data collection, data processing and modeling and simulation. Data collection is through on-site scanning and collection of the physical equipment, structures, and pipelines of the factory. Data collection usually uses some special sensing equipment, such as using three-dimensional laser point cloud collection equipment. By determining the original reference point, the data is collected into the factory. The real object emits laser and collects various reflection points of the real object by collecting laser points. Data processing is to perform noise reduction, deduplication, repair, etc. on the data so that the data can be used in subsequent modeling steps. The process can also assist in a certain data analysis process, that is, to determine the actual location of multiple physical scan data in the actual factory. The mutual relationships among them, such as pipeline connection relationships, etc. The modeling and simulation process refers to importing processed data into a mathematical modeling program, and building a virtual factory in the virtual world that is a one-to-one replica of the real factory by establishing the physical and chemical relationship between the virtual and real worlds. The above process is the general process of establishing a digital twin factory.

As shown in Figure 2, when selecting equipment, you need to first understand the working parameters of the current equipment in the factory, and determine the expected parameters of a replacement equipment based on various factors such as expected production goals. Taking a pump as an example, when selecting a model, you need to pay attention to the technical requirements data, design result data, rule settings and other data of the existing pump, such as type, design flow, head, motor power, etc., and finally output the relevant information of the pump. Data is stored in a structured database, including but not limited to the following data:

1. Technical requirements document: including the working environment of the pump, such as medium and properties (viscosity, density, temperature, solid particle content of the liquid, etc.); energy supply (power of the pump and power supply type, such as 5KW, AC, and voltage of the power supply) and frequency requirements); design flow, head; temperature, humidity, altitude, explosion-proof requirements;

2. Pump selection calculation form: including pump model, flow, head, power and other parameters calculated according to technical requirements.

3. Pump performance curve: The performance curve drawn based on the pump’s flow rate, lift and other parameters is used to evaluate the pump’s working performance.

The data at the above points can be used as expected parameters, uploaded to the intermediate platform through the factory, and stored in the database.

The intermediate platform then associates it with the digital twin factory through master data (such as equipment tags) based on the expected parameter information structured and stored in the database.

Upload the expected performance parameters from the factory to the intermediate platform, and the intermediate platform performs the following retrieval process: perform a first retrieval based on the target device type; perform a second retrieval within the first retrieval results based on the expected performance parameters; retrieve and retrieval results from the database Related product twin plug-ins, assign a digital twin factory copy to each twin plug-in; replace the target device with a product twin plug-in in each digital twin factory copy; perform simulation for each digital twin factory copy to obtain simulation results ;Based on the simulation results of each digital twin factory copy, filter or sort and output to the factory.

The product twin plug-in is a digital twin model of a certain product device, which is constructed from the product data uploaded to the intermediate platform by the supplier. When suppliers launch new products, they upload the relevant parameters of the product, especially the three-dimensional model data, to the intermediate platform. The intermediate platform creates a digital twin model based on the digital twin algorithm. Product twin plug-ins can be formed according to the smallest equipment unit in the factory production line. The smallest equipment unit may refer to the lowest level equipment in the production line that can be replaced individually, such as a valve part, a base part, etc. Product twin plug-ins can be provided by multiple suppliers of multiple product types. As the number of participating suppliers increases, the number and variety of product twin plug-ins stored in the system's database will also increase.

The specific operation of performing the first screening above is: based on the input target device type, the intermediate platform retrieves only the digital twin module data set that matches the target device type from the database. Specifically, the target device type information can be included in the input expected performance parameters, or it can be automatically generated by the system after the target device is selected in the early stage. The target device type is a query key value information. Through key value query, the corresponding product twin plug-in data set can be filtered out from the many product twin plug-ins that have been stored in the database.

The specific operation of the intermediate platform for secondary retrieval is to retrieve product twin plug-ins that meet the rules from the results of the first retrieval according to the preset rules. Preset rules can be designed manually, especially by craftsmen at the factory. The rules that can be selected include but are not limited to: similar usage conditions, product twin plug-in parameter values greater than or equal to expected performance parameters, similar working performance, and appropriate cost. wait. You can select one of the above rules according to specific needs, or select multiple rules for combined retrieval. Taking a pump as an example, the adoption rules may include: similar usage conditions; the design flow rate of the digital twin module is greater than the design flow rate of the current equipment to be replaced; the head of the digital twin module is greater than the head of the current equipment to be replaced; the equipment power of the digital twin module is currently Multiples of the specified number of devices to be replaced, the working performance is basically the same.

