WO2023044631A1

WO2023044631A1 - A device, system, method and storage medium for ai application deployment

Info

Publication number: WO2023044631A1
Application number: PCT/CN2021/119718
Authority: WO
Inventors: Hongyang Zhang; Bingchao TANG
Original assignee: Siemens Aktiengesellschaft; Siemens Ltd., China
Priority date: 2021-09-22
Filing date: 2021-09-22
Publication date: 2023-03-30
Also published as: EP4359908A1

Abstract

A device, system, method and computer readable storage medium for an Al application deployment are provided. The device includes: an environment configuration module (lll), to provide a visualized environment configuration interface, and set an Al application running device (120) as a target industrial server based on a user's running environment configuration; a parameter configuration module (112), to provide a visualized parameter configuration interface and receive input and output parameter configuration of Al application from the user; a method configuration module (113), to provide a visualized method configuration interface and receive control method configuration of Al application from the user; a first deployment module (114), to generate an Al application according to the input and output parameter configuration and the control method configuration, and deploy the Al application to the Al application running device (120); and a second deployment module (115), to generate a description file of the Al application according to the input and output parameter configuration and the control method configuration, and deploy the description file to corresponding industrial client device (140).

Description

A DEVICE, SYSTEM, METHOD AND STORAGE MEDIUM FOR AI APPLICATION DEPLOYMENT

FIELD

The present disclosure relates to industrial automation technologies, and more particularly, to a device, system, method and computer readable storage medium for an artificial intelligence (AI) application deployment.

BACKGROUND

Artificial intelligence (AI) is widely used in the fields of computer vision, robotics and intelligent recommendation. More and more automation companies are trying to introduce AI into industrial automation systems.

However, the current problem is that it is difficult to deploy and integrate an AI application into an industrial automation system. Traditional automation devices such as programmable logic controller (PLC) , human machine interface (HMI) and industrial personal computer (IPC) have been widely used in factories, and automation engineers are also very familiar with them. However, AI applications are usually developed by AI experts, and the interfaces of AI applications are usually complex and different. Additional coding is required before deployment, as different automation applications may require different inputs and outputs of AI applications. These configuration and deployment work of AI applications are not familiar to automation engineers and it is difficult for the AI applications to communicate with other automation devices. These difficulties greatly limit the process of integrating AI technology into the automation field.

SUMMARY

According to examples of the present disclosure, a device, system, method and computer readable storage medium for AI application deployment is provided to reduce the complexity of AI application deployment and improve the deployment efficiency.

The device for AI application deployment provided by examples of the present disclosure includes: an environment configuration module, to provide a visualized environment configuration interface to a user, receive running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device via the environment configuration interface, and set the AI application operation device as a target industrial server based on the running environment configuration; a parameter configuration module, to provide a visualized parameter configuration interface to the user and receive input and output parameter configuration of AI application conducted by the user via the parameter configuration interface; a method configuration module, to provide a visualized method configuration interface to the user, and receive control method configuration of AI application conducted by the user via the method configuration interface; a first deployment module, to generate an AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploy the AI application to the AI application running device; and a second deployment module, to generate a description file of the AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploy the description file to corresponding industrial client device.

In an example, the parameter configuration module import a pre trained AI model and a corresponding description file of the AI model, analyze the AI model in combination with the description file, determine input and output candidate parameter variables of the AI model, and present the input and output candidate parameter variables to the user in the form of visualized structure through the parameter configuration interface to make the user select current input and output parameter variables for the target industrial communication protocol to complete the input and output parameter configuration of the AI application.

In an example, the second deployment module map current input and output parameter variables obtained from the input and output parameter configuration and current control method obtained from the control method configuration to an information model satisfying the target industrial communication protocol, and the current input-output parameter variables and processes in the current control method are mapped to nodes of the information model; based on the information model, to generate a description file in a corresponding format according to the standard of the target communication protocol.

