Industrial data acquisition equipment-oriented special MCP tool library construction method and system

Abstract

The invention discloses a special MCP tool library construction method and system for industrial data acquisition equipment, and the method comprises the steps: building a unified function description model for the data acquisition equipment of an industrial site, and developing a standardized equipment access interface based on the model; equipment communication protocols are automatically identified through a multi-protocol identification mechanism, corresponding parsers are loaded, and original equipment data are converted into standardized data; providing services for the standardized data through an equipment access interface, and packaging a special tool based on an MCP tool specification; and registering the packaged tool to an MCP server to form the special tool library for industrial data acquisition. The problems that a general tool library cannot meet the professional requirements of industrial equipment and integration is difficult due to diversity of industrial protocols are solved, a unified and reliable tool calling interface can be provided for an industrial agent, and the efficiency and the intelligent level of industrial data collection and equipment control are remarkably improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of intelligent agents in artificial intelligence technology, and in particular to an industrial data acquisition intelligent agent construction method and system based on a self-defined MCP tool library. BACKGROUND

[0002] A model context protocol (MCP) tool library is a core component for realizing intelligent agent tool calling in the field of artificial intelligence. The MCP tool library encapsulates external functions as standardized tools, so that intelligent agents can understand and execute various operations. In the field of industrial automation, data acquisition (referred to as "data acquisition" for short) is a basic link for realizing device monitoring and production management. The core task of data acquisition is to obtain running parameters and state information from various industrial data acquisition devices (such as PLCs, sensors, industrial gateways, etc.) in real time. Unlike general software service calling, industrial data acquisition needs to directly interact with physical devices. These devices usually use different communication protocols (such as MQTT, Modbus, OPC UA, etc.), have their own special instruction sets and data point table structures. Although the MCP tool library has formed a standardized solution in the general field (such as database query, chart generation, etc.), due to the diversity of industrial device communication protocols and the complexity of control logic, the general MCP tool library is difficult to meet the professional requirements of industrial data acquisition scenarios for accurate device control and reliable data acquisition. A special MCP tool library for industrial data acquisition has important application value in the field of industrial intelligence: 1) device state monitoring: real-time acquisition of device running parameters, such as "query the current running state of No. 1 machine tool", etc.; 2) device remote control: execution of high-frequency operation and maintenance operations, such as "remote restart of the industrial gateway on No. 2 production line" and other precise control instructions.

[0003] The current MCP tool library construction method in the industrial data acquisition field mainly develops around two directions of general tool encapsulation and special tool development. Among them, the general tool encapsulation mainly converts the general software service interface into the MCP tool through the standard interface description; the special tool development writes independent tool implementation according to the specific industrial equipment control demand. However, these methods face two core challenges: firstly, in the general tool encapsulation aspect, since the industrial data acquisition equipment adopts diversified communication protocols (such as MQTT, Modbus, OPC UA, etc.), and the instruction set and data point table structure of each manufacturer's equipment are different, the general MCP tool lacks professional support for the industrial equipment specific protocol and control, and cannot directly adapt to diversified industrial data acquisition equipment. Secondly, in the special tool development aspect, since there is a lack of unified device access standard and function abstraction specification, it is necessary to repeatedly develop functionally similar tools for each device model, resulting in a large number of special tools with different interfaces, causing waste of development resources and increase of maintenance cost. These two challenges jointly restrict the systematic construction and large-scale application of the special MCP tool library in the industrial data acquisition field. SUMMARY

[0004] In view of the deficiencies in the prior art, the present application provides a special MCP tool library construction method and system for industrial data acquisition equipment. The present application first realizes the unified access and control of various industrial data acquisition equipment through a self-developed backend system, encapsulates the device functions of different protocols (including MQTT, Modbus, OPC UA, etc.) into a group of standardized application program interfaces; and then systematically converts these backend interfaces into special MCP tools that can be directly called by intelligent agents through the established MCP tool encapsulation specification. This method not only solves the problem that general MCP tools cannot adapt to special industrial equipment, but also avoids repeated development through a standardized tool encapsulation mode, and finally forms a complete, unified and expandable MCP tool library for industrial data acquisition.

