CN117931888A - Digital twin data acquisition method and device and computer equipment - Google Patents

Abstract

The application relates to a digital twin data acquisition method, a digital twin data acquisition device and computer equipment. The method comprises the following steps: preprocessing and screening the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data corresponding to the characteristic engineering data; directionally screening the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine; and carrying out stream computing processing on the target data to obtain model data corresponding to the robot digital twin model. The method can realize degradation of the data volume required by constructing the digital twin model, and improve the stability of the whole data, so that the modeling of the digital twin model of the robot can be realized on industrial application.

Description

Digital twin data acquisition method and device and computer equipment

Technical Field

The present application relates to the field of automation technologies, and in particular, to a digital twin data acquisition method, apparatus, and computer device.

Background

With the development of automation technology, the trend of replacing manual work by robots is more obvious, and further, the robots are required to be controlled and managed based on the robot digital twin body model construction technology.

In the prior art, the data collection and analysis research work related to robots is relatively less. Because the related products of the robot are less in floor, the data characteristic engineering of the robot is difficult to construct, the data acquisition forms are different due to different specific businesses, and the data uploading operation is carried out after the data acquisition. According to the normal automobile production line (more than 200), the data generation amount of the robot is approximately T-level every day. If the data is directly collected and uploaded, huge uploading data volume is generated, and maintaining the huge uploading data volume requires a large amount of hardware resources and stable network bandwidth, which is different from the network environment of the actual industrial site. The modeling mode and the access mode of the digital twin system are more and less, mainly because the motion mode of the robot is difficult to predict and the sampling data points are too many, so that if twin modeling is realized, a large amount of data IO (Input/Output) is consumed in transmission and calculation, and the landing on industrial application is difficult to realize.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a digital twin data acquisition method, apparatus, computer device, computer readable storage medium, and computer program product that enable data volume degradation.

In a first aspect, the present application provides a digital twin data acquisition method, comprising:

Preprocessing and screening the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data of the corresponding characteristic engineering data;

directionally screening the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine;

And carrying out stream computing processing on the target data to obtain model data corresponding to the robot digital twin model.

In one embodiment, performing a streaming computation on the target data to obtain model data corresponding to a robot digital twin model, where the model data includes:

determining a data source analysis rule of the target data based on the data source input of the target data;

and analyzing the target data into extreme point data of a corresponding type according to the data source analysis rule to obtain model data.

In one embodiment, the method includes directionally screening the original time sequence data by a rule engine to obtain target data meeting the screening condition of the rule engine, including:

And directionally screening the original time sequence data by using the sphere parameter of the rule engine to obtain target data meeting the screening condition of the sphere parameter.

In a second aspect, the present application also provides a digital twin data acquisition device, including:

The preprocessing module is used for preprocessing and screening the robot data based on the data types of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data of the corresponding characteristic engineering data;

the directional screening module is used for carrying out directional screening on the original time sequence data through the rule engine to obtain target data meeting the screening conditions of the rule engine;

And the streaming processing module is used for carrying out streaming calculation processing on the target data to obtain model data corresponding to the digital twin model of the robot.

In a third aspect, the present application also provides a digital twin construction system comprising: edge servers and cloud center servers;

The edge server is used for executing the method so as to transmit the model data to the cloud center server;

The cloud center server is used for constructing a digital twin model corresponding to the robot based on the model data.

In one embodiment, an edge server is configured with:

the MQTT is used for receiving robot data;

the edge data ETL tool ekuiper is connected with the MQTT and is used for processing the robot data to obtain model data;

The cloud center server is configured with kafka for receiving the model data transmitted by the edge data ETL tool ekuiper and transmitting to the digital twin construction module of the cloud center server.

In one embodiment, a digital twinning construction module includes:

Model database, dispatcher and robot model library;

the model database is used for receiving and storing the model data transmitted by the kafka;

The dispatcher is used for constructing a robot model library according to the data in the model database, and the robot model library stores digital twin models corresponding to the robots.

In a fourth aspect, the present application also provides a computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.

In a fifth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method described above.

In a sixth aspect, the application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described above.