The system for AI application deployment provided by examples of the present disclosure includes: an AI application running device, an industrial client device, and a device for AI application deployment above mentioned; wherein, the AI application running device is to run the AI application, receive input parameters from the industrial client device according to running requirements, and provide obtained output result to the industrial client device; the industrial client device is to interact with the AI application running device according to the description file, provide corresponding input parameters to the AI application running device, and receive the output result from the AI application running device.

The method for AI application deployment provided by examples of the present disclosure includes: providing a visualized environment configuration interface to a user, receiving running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device via the environment configuration interface, and setting the AI application operation device as a target industrial server according to the running environment configuration; providing a visualized parameter configuration interface to the user, and receiving input and output parameter configuration of AI application conducted by the user via the parameter configuration interface; providing a visualized method configuration interface to the user, and receiving control method configuration of AI application conducted by the user via the method configuration interface; generating an AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploying the AI application to the AI application running device; generating a description file of the AI application according to the input and output parameter configuration and the control method configuration, and deploying the description file to corresponding industrial client device.

In an example, wherein providing a visualized parameter configuration interface to the user, and receiving input and output parameter configuration of AI application conducted by the user via the parameter configuration interface, comprises: importing a pre trained AI model and a corresponding description file of the AI model; analyzing the AI model in combination with the description file, and determining input and output candidate parameter variables of the AI model; presenting the input and output candidate parameter variables to the user through the parameter configuration interface in the form of visualized structure; receiving the input and output parameter configuration of AI application by selecting, by the user, current input and output parameter variables for the target industrial communication protocol from the candidate parameter variables.

In an example, wherein generating a description file of the AI application according to the input and output parameter configuration and the control method configuration, and deploying the description file to corresponding industrial client device, comprises: mapping the current input and output parameter variables and current control method obtained from the control method configuration to an information model satisfying the target industrial communication protocol, the input and output parameter variables and processes in the current control method are mapped to nodes of the information model; based on the information model, generating a description file in a corresponding format according to standard of the target communication protocol.

In an example, wherein the target industrial communication protocol is OPC UA, the description file of the AI application is a description file in XML format.

Another system AI application deployment provided by examples of the present disclosure includes: at least one memory, to store a computer program; and at least one processor, to call the computer program stored in the at least one memory to perform a method for AI application deployment mentioned above.

The non-transitory computer-readable storage medium, on which a computer program is stored, the computer program is to be executed by a processor to implement a method for AI application deployment mentioned above.

It can be seen from above mentioned technical solutions in embodiments of the present disclosure, the user only need to make simple configuration through the visualized configuration interface, the configuration of corresponding target industrial server, the generation and deployment of the description file and AI application may be completed by the AI application deployment device, thus the complexity of AI application deployment is reduced and the deployment efficiency is improved.

In addition, through the automatic analysis of the AI model, the input and output parameter variables supported by the AI model are determined, and the input and output parameter variables are presented to the user in a visualized structure, so that the user can complete the input and output parameter configuration for the current industrial communication protocol only by simple clicking, and other configuration processes can also be completed by simple clicking. It can be seen that the configuration process also simplifies the user's operation and reduces the requirements for the user's professional knowledge.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the present disclosure, reference should be made to the Detailed Description below, in conjunction with the following drawings in which like reference numerals refer to corresponding parts throughout the figures.

Fig. 1A is a schematic diagram illustrating a system for AI application deployment according to embodiments of the present disclosure.

Fig. 1B is a schematic diagram illustrating a device for AI application deployment according to embodiments of the present disclosure.

Fig. 2 is a schematic diagram illustrating an AI application conforming to the OPC UA standard according to an example of the present disclosure.

Fig. 3 is a schematic diagram illustrating an information model conforming to the OPC UA standard according to an example of the present disclosure.

Fig. 4 is flow diagram illustrating a method for AI application deployment according to embodiments of the present invention.

Fig. 5 is a schematic diagram illustrating another system for AI application deployment according to embodiments of the present disclosure.