[0005] The present application achieves the above technical object by the following technical means.

[0006] A special MCP tool library construction method for industrial data acquisition equipment:

[0007] A unified function description model is established for the data acquisition equipment in the industrial field, and a data acquisition equipment function vector is defined;

[0008] A standardized device access interface is developed based on the device function vector;

[0009] The communication protocol type of the data acquisition equipment is identified through a protocol matching algorithm, the corresponding protocol analysis plug-in is loaded according to the identified communication protocol type, and the original data of the data acquisition equipment is converted into standardized data;

[0010] The standardized data is provided through the device access interface, and the special MCP tool is packaged based on the MCP tool specification and the MCP protocol adapter;

[0011] The packaged MCP tool is registered to the MCP server to form an industrial data acquisition special tool library.

[0012] Further, the device function vector Wherein, is a data acquisition device identifier, is a data acquisition device type, is a data acquisition device parameter set, is a data acquisition device instruction set.

[0013] Further, the data acquisition device parameter set is: Wherein, is a tool name, is a data type, is a data unit, is a numerical range, is a data update frequency.

[0014] Further, the protocol matching algorithm realizes the identification of the communication protocol type by calculating the similarity of the industrial data acquisition device response data and the protocol feature library, and the calculation formula of the similarity is: Wherein is an industrial data acquisition device response feature, is a protocol standard feature, is the total number of device response features or protocol standard features.

[0015] Further, the original data of the data acquisition device is converted into standardized data, and the conversion process adopts a data mapping function: Wherein, is the original data of the industrial data acquisition device, is an offset, is a scaling ratio, is a reference value.

[0016] Further, the parameters in the data mapping function are dynamically configured according to the type of the data acquisition device, and the configuration rule is:

[0017]

[0018]

[0019]

[0020] Wherein, and Max, Min are the maximum value, minimum value of the engineering value after data standardization, and Max, Min are the maximum value, minimum value of the original value of the industrial data acquisition equipment.

[0021] Further, the MCP tool packaging process defines a tool descriptor: Wherein name is the tool name, inputs is the input parameter, and outputs is the output data format.

[0022] A special MCP tool library construction system for industrial data acquisition equipment, comprising:

[0023] A protocol adaptation module for identifying the communication protocol type of the industrial data acquisition equipment and loading the corresponding protocol analysis plug-in;

[0024] A data standardization module for converting the original data of the industrial data acquisition equipment into standardized data;

[0025] A tool packaging module for packaging special tools according to MCP specifications based on standardized data interfaces;

[0026] A service deployment module for registering the packaged tools to the MCP server to form a special tool library.

[0027] An electronic device comprising a memory and a processor;

[0028] The memory is used to store a computer program;

[0029] The processor is used to execute the computer program and implement the above-mentioned special MCP tool library construction method for industrial data acquisition equipment when executing the computer program.

[0030] A storage medium storing a computer program, which, when executed by a processor, causes the processor to execute the above-mentioned special MCP tool library construction method for industrial data acquisition equipment.

[0031] The beneficial effects of the present application are:

[0032] (1) The present application establishes a device function description model and a standardized interface to construct a special MCP tool library for industrial data acquisition, solving the problem that general tool libraries cannot meet the special requirements of industrial equipment. The method designs a special tool packaging specification for the control instruction set and data point table structure of industrial equipment, so that each tool can accurately understand the operation semantics of industrial equipment. Through unified device function description, the parameters and control instructions of different manufacturer equipment are standardized packaged, ensuring the professionalism and reliability of the tool library in industrial scenarios, and realizing precise control and data acquisition of industrial equipment.