According to the digital twin data acquisition method, the digital twin data acquisition device and the computer equipment, the robot data are preprocessed and screened based on the data types of the characteristic engineering data required by constructing the robot digital twin model, so that the original time sequence data corresponding to the characteristic engineering data are obtained; directionally screening the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine; and carrying out streaming calculation processing on the target data to obtain model data corresponding to the digital twin model of the robot, so that degradation of the data quantity required by constructing the digital twin model is realized, and the stability of the whole data is improved, and the modeling of the digital twin model of the robot can be realized in industrial application.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related art, the drawings that are required to be used in the embodiments or the related technical descriptions will be briefly described, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for those skilled in the art.

FIG. 1 is a flow diagram of a method of digital twin data acquisition in one embodiment;

FIG. 2 is a schematic diagram of the composition of robot data in one embodiment;

FIG. 3 is a flow diagram of a process for streaming target data in one embodiment;

FIG. 4 is a schematic diagram of an edge cloud architecture in one embodiment;

FIG. 5 is a schematic diagram of a digital twin data composition of a robot in one embodiment;

FIG. 6 is a schematic diagram of the composition of an AML file;

FIG. 7 is a flow diagram of behavior modeling and motion modeling based on raw data in one embodiment;

FIG. 8 is a block diagram of a digital twin data acquisition device in one embodiment;

fig. 9 is an internal structural diagram of a computer device in one embodiment.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.

It will be appreciated that terms such as "first," "second," and the like, are used herein merely to distinguish between similar objects and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated.

It is to be understood that in the following embodiments, "connected" is understood to mean "electrically connected", "communicatively connected", etc., if the connected circuits, modules, units, etc., have electrical or data transfer between them.

It is understood that "at least one" means one or more and "a plurality" means two or more.

As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," and/or the like, specify the presence of stated features, integers, steps, operations, elements, components, or groups thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or groups thereof. Also, the term "and/or" as used in this specification includes any and all combinations of the associated listed items.

The related concepts in the present application are explained in terms of the following:

AML: automation ML (Automation Markup Language) is an XML-based data interchange format for factory engineering data. AML is primarily intended to support data exchange between heterogeneous engineering devices, the goal of which is to exchange data interconnections in different fields such as mechanical engineering, electrical design, machining engineering, process control engineering, HMI, PLC programming, robot programming, etc. It can be applied to all industrial fields requiring data exchange, such as discrete industries or process industries.

BOM: BOM (Bill of Material) bill of materials, that is, a file describing the structure of a product in a data format, is a product structure data file that can be identified by a computer, and is also a dominant file of ERP. BOM makes the system recognize the product structure and is also the tie for connecting and communicating various business of enterprises. The types of BOM in ERP systems mainly include 5 types: the system comprises a condensed BOM, a summarized BOM, a back check BOM, a cost BOM and a planning BOM.

Profile database (Neo 4 j): neo is a network-oriented database-that is, it is an embedded disk-based Java persistence engine with full transactional properties, but it stores structured data on the network rather than in tables. The network (mathematically called a graph) is a flexible data structure that can be used in a more agile and fast development mode. He supports the data expression of the pattern and the construction of the derivatization scheme.

Digital twinning: the digital twin is to fully utilize data such as a physical model, sensor update, operation history and the like, integrate simulation processes of multiple disciplines, multiple physical quantities, multiple scales and multiple probabilities, and complete mapping in a virtual space, thereby reflecting the full life cycle process of corresponding entity equipment. Digital twinning is a universally adapted theoretical technology system, can be applied in a plurality of fields, and has more application in the fields of product design, product manufacturing, medical analysis, engineering construction and the like. The most deep application in China is in the engineering construction field, the highest attention is paid, and the hottest research is in the intelligent manufacturing field.

Zeebe-Zeebe is an open source workflow engine for executing and coordinating distributed workflows. It is a distributed workflow engine introduced by Camunda, which aims to handle large-scale, highly concurrent workflow applications. Zeebe are designed to support scalability, high performance, and fault tolerance. The system is based on an event-driven architecture, adopts a distributed deployment mode, and allows workflow instances to be distributed to a plurality of nodes for execution, so that horizontal expansion and load balancing are realized.

EdgeX Foundry (edgex): edgeX Foundry is a hardware and operating system independent open-source neutral edge computing micro-service framework of a Linux foundation, and is used for unifying an ecosystem of an industrial internet of things edge computing solution.