After a second search, the results obtained were close to the list of replaceable products expected by the factory. However, it is unclear to both the supplier and the factory whether the specific performance of the product after it is actually installed on the production line will be as expected. The problem in the existing technology is that for some special equipment products or for some special production lines, the conventional selection and matching system will also recommend products whose nominal parameters meet the expected performance parameters proposed by the factory. However, after purchasing products and equipment as recommended and replacing them, problems may still arise in which the actual performance of the equipment does not meet expectations. On the one hand, the reason is that the expected performance parameters provided by the factory do not match the actual situation of the factory, or there are certain differences between the product twin plug-in produced based on the data provided by the supplier and the actual product. Therefore, after the second search, the present invention assigns a digital twin factory copy to each product twin plug-in according to the results, replaces the corresponding target device in each digital twin factory copy with the product twin plug-in, and performs simulation, thereby Obtain simulation results. Preferably, the digital twin factory copy may not be copied for the entire factory, but only for the process area where the target equipment is located. The process area refers to the part associated with the target equipment in the production line, such as the parts that belong to the same process link. Taking the target equipment as a pump as an example, the process area can include upstream tanks, downstream tanks, and intermediate pipelines. As well as the pump itself. The selection of process areas can be manually specified or automatically derived by artificial intelligence based on pre-analysis of the correlation of each production line in the digital twin factory model.

Detailed rules for secondary retrieval can be formulated by the factory. For example, the intermediate platform can open a rule formulation window to the factory, and the factory can choose, for example, economic priority (lower-priced products are ranked first), quality priority (can fully cover expectations). Products with performance parameters and larger differences are placed first), and more complex rules can also be formulated.

In the simulation of the digital twin factory copy, the product twin plug-in is replaced with the selected target equipment (or multiple corresponding product twin plug-ins are replaced with the corresponding equipment in the selected target area). The replacement includes the replacement of the model and the replacement of process parameters. Based on the replaced digital twin factory copy, simulation calculations are performed according to the process operation logic. Process operation logic refers to a set of pre-stored algorithms that can simulate the changing state of materials in the digital twin factory production line from basic scientific levels such as physics and chemistry. Taking fluid passing through a pipeline as an example, the algorithm will include a fluid flow model. , pipeline constraint model, pump actuation dynamic model, etc. The process operation logic algorithm can be pre-made by humans and computers based on the specific characteristics of the factory's production line. Combined with a large amount of prior basic physical, chemical and biological knowledge, a relatively complete process operation logic algorithm can be obtained. After replacing the product twin plug-in, it is equivalent to changing some calculation parameters in the process operation logic. Based on the update of calculation parameters, new simulation results can be output. The simulation results are displayed to the users at the factory as one of the outputs of this system. Factory personnel can decide which product to purchase based on the simulation results.

DCS real-time monitoring system can be used to associate plug-ins with digital twin factories. The system can compare expected performance data with simulation parameters of real-time simulation to form early warning results.

Preferably, the intermediate platform or factory side may have a material management system. Material management systems are used to manage equipment already in the factory, such as inventory equipment. In the case where the factory side can select, the intermediate platform will recommend to the factory side based on the simulation results or the intermediate platform will give priority to the equipment that has been stored in the material management system during the first and second retrieval, so as to optimize the use of inventory and reduce the cost. cost.

As shown in Figure 3, the present invention also provides a supporting end-to-end production line equipment optimization method based on digital twins. The method may include the following: establishing a digital twin factory that is a digital twin of the physical factory; retrieving product twin plug-ins that match the selected equipment (or equipment within the selected area) based on the formed expected performance parameters; The plug-in allocates a copy of the digital twin factory and performs a simulation based on the copy, in which the product twin plug-in replaces the selected equipment; the simulation results are output to factory personnel for selection.