[0033] (2) The application effectively solves the integration problem caused by the coexistence of multiple communication protocols in industrial field through a multi-protocol identification mechanism and dynamic analysis plug-in. The system can automatically identify the type of communication protocol used by the device and dynamically load the corresponding protocol analyzer to convert raw device data of different formats into unified structured data. This method can adaptively process multiple industrial protocols such as Modbus, OPC UA, MQTT, etc., shielding the complexity of the underlying protocol, significantly reducing the technical threshold of industrial device access to intelligent agent systems, and avoiding the problem of repeated development caused by developing interfaces for each device.

[0034] (3) The special MCP tool library construction method of the application adopts modular design, has good scalability and practicality. Through standardized tool encapsulation process and service deployment mechanism, new industrial data acquisition devices can quickly access the system and generate corresponding MCP tools. The dynamic loading of protocol analysis plug-in and the elastic deployment of tools provide a complete special MCP tool library for industrial data acquisition for industrial intelligent agents, greatly promoting the large-scale application of artificial intelligence technology in industrial field, and providing reliable technical support for intelligent manufacturing. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 The special MCP tool library construction method for industrial data acquisition equipment of the application is shown in the framework diagram.

[0036] Figure 2 The complete construction process diagram of the special MCP tool library of the application is shown. DETAILED DESCRIPTION

[0037] The application will be further described below in conjunction with the drawings and specific embodiments, but the protection scope of the application is not limited thereto.

[0038] As shown in Figure 1 , the system of the application constructs a complete architecture from the bottom layer industrial data acquisition equipment to the top layer intelligent agent application. Various types of industrial data acquisition equipment (such as mechanical arm, machine tool, PLC, etc.) access the back-end service layer of the system through various device communication protocols. The back-end service layer is the core, and the protocol dynamic adaptation engine built-in is responsible for unified processing of different device communication protocols, and through the data standardization conversion module, the raw data of industrial data acquisition equipment is converted into regular and unified engineering data. The unified API interface is systematically encapsulated into a special MCP tool library, providing data access, data analysis, device control and other standardized tools for upper layer intelligent agent application.

[0039] The API interface includes three types, namely a data access API interface, a device control API interface and a data analysis API interface. Taking "obtaining a robot arm state" as an example, the system works according to the following flow: first, the protocol adaptation module automatically identifies that the robot arm uses the Modbus protocol for communication through the multi-protocol identification engine, and dynamically loads the corresponding Modbus parser; then, the data standardization module reads the value from the original register of the robot arm (for example, the value of address 40111 is 4550), and calls the data mapping function to convert it into engineering data (45.5 degrees) with clear physical meaning according to the preset rule (for example: engineering value = (original value - 0) * 0.01 + 0). At this point, the original signal of the industrial data acquisition device has been processed and packaged into a standard API interface (such as GET / api / device / robot_arm_001 / status).

[0040] As shown in Figure 2 The special MCP tool library of the application is constructed according to a clear automation process: starting from the initialization of industrial data acquisition device connection, the physical connection is established. Then, the system starts the multi-protocol automatic identification process to determine the specific communication protocol type used by the industrial data acquisition device. According to the identification result, the system dynamically loads the corresponding protocol parsing module. The parsed industrial data acquisition device raw data enters the data standardization conversion link and is converted into engineering values with clear physical meaning. These standardized data are provided to the outside through a unified device control API interface. Finally, the system encapsulates these API interfaces into special MCP tools that can be directly called by agents according to the MCP specification, completing the entire construction process.

[0041] The construction system of the application specifically includes a protocol adaptation module, a data standardization module, a tool encapsulation module and a service deployment module. The protocol adaptation module includes a multi-protocol identification engine and a protocol parsing plug-in library, the data standardization module includes a data cleaning unit and a format conversion unit, the tool encapsulation module includes a tool definition unit and a tool generation unit, and the service deployment module includes a tool registration unit and a version management unit.