The reference numerals are as follows:

Reference numeral	Object
110	AI application deployment device
111	environment configuration module
112	parameter configuration module
113	method configuration module
114	first deployment module
115	second deployment module
120	AI application running device
130	data obtaining device
131	camera

132	sensor
140	industrial client device
141	PLC
142	HMI
143	MES
144	SCADA
21	start
22	image capture
23	image preprocessing
24	image classification
241	call image classification AI model
242	model structure visualization
243	input/output layer recognition
25	classification result post-processing
26	end
S401～S404	processes
51	memory
52	processor
53	displayer
54	bus

DETAILED DESCRIPTION

In embodiments of the present disclosure, it is considered that the current common method is that the automation company provides special devices and corresponding template solutions, and uses a pre trained AI model according to a specific use case. This means that the functionality is very limited and automation engineers cannot use AI models with different inputs or outputs. Another method is to deploy an AI application to a personal computer (PC) or an edge device using general operating systems such as windows or Linux. The interface of an AI application is usually API or fixed network communication protocol, which is also difficult to be integrated into the automation system.

Therefore, in the embodiments of the present disclosure, it is considered to provide a scheme that can effectively integrate an AI application into the automation system, so that the automation engineer can easily configure and deploy an AI application to an IPC device or an edge device. For example, an AI application can be integrated into an automation system which is based on an industrial communication protocol such as OPC UA.

Reference will now be made in detail to examples, which are illustrated in the accompanying drawings. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. Also, the figures are illustrations of an example, in which modules or procedures shown in the figures are not necessarily essential for implementing the present disclosure. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the examples.

Fig. 1A is a schematic diagram illustrating a system for AI application deployment according to embodiments of the present disclosure. As shown in Fig. 1A, the system may include an AI application deployment device 110, an AI application running device 120, a data obtaining device 130, and an industrial client device 140.

The AI application deployment device 110 is configured to provide a user a visualized environment configuration interface, a visualized parameter configuration interface and a visualized method configuration interface; to set the AI application running device 120 as a target industrial server according to running environment configuration for a target industrial communication protocol conducted by the user for an AI application running device 120 via the environment configuration interface; to generate a AI application of an AI model according to input and output parameter configuration of the AI application via the parameter configuration interface and the method configuration of the AI application via the control method configuration interface, and to deploy the AI application to the AI application running device 120; at the same time, to generate a description file of the AI application according to the input and output parameter configuration of the AI application and the method configuration of the AI application, and to deploy the description file to an corresponding industrial client device 140.

The running environment configuration may include the configuration of industrial server name, industrial server address, port, maximum number of industrial sessions, registered nodes, etc.

The AI application deployment device 110 may be configured to import a pre trained AI model and its corresponding description file, analyze the AI model in combination with the description file, determine input and output candidate parameter variables of the AI model, and present determined input and output candidate parameter variables to the user in a visualized structure through the parameter configuration interface, The user can click to select current input and output parameter variables for the target industrial communication protocol, so as to complete the input and output parameter configuration of the AI application.

The method configuration of AI application refers to the configuration of control methods including the operation and/or state feedback of an AI application, and a current control method for AI application is obtained.

The AI application deployment device 110 may map the current input and output parameter variables and the current control method to an information model satisfying the target industrial communication protocol. The current input and output parameter variables and the current control method are mapped to a node of the information model. The AI application deployment device 110 may generate a description file in a corresponding format according to a standard of the target communication protocol according to the information model.

In an example of the present disclosure, the target industrial communication protocol may be OPC UA, and accordingly, the description file in the corresponding format may be a description file in XML format.

Fig. 2 is a schematic diagram illustrating an AI application conforming to the OPC UA standard according to an example of the present disclosure. As shown in Fig. 2, the AI model in this embodiment is an image classification AI model. The process of AI application includes: start 21, image capture 22, image preprocessing 23, image classification 24, classification result post-processing 25 and end 26.