Edge calculation: edge computing refers to providing near-end services by adopting an open platform with integrated network, computing, storage and application core capabilities on the side close to the object or data source. The application program is initiated at the edge side, and faster network service response is generated, so that the basic requirements of the industry in the aspects of real-time service, application intelligence, security, privacy protection and the like are met. Edge computation is between a physical entity and an industrial connection, or at the top of a physical entity. The cloud computing can still access the historical data of the edge computing.

Cloud computing: cloud computing (clouding) is one type of distributed computing, which refers to decomposing a huge data computing process program into numerous small programs through a network "cloud", and then processing and analyzing the small programs through a system composed of multiple servers to obtain results and returning the results to users. Early cloud computing, simply referred to as simple distributed computing, solves task distribution, and performs merging of computing results. Thus, cloud computing is also known as grid computing. By this technique, processing of tens of thousands of data can be completed in a short time (several seconds), thereby achieving a powerful network service.

Bian Yun cooperate: bian Yun collaboration is that edge computing most deployments and application scenarios require edge side collaboration with the central cloud. Including resource collaboration, application collaboration, data collaboration, intelligent collaboration, and the like.

Message Queue (MQ): is a data structure that is "first in first out" in the underlying data structure. The method is generally used for solving the problems of application decoupling, asynchronous message, flow peak clipping and the like, and realizing a high-performance, high-availability, scalable and final consistency architecture.

MQTT (Message Queuing Telemetry Transport): the MQTT protocol, collectively referred to as Message Queuing Telemetry Transport, translates to message queue transport probing, which is a publish-subscribe mode based message protocol under the ISO standard, is based on the TCP/IP protocol suite, and is designed to improve the performance of the network device hardware and the performance of the network. MQTT is commonly used on IoT, internet of things, and is widely used in industrial-level application scenarios such as automotive, manufacturing, oil, gas, etc.

Kafka: kafka is an open source stream processing platform developed by the Apache software foundation, written by Scala and Java. Kafka is a high-throughput distributed publish-subscribe messaging system that can handle all action flow data for consumers in a web site. Such actions (web browsing, searching and other user actions) are a key factor in many social functions on modern networks. These data are typically addressed by processing logs and log aggregations due to throughput requirements. This is a viable solution for log data and offline analysis systems like Hadoop, but with the limitation of requiring real-time processing. The purpose of Kafka is to unify on-line and off-line message processing through the Hadoop parallel loading mechanism, and also to provide real-time messages through the clusters.

OLAP (for online analytical processing) is software for high-speed multidimensional analysis of large amounts of data from data warehouses, data marts, or some other unified centralized data store.

Clickhouse: clickHouse are collectively CLICK STREAM, data WareHouse. It is a russian Yandex open-source columnar store database for online analytical processing query (OLAP) MPP architecture that can generate analytical data reports in real-time using SQL queries.

ETL: the Extract-Transform-Load is used to describe the process of extracting (Extract), converting (Transform), and loading (Load) data from the source to the destination. The term ETL is more commonly used in data warehouses, but its objects are not limited to data warehouses.

Stream processing: streaming processing (sometimes referred to as event processing) may be described simply as continuous processing of unbounded data or events. A stream or event handling application may be more or less described as a directed graph and is typically described as a Directed Acyclic Graph (DAG). In such a graph, each edge represents a stream of data or events, each vertex represents an operator, and data or events from adjacent edges are processed using logic defined in the program. There are two special types of vertices, commonly referred to as sources and sinks. sources read external data/events into the application, and sinks typically gathers results generated by the application.

DAG: directed acyclic graph DIRECTED ACYCLIC GRAPH (DAG), a finite directed graph with no directed loops. It consists of a finite number of vertices and directed edges, each directed edge pointing from one vertex to the other; starting from any vertex, the original vertex cannot be returned by the directed edges. The directed acyclic graph starts from any point in a graph, and no matter how many bifurcation intersections are walked through, the directed acyclic graph has no possibility of returning to the point.

Edge flow processing engine (ekuiper): LF Edge eKuiper is lightweight internet of things edge analysis and stream processing open source software realized by Golang, and can be operated on various kinds of edge equipment with limited resources. eKuiper is to provide a streaming software framework (similar to APACHE FLINK (open new window)) at the edge. eKuiper a rules engine allows users to provide SQL-based or graphics-based (similar to Node-RED) rules, creating Internet of things edge analysis applications within a few minutes.