[0042] The protocol adaptation module automatically identifies the communication protocol type of the industrial data acquisition equipment through a multi-protocol identification engine. The identification process follows a preset protocol detection sequence, and attempts industrial commonly used protocols such as Modbus, OPC UA, MQTT, and the like in turn. After identification is completed, the system dynamically loads corresponding protocol analysis plug-ins from the plug-in library according to the communication protocol type. These plug-ins are designed using a unified pluggable architecture, and support hot loading and dynamic updating. The protocol analysis plug-in library is a pluggable component set (local storage library) that has pre-stored Modbus, OPC UA, MQTT, and other industrial protocol analysis codes. The system dynamically calls the corresponding analysis code according to the identified communication protocol type, and converts the raw data of the industrial data acquisition equipment into a standardized format.

[0043] The data standardization module performs secondary processing and refinement on the raw data of the industrial data acquisition equipment from the protocol adaptation module through its internal data cleaning unit and format conversion unit. The processing process is as follows: first, the data cleaning unit checks, filters, and repairs the raw data of the industrial data acquisition equipment to eliminate invalid values and abnormal points, and ensures data quality and reliability; then, the format conversion unit converts the cleaned raw data values into standard values with clear engineering significance and unified dimensions according to the predefined device data point table and mapping rules, and finally outputs a structured data format that is uniformly used by the system.

[0044] The tool packaging module internally has a tool definition unit and a tool generation unit, which systematically convert the unified structured data interface provided by the data standardization module into a special tool that conforms to the MCP specification. The packaging process is as follows: first, the tool definition unit performs prototype design of the tool, and defines the name, description, input parameters, and output data format of the tool according to the MCP specification for the standardized device function; then, the tool generation unit generates executable tool code based on this definition through the MCP protocol adapter, which embeds the calling logic for the back-end standardized API, and finally outputs a complete standardized tool that can be loaded by the MCP server and called by the intelligent agent. The MCP protocol adapter is responsible for protocol conversion, and does not need to rewrite the back-end code to adapt to MCP. Only an API description document needs to be provided to automatically generate an MCP-compatible tool.

[0045] The service deployment module is responsible for deploying the standardized MCP tool produced by the tool encapsulation module to the MCP server in a safe and orderly manner through its internal tool registration unit and version management unit, and managing the full life cycle of the MCP tool. The deployment process is as follows: first, the tool registration unit registers the tool instance to the server runtime environment and updates the service discovery mechanism, so that the agent can perceive and call the new tool immediately; then, the version management unit establishes a version tracking strategy for each tool, supports multi-version coexistence and gray release, ensures smooth and controllable tool upgrade process, and finally builds a stable, reliable and sustainable evolving MCP tool library dedicated to industrial data acquisition.

[0046] The specific implementation process is as follows:

[0047] I. Modeling of industrial data acquisition equipment and standardization of interface

[0048] 1) First, a unified functional description model of industrial data acquisition equipment is established, and the industrial data acquisition equipment function vector is defined , wherein is the industrial data acquisition equipment identifier, is the industrial data acquisition equipment type, is the industrial data acquisition equipment parameter set, is the industrial data acquisition equipment instruction set; the industrial data acquisition equipment parameter set is specifically defined as: , wherein is the data type (such as integer, floating point), is the device data unit (such as ℃, MPa), is the numerical range of device data (such as [0, 100]), is the data update frequency of device data (such as 100 ms); the device instruction set includes operations supported by the device, such as "read", "write", "reset", etc.

[0049] 2) Based on the industrial data acquisition equipment function vector, a standardized device access interface is developed in Java language and in accordance with a self-defined interface design specification, including a data access interface, a data analysis interface and a device control interface. The data access interface is used to read industrial data acquisition equipment parameters in real time, the data analysis interface is used to query the running status of industrial data acquisition equipment, and the device control interface is used to execute industrial data acquisition equipment instructions. The design specification is a structured mapping of the industrial data acquisition equipment function vector. The interface design adopts the RESTful style and uses the JSON format for data exchange.