The process of the AI application is configured by the user through the method configuration interface, and the input and output parameter variables of the AI model involved in the process may be configured by the user through the parameter configuration interface. For example, for the configuration of start 21, the leftmost table at the bottom in Fig. 2 may be presented to the user through the method configuration interface for the user to select whether to read and write by the OPC UA device to start operation and feedback state operation. As another example, for the configuration of image capture 22, the table in the lower middle of Fig. 2 may be presented to the user through the parameter configuration interface for the user to select whether to read and write resolution width and resolution height by OPC UA device. As another example, for the configuration of the image classification 24, it may be configured to call the corresponding image classification AI model 241. Further, the model structure may be visualized 242 and the input/output layer of the model may be recognized 243. For another example, for the classification result post-processing 25, it may determine the parameters to be output according to the user's configuration of the output parameter variables, that is, in the parameter configuration stage, the rightmost table at the bottom in Fig. 2 may be presented to the user through the parameter configuration interface for the user to select the array read by the OPC UA device from the structure array with a length of 1000, In the figure, the user selects the first two arrays of data as an example.

Fig. 3 is a schematic diagram illustrating an information model conforming to the OPC UA standard according to an example of the present disclosure. As shown in Fig. 3, in OPC UA address space, the information model of AI application is automatically generated according to user configuration. The current input and output parameter variables and processes in the current control method configured by the user are mapped to corresponding nodes in the information model. Each model is configured as a node of "BaseObjectType" and added to the OPC UA address space. There may be multiple model nodes in an information model, such as the first model node Model1 and the second model node Model2 in Fig. 2. These model nodes may be child nodes of an “Objects” node. The “Objects” node may be a child node of a “Root” node, and child nodes of the “Objects” node may further include a “Server” node and a “Folder Type” node. Under each model node, two nodes with type “BaseObjectType” and name “Inputs” and “Outputs” may be added. The current input parameter variables selected by the user will be added to the "inputs" node, and the type of each "inputs" node will be mapped to the "DataType" of OPC UA. The same process is also applicable to the output of the model. For the current control method, basic methods such as "Start" , "End" and "Get Status" are also configured as nodes under the model node.

The AI application running device 120 is configured to run the AI application, receive input parameters from corresponding industrial client device 140 according to the running requirements, and interact with corresponding data obtaining device 130 to obtain data collected by the data obtaining device 130; to provide obtained output result to the corresponding industrial client device 140. In an example, the AI application running device 120 may be an IPC device or an edge device.

The data obtaining device 130 is configured to collect corresponding industrial data and provide collected data to the AI application operation device 120. In an example, the data obtaining device 130 may include a camera 131, a sensor 132, and the like.

The industrial client device 140 is configured to interact with the AI application running device 120 according to the description file, provide corresponding input parameters to the AI application running device 120, and receive the output result from the AI application running device 120. In an example, the industrial client device 140 may include a programmable logic controller (PLC) 141, a human-computer interaction interface device (HMI) 142, a manufacturing execution system (MES) 143, a supervisory control and data acquisition (SCADA) system 144, etc.

In one example, as shown in Fig. 1B, the AI application deployment device 110 may include an environment configuration module 111, a parameter configuration module 112, a method configuration module 113, a first deployment module 114 and a second deployment module 115

The environment configuration module 111 is configured to provide a user with a visualized environment configuration interface, receive running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device 120 via the environment configuration interface, and set the AI application operation device 120 as a target industrial server based on the running environment configuration.

The parameter configuration module 112 is configured to provide the user with a visualized parameter configuration interface and receive input and output parameter configuration of AI application conducted by the user via the parameter configuration interface. In an example, the parameter configuration module 112 may import a pre trained AI model and a corresponding description file thereof as described above, analyze the AI model in combination with the description file, determine input and output candidate parameter variables of the AI model, and present determined input and output candidate parameter variables to the user in the form of visualized structure through the parameter configuration interface. The user may click to select current input and output parameter variables for the target industrial communication protocol to complete the input and output parameter configuration of the AI application.