Influxdb: influxDB is an open source time-sequential database developed by InfluxData. It is written by Go and is focused on searching and storing time-series data with high performance. InfluxDB is widely used in the fields of monitoring data of storage systems, real-time data of the IoT industry, and the like.

Heterogeneous calculation (Heterogeneous Computing): heterogeneous computing technology is generated from the middle 80 s, and has become one of research hotspots in the parallel/distributed computing field because of the capability of economically and effectively acquiring high-performance computing power, good expandability, high computing resource utilization rate and huge development potential. With the rapid development of communication and network technologies, network computing concepts have evolved. Homogeneous network computing systems now or cow first emerge, and heterogeneous network computing systems soon emerge, making heterogeneous computing one of the main research hotspots in the parallel/distributed computing field in recent years.

Container (Docker): the Docker container is an open-source application container engine, so that the developer can package their applications and rely on packages to a portable container in a unified manner, then issue the packages to any server (including popular Linux machines and windows machines) provided with the Docker engine, and can also implement virtualization. The containers are completely using a sandbox mechanism without any interface to each other (an app like an iPhone). Almost no performance overhead, can be easily run in machines and data centers. Most importantly, they do not rely on any language, framework, including systems.

Container cluster management system (Kubernetes): kubernetes (commonly referred to as K8 s) is an open source container cluster management system from the Google cloud platform for automatically deploying, expanding and managing containerized (containerized) applications. The system builds a dispatch service for a container based on Docker.

Finite State Machine (FSM): finite state automata (FSM FINITE STATE MACHINE or FSA FINITE STATE automaton) is a computational model that is abstracted to study finite memory computational processes and certain language classes. Finite state automata possess a finite number of states, each of which can be migrated to zero or more states, and the input string determines which state to perform the migration. The finite state automaton may be represented as a directed graph. Finite state automata are the subject of investigation of automaton theory.

Dolphin scheduler: apache DolphinScheduler is a distributed, scalable, visual DAG workflow task scheduling open source system. The method is suitable for enterprise-level scenes, and provides a solution for visualizing operation tasks, workflows and full life cycle data processing processes. Apache DolphinScheduler are directed to resolving complex big data task dependencies and providing data and relationships in various OPS schemas for applications. The problem that the data research and development ETL depends on the complex and complicated state of health of the task can not be monitored is solved. DolphinScheduler assembles the task in a DAG (DIRECTED ACYCLIC GRAPH, DAG) streaming mode, can monitor the execution state of the task in time, and support the operations such as retrying, failure of the designated node to resume, suspending, resuming, and terminating the task.

In one embodiment, as shown in fig. 1, a method for constructing a digital twin body of a robot is provided, and the embodiment is applied to a terminal for illustration by the method, it is understood that the method can also be applied to a server, and can also be applied to a system comprising the terminal and the server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the steps of:

step 102, preprocessing and screening the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data of the corresponding characteristic engineering data.

The robot may be a robot for performing various tasks and operations in an automotive production line, and may be, for example, a welding robot, an assembly robot, a material handling robot, or the like. The robot data may be data such as working parameters, state parameters, motion parameters, and electrical parameters generated during the working process of the robot. As shown in fig. 2, in one particular embodiment, the robot data may be divided into internal data, which may be directly read by a controller of the robot itself, and external data, which may be acquired based on external sensor measurements. Specifically, the internal data includes robot start-stop data, robot axis angle, robot periodic energy consumption, robot axis data, robot alarm data, and other internal data; the robot axis data includes xyz angle of the robot axis, payload, what axis, current data, and voltage data. The external data includes vibration data, temperature data, current data, voltage data, power data, robot spatial axis data, and other external data; the robot spatial axis data includes xyz actual coordinates of the robot spatial axis, projection data, and multi-modal data.

In a practical industrial scenario, this data will be sampled very often, since the internal data sampling frequency of existing robots can go to within 10 ms. The method is equivalent to generating more than 100 pieces of data per second on average, and the data message type of a single robot can be split into five or more pieces, so that the data generated per second is equivalent to not less than 500 pieces.