Preferably, based on the above, this solution found that in the field of optimization solutions for using network platforms to query equipment, although it can more conveniently enable factory technicians to access products provided by more suppliers, it can also achieve automation to a certain extent. Search for products and equipment that may meet the requirements. However, whether it is from the perspective of whether the product can actually meet the production needs after installation, or from the perspective of the factory's own special needs, this plan believes that the customized needs of the factory (customer) also need to be paid attention to, that is, the supplier's products do not have to be installed before leaving the factory. Parameters tested. Therefore, when the expected performance parameters uploaded by the factory include custom parameters, this system first matches the relevant parameter key values based on semantic analysis. If there is a match, it will search in the database according to the key value. If there is no match, Then a query list is formed and sent to the supplier. The supplier here may be a supplier that has been screened, for example, a supplier that meets the target device type after the above first query. After receiving the inquiry list, the supplier can specifically test the parameter items recorded on the inquiry list to obtain answers to customized requirements. Subsequently, the new data obtained from the test is uploaded by the supplier to the intermediate platform to form a new product twin plug-in or update the original product twin plug-in, so that the product twin plug-in can be inserted into the ongoing optimization simulation for the target device Simulate the selection process. The above-mentioned optimization and selection process is continuously carried out on the intermediate platform. In other words, this search is not the type of "results are available as soon as the requirements are submitted", but can be continuously searched according to the factory's needs, and can even be reversely increased based on the factory's custom needs. The object's retrieval scheme. The simulation process is not able to perfectly simulate all the working performance of the equipment in the production line from the beginning. When some parameters are missing, even the realistic calculation program will only calculate and predict the equipment's performance according to conventional simulation methods. working performance. Take a pump as an example. A pump is a tool used to pump substances, especially fluid substances. If you want to simulate the working performance of a pump, the material data required are the pump's head, flow rate, medium properties, working environment, etc. These data can be prepared in advance. Provided by the factory or supplier. According to these data, normal simulation can also be carried out. However, if the production line of the factory is relatively special, for example, if the medium is magnetic or has variable fluid properties, although simulation results can be obtained using conventional methods, the simulation results will be different from the actual working conditions. Therefore, the present invention provides a customized parameter uploading solution to the factory. This not only enables suppliers to optimize the inspection project standards of their own products when they leave the factory, but also upgrades the simulation algorithm in digital twin simulation to enable calculations by adding conditions. The results are more convergent and better matched. Preferably, this system can also record custom parameters, corresponding target equipment types, and data fed back by the supplier after receiving the inquiry list, and use it as a system template for the replacement of other similar equipment. When other factory users upload similar requirements (preferably, the upstream and downstream data link integration functions of the digital twin factory can be used to make independent guesses), it is automatically recommended to the factory whether to add relevant custom parameters, so that the factory has the power to make design decisions. A better place to start. Preferably, products from suppliers with higher feedback query list speed and quality (evaluable by the factory) can be displayed with priority when the recommendation list is finally formed to the factory. This can encourage more suppliers to proactively understand factory needs, develop and provide more types of products that better meet needs, and promote product updates and iterations.

Preferably, this system can also provide optimized equipment query and selection for a selected part of the production line. Through the factory end, the target area of the production line is defined in the digital twin factory model. The intermediate platform automatically searches for all equipment in the target area, and the user uploads the expected performance parameters to the intermediate platform. The expected performance parameters are different for each device in the target area, and only the expected performance parameters of some devices may be given. Based on the process association between equipment in the target area in the digital twin factory, the intermediate platform automatically shares some of the expected performance parameters (especially custom parameters) to all associated equipment in the target area. And match the appropriate product twin plug-in for each device according to the expected performance parameters after the update. Process correlation can be the relationship between several devices on the production line. The most common relationship is the upstream and downstream relationships. Basically all equipment has an upstream and a downstream. Multiple devices can be associated through the upstream and downstream relationships. For example, tank A is connected to tank B through a pipeline, which is an upstream and downstream relationship, including three components: tank A, the pipeline, and tank B. Then they are also related to a certain extent in terms of parameter relationships. This plan believes that the expected performance parameters are also consistent among process-related equipment. For example, if a pump requires antimagnetic properties, then the storage tanks upstream and downstream of it have antimagnetic requirements. Therefore, this solution can automatically fill in the expected performance parameter requirements of the equipment of the associated production line based on correlation, thereby significantly reducing the labor cost of user input. More importantly, the special parameter requirements of each associated equipment can be updated to the supplier, so that the retrieval simulation results can be more organically and smoothly integrated within the factory's designated production line section, significantly reducing the probability of problems after optimization and replacement.