The method configuration module 113 is configured to provide the user with a visualized method configuration interface and receive control method configuration of AI application conducted by the user via the method configuration interface.

The first deployment module 114 is configured to generate an AI application of the AI model according to the input and output parameter configuration of the AI application and the control method configuration of the AI application, and deploy the AI application to the AI application running device 120.

The second deployment module 115 is configured to generate a description file of the AI application according to the input and output parameter configuration of the AI application and the control method configuration of the AI application, and deploy the description file to corresponding industrial client device 140. In an example, the second deployment module 115 may map current input and output parameter variables obtained from the input and output parameter configuration and current control method obtained from the control method configuration to an information model meeting the target industrial communication protocol, and the current input and output parameter variables and processes in the current control method are mapped to nodes of the information model; based on the information model, to generate a description file in a corresponding format according to the standard of the target communication protocol.

The system for AI application deployment system according to embodiments of the present disclosure is described in detail above, and the method for AI application deployment according to embodiments of the present disclosure will be described in detail hereinafter. The method for AI application deployment method according to embodiments of the present disclosure can be implemented on the system for AI application deployment according to embodiments of the present disclosure. For details not disclosed in the embodiments of the method of the present disclosure, please refer to the corresponding description in the embodiments of the system of the present disclosure, which will not be repeated here.

Fig. 4 is flow diagram illustrating a method for AI application deployment according to embodiments of the present invention. As shown in Fig. 4, the method may include the following processes:

At block S401, a visualized environment configuration interface is provided to a user, and running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device via the environment configuration interface is received, and the AI application operation device is set as a target industrial server according to the running environment configuration.

At this block, the information to be configured of the running environment configuration for a target industrial communication protocol may be presented to the user through the environment configuration interface in the form of visualized structure, such as drop-down menu, text box or list. The information to be configured may include industrial server name, industrial server address, port, maximum number of industrial sessions, registered nodes, etc.

At block S402, a visualized parameter configuration interface is provided to the user, and input and output parameter configuration of AI application conducted by the user via the parameter configuration interface is received; and a visualized method configuration interface is provided to the user, and control method configuration of AI application conducted by the user via the method configuration interface is received.

At this block, a pre trained AI model and a corresponding description file thereof may be imported, the AI model may be analyzed in combination with the description file, input and output candidate parameter variables of the AI model may be determined, and determined input and output candidate parameter variables may be presented to the user through the parameter configuration interface in the form of visualized structure, such as data table or list, The user can complete the input and output parameter configuration of the AI application by selecting current input and output parameter variables for the target industrial communication protocol from the candidate parameter variables.

In addition, the control method configuration process of AI application may be presented to the user through the method configuration interface in the form of visualized structure, such as drop-down menu or selectable list. The user may obtain the current control method for the target industrial communication protocol by clicking corresponding candidate options, so as to complete the method configuration of AI application.

At block S403, an AI application is generated according to the input and output parameter configuration of the AI application and the control method configuration of the AI application, and the AI application is deployed to the AI application running device.

At block S404, a description file of the AI application is generated according to the input and output parameter configuration and the control method configuration, and the description file is deployed to corresponding industrial client device.

At this block, the current input and output parameter variables and the current control method may be mapped to an information model satisfying the target industrial communication protocol, the input and output parameter variables and processes in the current control method may be mapped to nodes of the information model; based on the information model, a description file in a corresponding format is generated according to the standard of the target communication protocol.

In an example, the target industrial communication protocol is OPC UA, the description file of the AI application is a description file in XML format.

It should be noted that not all blocks and modules in the above flow and schematic diagrams are necessary, and some blocks or modules can be ignored according to actual needs. The execution sequence of blocks is not fixed and can be adjusted as needed. The division of modules is only functional division for the convenience of describing. In actual implementation, a module can be realized by multiple modules, and the functions of multiple modules can also be realized by one module. These modules can be located in the same device or in different devices.