Taking the data format of innodb engine type of Mysql as an example, the space occupied by every 1000 pieces of data with 54 varchar32 fields is about 0.6M, so that the data generated per second is about 0.3M-0.5M, the data amount generated per day is 26G-43G, and the data amount is very huge in view of cloud computing.

The external data sampling frequency of the robot is carried out according to the existing sensor, in order to ensure that the accuracy of data and models is not smaller than the sampling frequency of internal data in the application process, the data generated every day is equivalent to the internal data (26G-43G), and if projection data and multi-mode data, even picture data and video data are contained, the data amount is not less than 10 times of the existing internal data sampling (260G-430G).

In one embodiment, to construct a digital twin model of a robot, the feature engineering data needs to be synchronously collected with the data twin data. The characteristic engineering data of the robot refers to raw data collected from a sensor, an actuator and an environment of the robot, and in general, the characteristic engineering data can be used for related follow-up tasks such as machine learning or digital twin modeling after operations such as processing, conversion, selection and extraction. For different follow-up tasks, the characteristic engineering data required to be collected in the early stage are also different. Illustratively, with respect to the energy consumption business of the robot, the data types of the required characteristic engineering data include robot cycle energy consumption, robot axis data, current data, voltage data, power data, and xzy actual coordinates of the robot spatial axis. For the modeling task of the action model, the data types of the required feature engineering data include: robot axis angle, xyz angle, payload and what axis in the robot axis data, and xyz actual coordinates and projection data in the robot spatial axis data.

In order to reduce the data acquisition amount of the digital twin model of the robot, the embodiment of the application carries out preprocessing screening on the robot data based on the data type of the characteristic engineering data required by constructing the digital twin model of the robot to obtain the original time sequence data of the corresponding characteristic engineering data.

And 104, directionally screening the original time sequence data through the rule engine to obtain target data meeting the screening conditions of the rule engine.

A rule engine may refer, among other things, to a computer system or tool for managing and executing rule-based business logic. It can help organize and manage a large number of rules and evaluate and process input data based on these rules, thereby automating the decision process.

Illustratively, the original time sequence data can be directionally screened through the rule-based business logic set by the rule engine, so as to obtain target data meeting the screening condition of the rule engine. Through the directional screening of the rule engine, a large number of data writing which is inconsistent with the condition can be removed, so that the overall data output is greatly reduced.

And 106, performing streaming computing processing on the target data to obtain model data corresponding to the robot digital twin model.

Specifically, the target data is obtained by directional screening of the original time sequence data, so that the target data still consists of the time sequence data. And performing streaming calculation processing on the target data, namely performing streaming calculation on the time sequence data. By setting the rule of the streaming calculation, the real-time data output of the robot can be calculated in real time by the original data input of the robot, and the input of the real-time data is a large amount of time-series data, but the output data can be scaled by a reasonable rule.

Compared with all original time sequence data of all robot data types, the data volume of the data subjected to the flow calculation processing is greatly degraded, and the data can be further used for constructing a digital twin model of the robot in a subsequent task.

In the digital twin data acquisition method, preprocessing and screening are carried out on the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model, so as to obtain the original time sequence data corresponding to the characteristic engineering data; directionally screening the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine; and carrying out streaming computation on the target data to obtain model data corresponding to the digital twin model of the robot, so that the degradation of the data amount of the constructed digital twin model is realized, the stability of the whole data is improved, and the modeling of the digital twin model of the robot can be realized in industrial application.

In an exemplary embodiment, as shown in fig. 3, the streaming calculation process is performed on the target data to obtain model data corresponding to the robot digital twin model, which may include steps 302 to 304. Wherein:

Step 302, determining a data source parsing rule of the target data based on the data source input of the target data.

Specifically, the edge data ETL tool ekuiper may be used to stream the target data. ekuiper can accept the input of various protocols including OPCUA, MQTT and the like, then flow processing calculation is carried out in ekuiper, and the actual required data is calculated through the original time sequence data in the target data, namely model data corresponding to the digital twin model of the robot is constructed.

Wherein ekuiper receives the data source of the MQTT as the data source format of the MQTT, and all messages of the robot can be used as the data source format of the MQTT. Further, ekuiper may set different data source parsing rules (rules) according to different data source inputs (source).

And step 304, analyzing the target data into extreme point data of a corresponding type according to the data source analysis rule to obtain model data.