[0050] II. Protocol identification and data conversion

[0051] 1) Establish a multi-protocol identification mechanism, automatically identify the device communication protocol type through the protocol matching algorithm. The protocol matching algorithm is realized by calculating the similarity between the industrial data acquisition device response data and the protocol feature library. The similarity score The calculation formula is:

[0052] (1)

[0053] Wherein is the industrial data acquisition device response feature, is the protocol standard feature, is the total number of device response features or protocol standard features.

[0054] The protocol analysis plug-in library contains unique features of each communication protocol, such as Modbus protocol function code features (such as 03H read holding register), OPC UA protocol node identification features (such as ns=2;i=1) and MQTT protocol topic naming features (such as " / device / status"). In the process of identifying the type of device communication protocol, a safety protection mode is adopted to avoid affecting critical devices through a device whitelist system. The whitelist configuration is: , wherein is the allowed protocol list, is the number of retries, is the timeout duration.

[0055] 2) According to the identified communication protocol type, dynamically load the corresponding protocol analysis plug-in, and convert the industrial data acquisition device raw data into standardized data points; the conversion process uses a data mapping function:

[0056] (2)

[0057] Wherein is the industrial data acquisition device raw data, is the offset, is the scaling ratio, is the reference value, these parameters are dynamically configured according to the type of industrial data acquisition device, the configuration rules are:

[0058] (3)

[0059] (4)

[0060] (5)

[0061] Wherein and are the maximum and minimum values of the engineering values after data standardization (such as [0, 100]), and The maximum value and minimum value of the original value of the industrial data acquisition equipment (such as [0, 255]).

[0062] Through data mapping, the original data of the industrial data acquisition equipment of different communication protocols is unified into engineering values with physical meaning.

[0063] III. Tool packaging

[0064] The standardized data is provided through the device access interface, and the dedicated MCP tool is packaged based on the MCP tool specification and the MCP protocol adapter. The tool packaging process defines the tool descriptor :

[0065]

[0066] wherein is the tool name (such as "read_device_data"), is the input parameter of the MCP tool, is the output data format of the MCP tool. The inputs parameter is specifically defined as:

[0067] inputs = {"device_id": "string","operation": ["read", "write"],"parameters": {"address": "string","value": "number"}}

[0068] The tool generation unit automatically generates tool code according to the tool descriptor, and injects data access logic to ensure that the tool can correctly call the backend interface.

[0069] IV. Service deployment and operation

[0070] 1) Register the packaged tool to the MCP server to form an industrial data acquisition special tool library. The tool registration process adopts a version control mechanism, each tool contains a version number (such as "v1.0"), supports multiple versions coexistence and gray release. The service deployment module publishes the tool metadata to the MCP server through the tool registration unit, and updates the tool index, for dynamic discovery and calling by the agent.

[0071] 2) When the system is running, the protocol adaptation module continuously monitors the device connection state, and when a new device is detected, automatically triggers the protocol identification and data standardization process, and generates the corresponding MCP tool. The tool library supports dynamic expansion, and new protocol analysis plug-ins can be registered through the plug-in library without modifying the core code.

[0072] The special MCP tool library for industrial data acquisition equipment is an extensible tool set constructed based on standardized tool descriptors, and the input of the tool library is the standardized equipment data and services through the backend service layer, and the output is a series of industrial data acquisition special MCP tools that can be directly understood and called by intelligent agents. The construction method of the tool library fundamentally solves the core challenges of the lack of special MCP tools in the field of industrial data acquisition and the fragmentation of protocol analysis through a unified equipment function model and protocol adaptation engine; the standardized tool encapsulation specification can quickly generate accurate control and data acquisition tools for different equipment, promoting the standardization and large-scale application of industrial agent technology. The special MCP tool library constructed by the application plays a key role in realizing reliable and accurate interaction and control of intelligent agents in complex industrial sites.