It can be understood that the hardware modules in above embodiments can be implemented mechanically or electronically. For example, a hardware module may include specially designed permanent circuits or logic devices (such as special-purpose processors, such as FPGA or ASIC) to complete specific operations. The hardware module may also include a programmable logic device or circuit temporarily configured by software (such as including a general-purpose processor or other programmable processor) for performing specific operations. As for the specific use of mechanical mode, or special permanent circuit, or temporarily configured circuit (such as configured by software) to realize the hardware module, it can be determined according to the consideration of cost and time.

In fact, the system for AI application deployment provided by this implementation manner of the present disclosure may be specifically implemented in various manners. For example, the system for AI application deployment may be compiled, by using an application programming interface that complies with a certain regulation, as a plug-in that is installed in an intelligent terminal, or may be encapsulated into an application program for a user to download and use.

When compiled as a plug-in, the system for AI application deployment may be implemented in various plug-in forms. The system for AI application deployment provided by this implementation manner of the present disclosure may also be implemented by using a specific technology, such as a Flash plug-in technology, a RealPlayer plug-in technology, an MMS plug-in technology, a MIDI staff plug-in technology, or an ActiveX plug-in technology.

The method for AI application deployment provided by this implementation manner of the present disclosure may be stored in various storage mediums in an instruction storage manner or an instruction set storage manner. These storage mediums include, but are not limited to: a floppy disk, an optical disc, a DVD, a hard disk, a flash memory, a USB flash drive, a CF card, an SD card, an MMC card, an SM card, a memory stick, and an xD card.

Moreover, it should be clear that an operating system operated in a computer can be made, not only by executing program code read by the computer from a storage medium, but also by using an instruction based on the program code, to implement some or all actual operations, so as to implement functions of any embodiment in the foregoing embodiments.

For example, Fig. 5 is a schematic diagram illustrating another system for AI application deployment according to embodiments of the present disclosure. The system may be used to perform the method shown in figure 4, or to implement the system shown in figure 1. As shown in figure 5, the system may include at least one memory 51, at least one processor 52 and at least one displayer 53. In addition, some other components may be included, such as communication port, input/output controller, network communication interface, etc. These components communicate through bus 54, etc.

At least one memory 51 is configured to store a computer program. In one example, the computer program can be understood to include various modules of the system shown in figure 1. In addition, at least one memory 51 may store an operating system or the like. Operating systems include but are not limited to: Android operating system, Symbian operating system, windows operating system, Linux operating system, etc.

At least one processor 52 is configured to call the computer program stored in at least one memory 51 to perform a method for AI application deployment described in examples of the present disclosure. The processor 52 can be CPU, processing unit/module, ASIC, logic module or programmable gate array, etc. It can receive and send data through the communication port.

At least one display 53 is configured to display the respective configuration interfaces.

The I/O controller has an input device, which is used to input, output and display relevant data.

It should be understood that, as used herein, unless the context clearly supports exceptions, the singular forms "a" ( "a" , "an" , "the" ) are intended to include the plural forms. It should also be understood that, "and /or" used herein is intended to include any and all possible combinations of one or more of the associated listed items.

The number of the embodiments of the present disclosure are only used for description, and do not represent the merits of the implementations.

The foregoing description, for purpose of explanation, has been described with reference to specific examples. However, the illustrative discussions above are not intended to be exhaustive or to limit the present disclosure to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The examples were chosen and described in order to best explain the principles of the present disclosure and its practical applications, to thereby enable others skilled in the art to best utilize the present disclosure and various examples with various modifications as are suited to the particular use contemplated.