Specifically, ekuiper may be used to ultimately parse a data source (i.e., target data) into different types of data objects (sink) according to the parsing rules of the data source. Where sink means "receiver" or "destination" that refers to the final destination or storage location of the data stream. It is used to specify the final output location of the data stream and transfer the data from the process to the corresponding storage or target system.

For example, the data source parsing rule may be configured as an extreme point data calculation rule. Specifically, when one extreme point is calculated by one time series data, the data of the extreme point may be outputted as a result, and not all time series data are required.

Alternatively, using the ekuiper sliding window function, an extreme point can be considered to be reached if the maximum or minimum value is not the first and last value within the sliding window. Based on the sliding window function, extreme point data can be obtained through calculation through the original time sequence data and used as model data.

According to the digital twin data acquisition method provided by the application, the data source analysis rule of the target data is determined through the data source input of the target data, the target data is analyzed into the extreme point data of the corresponding type according to the data source analysis rule, the model data is obtained, the total data volume of the robot is converted into an uploading acceptable range after being preprocessed, and support work is provided for subsequent cloud computing.

In one embodiment, the method for directionally screening the original time sequence data through the rule engine to obtain the target data meeting the screening condition of the rule engine may include the following steps:

And directionally screening the original time sequence data by using the sphere parameter of the rule engine to obtain target data meeting the screening condition of the sphere parameter.

Where "is a key in the SQL (Structured Query Language ) language used to filter records that meet certain conditions. In SQL statements, "where" is typically followed by some logical expression that specifies the recording conditions that need to be screened. "where screening" is a conditional statement used in database queries for filtering and screening results. Using the WHERE screen, rows meeting the conditions can be selected according to the particular conditions. These conditions may include comparison operators (e.g., equal to, greater than, less than, etc.), logical operators (e.g., AND, OR, NOT), and other functions and expressions. By defining the conditions in the where clause, the data can be filtered according to the requirements, returning only records that satisfy the conditions.

For the original time sequence data, the sphere parameter of the rule engine can be used for directionally screening the original time sequence data to obtain target data meeting the screening condition of the sphere parameter.

According to the digital twin data acquisition method provided by the application, the write-in of a large number of data which are inconsistent in condition can be removed by carrying out directional screening on the sphere parameters of the rule engine, so that the overall data output is greatly reduced. In a specific embodiment, the whole data volume can be reduced to 5% -30% of the original data volume through the directional transmission of the data and the screening of the data.

In one exemplary embodiment, the present application also provides a digital twin construction system comprising: edge servers and cloud center servers;

The edge server is used for executing the method so as to transmit the model data to the cloud center server;

The cloud center server is used for constructing a digital twin model corresponding to the robot based on the model data.

Specifically, the embodiment of the application can jointly construct the digital twin system technology of the robot based on the digital twin modeling technology realized by AutomationML industrial modeling language and combining the edge computing technology, the cloud computing technology and the edge cloud cooperation technology. The digital twin construction system can be realized based on an edge cloud architecture, and the edge cloud architecture mainly comprises an edge end and a cloud end. For example, the cloud may be a cloud center server responsible for performing cloud computing. The edge may be an edge server responsible for edge computation.

In one embodiment, as shown in FIG. 4, an edge server is configured with:

the MQTT is used for receiving robot data;

the edge data ETL tool ekuiper is connected with the MQTT and is used for processing the robot data to obtain model data;

The cloud center server is configured with kafka for receiving the model data transmitted by the edge data ETL tool ekuiper and transmitting to the digital twin construction module of the cloud center server.

In particular, the MQTT is used for providing an MQTT service and receiving data uploading of robots in real time, and a single machine can receive data uploading of a plurality of robots, and the data uploading is dependent on the data volume of the robots.

The edge data ETL tool ekuiper serves as the core to carry over and pre-process the data while the original data is retained at influxdb as original data, typically 1-7 days.

Influxdb is used as a time sequence database of the edge terminal and mainly used for providing a data storage function of the edge terminal.

The kafka serves as a cloud end, can accept data transmission of the edge-end data ETL tool ekuiper, and can accept data transmission of a plurality of edge-end data ETL tools ekuiper due to the fact that the kafka components have the characteristic of being capable of being expanded transversely.