As shown in Figure 6, preferably, the factory end provided by this solution can obtain data in the factory's existing data asset management library in the form of accessing the factory database. At present, many factories have their own data management centers. A large amount of production data, construction data, operation and maintenance data, etc. in the factory will be summarized in the data management center, and some private content cannot be shared. This solution is based on the database connection protocol in the factory, configures data extraction and monitoring means, and only extracts data related to the equipment to be optimized. And basically, only the existing attribute data of the equipment is extracted, without involving the data collected during the production process ( That is, production data) or operation and maintenance data. Under optional circumstances, factory personnel can have open access to the above production data or operation and maintenance data. This solution is quite secure. In addition, the intermediate platform of this solution will display the independent interface of the digital twin (including the digital twin factory and product twin module) (that is, an independent graphical interface), and establish private connections with the corresponding factory and supplier sides respectively. The ordinary data interface can transmit the result data of whether the simulation matches, or the three-dimensional model data with low precision (hiding or blurring some model parameters or model shape as needed). On the one hand, this can reduce the congestion caused by data parallelism. On the other hand, it can avoid irrelevant personnel from being exposed to the fine model, which will lead to the leakage of important information (for example, the data of the data interface can be shared to all ports using the intermediate platform to facilitate personnel understanding and selection. , and plagiarists cannot accurately know the exact model of the product twin module to copy the product, thus protecting the interests of the supplier).

Preferably, the intermediate platform can be built in the form of pure software on the factory side and/or the supplier side, or it can be built on the cloud.

It should be noted that the above specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the scope of the present invention. within the scope of protection of the invention. Those skilled in the art should understand that the description of the present invention and the accompanying drawings are illustrative and do not constitute limitations on the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The description of the present invention contains multiple inventive concepts, such as "preferably", "according to a preferred embodiment" or "optionally" means that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to propose divisions based on each inventive concept. The right to apply. Throughout the text, the features introduced by "preferably" are only optional and should not be understood as mandatory settings. Therefore, the applicant reserves the right to waive or delete the relevant preferred features at any time.

Claims (10)

1. An end-to-end production line equipment optimization system based on digital twins, including: Factory side, which is used by the factory, Supplier side, which is used by suppliers who supply products to the factory, The intermediate platform serves as an intermediate data interactive processing platform and is connected to factory-side and supplier-side data respectively. It is characterized by, Based on the data uploaded by the factory, a digital twin of the factory is established on the intermediate platform, Based on the target equipment and/or target area that need to be optimized defined by the factory side in the digital twin factory, the intermediate platform generates a query program about the target equipment and/or all equipment in the target area. After the factory side obtains the expected performance parameters for at least one target device, the intermediate platform automatically shares some of the expected performance parameters to the associated devices based on the process association between the selected devices, Having determined the expected performance parameters, the intermediary platform performs a matching operation of the product twin plug-in from the database. 2. The system according to claim 1, wherein the product twin plug-in is formed by a digital twin from the data uploaded by the supplier to the intermediate platform, and has at least one three-dimensional model. 3. The system according to claim 1, wherein the expected performance parameters can include custom parameters filled in independently by the factory, and the project names and values of the custom parameters can be input independently by the factory. 4. The system according to claim 1, characterized in that the intermediate platform allocates at least one digital twin factory copy to each matched product twin plug-in, and replaces the target device in the digital twin copy with the product twin plug-in. , perform simulation on the replaced digital twin factory copy, and output the simulation results to the factory. 5. The system according to claim 1, characterized in that, when the items of the expected performance parameters do not correspond or do not completely correspond to the product twin plug-in, the intermediate platform selects the items that do not correspond to generate an inquiry list and sends it to The supplier side enables the supplier side to provide updated detection data to the intermediate platform based on the query list. 6. The system according to claim 5, characterized in that, after obtaining updated detection data, the intermediate platform updates the simulation algorithm based on this, so that it can output to the factory a simulation that is more consistent with its expected formation parameters. Simulation results. 7. The system according to claim 1, characterized in that, after receiving the expected performance parameters uploaded by the factory side, the intermediate platform continuously queries the matching product twin plug-in, and updates or adds a new product twin plug-in at the supplier side. After plug-in, the updated or new product twin plug-in can be added to the database for query matching. 8. The system according to claim 1, characterized in that when the factory side selects the target equipment or target area that needs to be optimized, the intermediate platform performs a first search based on all equipment types in the target equipment or target area to filter. Product twin plugin that conforms to the type. 9. The system according to claim 4, characterized in that when replacing the product twin plug-in, the model and the process parameters of the equipment are replaced. 10. The system according to claim 8, characterized in that the intermediate platform performs a second search according to preset rules for the first search results, and the preset rules can include similar usage conditions, similar working performance, and product twin plug-ins. The process parameters are greater than or equal to the expected performance parameters.