Claims

A device for AI application deployment, characterized in that, comprises:

an environment configuration module (111) , to provide a visualized environment configuration interface to a user, receive running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device via the environment configuration interface, and set the AI application operation device as a target industrial server based on the running environment configuration;

a parameter configuration module (112) , to provide a visualized parameter configuration interface to the user and receive input and output parameter configuration of AI application conducted by the user via the parameter configuration interface;

a method configuration module (113) , to provide a visualized method configuration interface to the user, and receive control method configuration of AI application conducted by the user via the method configuration interface;

a first deployment module (114) , to generate an AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploy the AI application to the AI application running device; and

a second deployment module (115) , to generate a description file of the AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploy the description file to corresponding industrial client device.
The device according to claim 1, characterized in that, the parameter configuration module (112) import a pre trained AI model and a corresponding description file of the AI model, analyze the AI model in combination with the description file, determine input and output candidate parameter variables of the AI model, and present the input and output candidate parameter variables to the user in the form of visualized structure through the parameter configuration interface to make the user select current input and output parameter variables for the target industrial communication protocol to complete the input and output parameter configuration of the AI application.
The device according to claim 2, characterized in that, the second deployment module (115) map current input and output parameter variables obtained from the input and output parameter configuration and current control method obtained from the control method configuration to an information model satisfying the target industrial communication protocol, and the current input-output parameter variables and processes in the current control method are mapped to nodes of the information model; based on the information model, to generate a description file in a corresponding format according to the standard of the target communication protocol.
A system for AI application deployment, characterized in that, comprises: an AI application running device (120) , an industrial client device (140) , and a device for AI application deployment (110) according to any one of claims 1 to 3; wherein,

the AI application running device (120) is to run the AI application, receive input parameters from the industrial client device (140) according to running requirements, and provide obtained output result to the industrial client device (140) ;

the industrial client device (140) is to interact with the AI application running device (120) according to the description file, provide corresponding input parameters to the AI application running device (120) , and receive the output result from the AI application running device (120) .
A method for AI application deployment, characterized in that, comprises:

providing a visualized environment configuration interface to a user, receiving running environment configuration for a target industrial communication protocol conducted by the user for an AI application operation device via the environment configuration interface, and setting the AI application operation device as a target industrial server according to the running environment configuration;

providing a visualized parameter configuration interface to the user, and receiving input and output parameter configuration of AI application conducted by the user via the parameter configuration interface;

providing a visualized method configuration interface to the user, and receiving control method configuration of AI application conducted by the user via the method configuration interface;

generating an AI application according to the input and output parameter configuration of AI application and the control method configuration of AI application, and deploying the AI application to the AI application running device;

generating a description file of the AI application according to the input and output parameter configuration and the control method configuration, and deploying the description file to corresponding industrial client device.
The method according to claim 5, characterized in that, wherein providing a visualized parameter configuration interface to the user, and receiving input and output parameter configuration of AI application conducted by the user via the parameter configuration interface, comprises:

importing a pre trained AI model and a corresponding description file of the AI model;

analyzing the AI model in combination with the description file, and determining input and output candidate parameter variables of the AI model;

presenting the input and output candidate parameter variables to the user through the parameter configuration interface in the form of visualized structure;

receiving the input and output parameter configuration of AI application by selecting, by the user, current input and output parameter variables for the target industrial communication protocol from the candidate parameter variables.
The method according to claim 6, characterized in that, wherein generating a description file of the AI application according to the input and output parameter configuration and the control method configuration, and deploying the description file to corresponding industrial client device, comprises:

mapping the current input and output parameter variables and current control method obtained from the control method configuration to an information model satisfying the target industrial communication protocol, the input and output parameter variables and processes in the current control method are mapped to nodes of the information model; based on the information model, generating a description file in a corresponding format according to standard of the target communication protocol.
The method according to any one of claims 5 to 7, characterized in that, wherein, the target industrial communication protocol is OPC UA, the description file of the AI application is a description file in XML format.
A system for AI application deployment, characterized in that, comprises:

at least one memory (51) , to store a computer program; and

at least one processor (52) , to call the computer program stored in the at least one memory (51) to perform a method for AI application deployment according to any one of claims 5 to 8.
A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, the computer program is to be executed by a processor to implement a method for AI application deployment according to any one of claims 5 to 8.