In one embodiment, a digital twinning construction module includes:

Model database, dispatcher and robot model library;

the model database is used for receiving and storing the model data transmitted by the kafka;

The dispatcher is used for constructing a robot model library according to the data in the model database, and the robot model library stores digital twin models corresponding to the robots.

In particular, the model database may be clickhouse database. clickhouse have two roles: the first point is that clickhouse, which is data that consumers accept kafka, has high compatibility with kafka due to being one of the components of cloud computing big data. The second point is to provide the function of OLAP, and to use SQL to perform real-time online analysis work, and to provide a faster cloud data processing function. The scheduler may be a dolphin scheduler, which is mainly used for performing ETL operation, and is deeply combined with clickhouse in the cloud computing process, and obtains the configuration of the python code and the data source of the core according to specific service requirements, and finally calculates the model required by the service through the cloud-up robot data. The method can gradually build an algorithm model library (an action model, a process model, a diagnosis model, a mechanism model and an energy consumption model) based on the robot, and provides an implementation method for the development of related services of the subsequent robot, the construction of a digital twin system of the robot and the construction of a lean production system based on digitalization.

Based on the above architecture, by combining the above edge data preprocessing cases with the original characteristic engineering preprocessing and the where screening processing, the total data size of the robot is reduced to 1%to 5% again, so that the total data size of the robot is reduced to 200M to 20G from the original huge data of 26G-430G per day, the preprocessed total data size is changed into an uploading acceptable range, and supporting work is provided for cloud computing.

To further illustrate the solution of the embodiment of the present application, the following description is made with reference to a specific example, as shown in fig. 5, where the construction of a digital twin model of a robot requires acquisition of characteristic engineering data and digital twin data.

Wherein the digital twinning data includes attributes and capabilities of the robot defined in AutomationML files. As shown in fig. 6, the robot is described with AutomationML, and attribute information of the robot includes main body information, sub-element information, interface information, character information, and parameter information. Specifically, the attribute information corresponds to the following modules (see table below) of AutomationML language, respectively to the classifications and functions in the table below. Wherein INSTANCEHIERARCHY corresponds to the body information; INTERNALELEMENT corresponding to subelement information; interface corresponding Interface information; role corresponding Role information; the attribute corresponds to the parameter information. Specifically, the robot is one type of equipment, and is composed of a plurality of parts, and the subelement information is part information.

Further, the motion modeling and the behavior modeling may be performed by collecting robot data through processes of the connection component, the stream processing component, the behavior component, and the motion component through the flow shown in fig. 7. And combining AutomationML files to jointly form digital twin data in the robot digital twin model.

The digital twin data acquisition method provided by the embodiment of the application is used for acquiring the characteristic engineering data in the robot digital twin model based on the digital twin construction system.

Specifically, data screening is first required according to service characteristics. For example, regarding the energy consumption service, only the characteristic engineering data related to the robot is required to be preprocessed and screened, so that irrelevant data is avoided from being extracted. According to specific business characteristic engineering, the whole data volume is reduced to 20% -50% of the original data volume.

Second, data screening may be performed according to rules. The write-in of a large number of data which are inconsistent in condition can be removed by carrying out directional screening through the sphere parameter of the rule engine, so that the output of the whole data quantity is greatly reduced. The whole data quantity is reduced to 5% -30% of the original data again through the directional transmission of the data and the screening of the data. The specific rule number is recorded in ekuiper transmission process, in a specific application scene, the original input is 1683766, the data is reduced to 410002 after the sphere screening of the data, and the data size is removed by 3/4.

Again, the preprocessing of the data may be performed according to rules, namely a streaming computing function. This again reduces the data to 1% -10% of the original. When one time sequence data is needed to calculate one extreme point, the data of the extreme point can be output as a result, and not all time sequence data are needed. Alternatively, using the ekuiper sliding window function, an extreme point can be considered to be reached if the maximum or minimum value is not the first and last value within the sliding window. Based on the sliding window function, extreme point data can be obtained through calculation through the original time sequence data and used as model data.