[0073] Based on the same inventive concept as the special MCP tool library construction method for industrial data acquisition equipment, the present application also provides an electronic device including one or more processors and one or more memories, the memories storing computer readable code, wherein the computer readable code, when executed by the one or more processors, performs the implementation of the special MCP tool library construction method for industrial data acquisition equipment. The memory can include a non-volatile storage medium and an internal memory; the non-volatile storage medium can store an operating system and computer readable code. The computer readable code includes program instructions that, when executed, cause the processor to perform any special MCP tool library construction method for industrial data acquisition equipment. The processor is used to provide computing and control capabilities to support the operation of the entire electronic device. The memory provides an environment for the computer readable code in the non-volatile storage medium to run, and the computer readable code, when executed by the processor, causes the processor to perform any special MCP tool library construction method for industrial data acquisition equipment.

[0074] It should be understood that the processor can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0075] The computer readable storage medium can be an internal storage unit of the electronic device, such as a hard disk or a memory of the computer device. The computer readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc.

[0076] The embodiments are preferred embodiments of the present application, but the present application is not limited to the above embodiments. Any obvious improvement, replacement or modification made by those skilled in the art without departing from the essential content of the present application shall fall within the protection scope of the present application.

Claims

1. A method for constructing a dedicated MCP tool library for industrial data acquisition equipment, characterized in that: Establish a unified functional description model for data acquisition equipment in industrial settings and define the functional vector of the data acquisition equipment; Develop standardized device access interfaces based on the aforementioned device function vectors; The communication protocol type of the data acquisition device is identified by a protocol matching algorithm. Based on the identified communication protocol type, the corresponding protocol parsing plugin is loaded to convert the raw data of the data acquisition device into standardized data. The standardized data is provided as a service through the device access interface, and a dedicated MCP tool is encapsulated based on the MCP tool specification and MCP protocol adapter. The packaged MCP tools are registered to the MCP server to form a dedicated tool library for industrial data acquisition.

2. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 1, characterized in that, The device function vector ,in, For data acquisition device identifiers, For data acquisition equipment type, This is a set of parameters for data acquisition equipment. This refers to the set of instructions for data acquisition equipment.

3. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 2, characterized in that, The set of parameters of the data acquisition equipment for: ,in, The tool name. For data types, For data units, For the numerical range, This refers to the data update frequency.

4. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 1, characterized in that, The protocol matching algorithm identifies the communication protocol type by calculating the similarity between the response data of the industrial data acquisition equipment and the protocol feature library. The formula for calculating the similarity is: ,in For the response characteristics of industrial data acquisition equipment, As a feature of the protocol standard, This represents the total number of device response characteristics or protocol standard characteristics.

5. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 1, characterized in that, The raw data from the data acquisition equipment is converted into standardized data. The conversion process uses a data mapping function. ,in, Raw data from industrial data acquisition equipment. This is the offset. For scaling ratio, This is the baseline value.

6. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 5, characterized in that, The parameters in the data mapping function are dynamically configured according to the data acquisition device type, and the configuration rules are as follows: , , ,in, and These are the maximum and minimum values ​​of the standardized engineering data. and These represent the maximum and minimum values ​​of the original data from industrial data acquisition equipment.

7. The method for constructing a dedicated MCP tool library for industrial data acquisition equipment according to claim 1, characterized in that, The tool descriptor is defined in the MCP-based tool encapsulation process: Where name is the tool name, inputs are the input parameters, and outputs are the output data format.

8. A system for implementing the method for constructing a dedicated MCP tool library for industrial data acquisition equipment as described in any one of claims 1-7, characterized in that, include: The protocol adaptation module is used to identify the communication protocol type of industrial data acquisition equipment and load the corresponding protocol parsing plugin; The data standardization module converts raw data from industrial data acquisition equipment into standardized data. The tool encapsulation module encapsulates specialized tools based on standardized data interfaces and in accordance with the MCP specification. The service deployment module registers the packaged tools to the MCP server to form a dedicated tool library.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is used to execute the computer program and, in executing the computer program, implement the method for constructing a dedicated MCP tool library for industrial data acquisition equipment as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the method for constructing a dedicated MCP tool library for industrial data acquisition equipment as described in any one of claims 1-7.

Images