In the above method, the robot data is received through the MQTT, kuiper serves as a core, the data is accepted and preprocessed, and at the same time, the original data is reserved as the original data at influxdb. kafka serves as a cloud end, can accept ekuiper data transmission, and is further used for receiving and storing model data transmitted by kafka through a Clickhouse database; the dolphin dispatcher is used for constructing a robot model library according to data in the Clickhouse database, and the robot model library stores digital twin models corresponding to the robots. Finally, the construction of a digital twin model of the robot in an industrial scene is realized.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a digital twin data acquisition device for realizing the above-mentioned digital twin data acquisition method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in the embodiments of the digital twin data acquisition device or devices provided below may be referred to the limitation of the digital twin data acquisition method hereinabove, and will not be repeated herein.

In one exemplary embodiment, as shown in FIG. 8, a digital twin data acquisition device 800 is provided, comprising:

the preprocessing module 801 is configured to perform preprocessing screening on the robot data based on the data type of the feature engineering data required for constructing the robot digital twin model, so as to obtain original time sequence data of the corresponding feature engineering data;

The directional screening module 802 is configured to perform directional screening on the original time sequence data through the rule engine, so as to obtain target data that meets the screening condition of the rule engine;

the streaming processing module 803 is configured to perform streaming computation processing on the target data, so as to obtain model data corresponding to the robot digital twin model.

In one embodiment, the streaming processing module is further configured to:

determining a data source analysis rule of the target data based on the data source input of the target data;

and analyzing the target data into extreme point data of a corresponding type according to the data source analysis rule to obtain model data.

In one embodiment, the directional screening module is configured to:

And directionally screening the original time sequence data by using the sphere parameter of the rule engine to obtain target data meeting the screening condition of the sphere parameter.

The modules in the digital twin data acquisition device can be realized in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one exemplary embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 9. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is for storing digital twin build data. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by a processor implements a digital twin data acquisition method.

It will be appreciated by persons skilled in the art that the architecture shown in fig. 9 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In an exemplary embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the steps of the method described above when executing the computer program.

In one embodiment, a computer readable storage medium is provided, having stored thereon a computer program which, when executed by a processor, implements the steps of the method described above.

In an embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, implements the steps of the method described above.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims (10)

1. A digital twin data acquisition method, the method comprising:

Preprocessing and screening the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data corresponding to the characteristic engineering data;

Directionally screening the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine;

and carrying out stream computing processing on the target data to obtain model data corresponding to the robot digital twin model.

2. The method according to claim 1, wherein the performing a streaming calculation on the target data to obtain model data corresponding to the digital twin model of the robot includes:

Determining a data source parsing rule of the target data based on data source input of the target data;

And according to the data source analysis rule, analyzing the target data into extreme point data of a corresponding type to obtain the model data.

3. The method according to claim 1, wherein the directionally screening the original time sequence data by a rule engine to obtain target data meeting a screening condition of the rule engine comprises:

And directionally screening the original time sequence data by using a sphere parameter of a rule engine to obtain target data meeting the screening condition of the sphere parameter.

4. A digital twin data acquisition device, the device comprising:

The preprocessing module is used for preprocessing and screening the robot data based on the data type of the characteristic engineering data required by constructing the robot digital twin model to obtain the original time sequence data corresponding to the characteristic engineering data;

The directional screening module is used for carrying out directional screening on the original time sequence data through a rule engine to obtain target data meeting the screening conditions of the rule engine;

And the stream processing module is used for carrying out stream calculation processing on the target data to obtain model data for constructing the digital twin model of the robot.

5. A digital twinning construction system, the system comprising: edge servers and cloud center servers;

the edge server is configured to perform the method of any one of claims 1 to 3 to transmit the model data to the cloud center server;

The cloud center server is used for constructing a digital twin model corresponding to the robot based on the model data.

6. The system of claim 5, wherein the edge server is configured with:

the MQTT is used for receiving the robot data;

An edge end data ETL tool ekuiper connected with the MQTT and used for processing the robot data to obtain the model data;

The cloud center server is configured with kafka, and is configured to receive the model data transmitted by the edge data ETL tool ekuiper, and transmit the model data to a digital twin construction module of the cloud center server.

7. The system of claim 6, wherein the digital twinning construction module comprises:

Model database, dispatcher and robot model library;

The model database is used for receiving and storing the model data transmitted by the kafka;

The scheduler is used for constructing the robot model library according to the data in the model database, and the robot model library stores digital twin models corresponding to the robots.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 3 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 3.

10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method of any one of claims 1 to 3.