A service discovery method and system based on machine learning in BMS communication

Abstract

The present invention relates to technical fields such as machine learning and BMS board protocol communication, and provides a service discovery method and system based on machine learning in BMS communication. A protocol description library supporting updates is created, and the protocol description library includes a protocol description and a functional interface description. A dynamic detection module is built in to send a detection package to a lower computer. The lower computer returns response information according to the detection package, and the response information is input into a trained machine learning model. The machine learning model intelligently analyzes the response information to identify a target protocol required for communication with the lower computer, and matches the target protocol with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol to generate configuration parameters and interface codes required for communication with the lower computer, thereby reducing the difficulty and cost of system integration, improving the protocol compatibility performance in service discovery, improving the flexibility of service discovery, and improving development and maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to technical fields such as machine learning and BMS board protocol communication, and in particular to a service discovery method and system based on machine learning in BMS communication. Background Art

[0002] In industrial automation, Internet of Things (IoT) and distributed systems, host computers (such as control centers, host computers, etc.) usually need to communicate with various types of lower computers (such as sensors, actuators, smart devices, etc.). Since different lower computers may use different communication protocols and functional interfaces, the host computer faces complexity and compatibility issues when integrating and managing these devices. Specifically in the field of BMS board protocol communication, a host computer needs to connect to different models of BMS boards for communication, so it is necessary to be able to identify and discover the specific BMS board protocol in the BMS protocol group formed by different models of BMS board protocols. At present, the discovery technology for the specific BMS board protocol in the BMS protocol group mainly adopts the project inquiry method, that is, the host computer sends an instruction to inquire about the project code of the BMS board. After the BMS board responds to the project code, the host computer discovers the configuration information of the project code according to the project code. These configuration information reflect the function set supported by the BMS board, so that the host computer can display the function page according to the function set supported by the BMS board and communicate with the functions in the function set. However, in the project inquiry method, the system integration of the upper computer communicating with various types of lower computers is difficult and costly, and the protocol compatibility performance is low, resulting in poor flexibility in service discovery and low development and maintenance efficiency.

[0003] In summary, the existing machine learning-based service discovery technology in BMS communication has technical problems such as difficult system integration, high cost, low protocol compatibility, poor service discovery flexibility, and low development and maintenance efficiency. Summary of the invention

[0004] In view of the deficiencies in the above-mentioned prior art, the present invention provides a service discovery method and system based on machine learning in BMS communication, so as to reduce the difficulty and cost of system integration, improve the protocol compatibility in service discovery, improve the flexibility of service discovery, and improve the efficiency of development and maintenance.

[0005] In a first aspect, the present invention provides a service discovery method based on machine learning in BMS communication, comprising:

[0006] Creating a protocol description library that supports updates, the protocol description library including protocol descriptions of different BMS communication protocols used when different BMS boards are used as slave computers and functional interface descriptions of the different BMS communication protocols;

[0007] A built-in dynamic detection module is provided, wherein the dynamic detection module sends a detection packet to the lower computer, wherein the detection packet has a basic support query for the different BMS communication protocols, and the lower computer returns a response information according to the basic support query;

[0008] Inputting the response information into a trained machine learning model, wherein the machine learning model performs intelligent analysis on the response information to identify the target protocol required for communication with the lower computer;

[0009] The target protocol is matched with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol, and the configuration parameters and interface code required for communication with the lower computer are generated according to the target protocol description and the functional interface description adapted to the target protocol.

[0010] In a second aspect, the present invention provides a service discovery system based on machine learning in BMS communication, comprising:

[0011] The host computer runs the machine learning-based service discovery method in the BMS communication described above;

[0012] The lower computer communicates with the upper computer through the service discovery method based on machine learning in the above-mentioned BMS communication.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] The present invention provides a service discovery method and system based on machine learning in BMS communication. The method and system create a protocol description library that supports updates. The protocol description library includes protocol descriptions of different BMS communication protocols used when different BMS boards are used as lower computers and functional interface descriptions of the different BMS communication protocols. A dynamic detection module is built in. The dynamic detection module sends a detection package to the lower computer. The detection package has basic support inquiries for the different BMS communication protocols. The lower computer returns response information according to the basic support inquiries. The response information is input into a trained machine learning model. The machine learning model intelligently analyzes the response information to identify a target protocol required for communication with the lower computer. The target protocol is matched with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol. According to the target protocol description and the functional interface description adapted to the target protocol, configuration parameters and interface codes required for communication with the lower computer are generated, thereby reducing the difficulty and cost of system integration, improving the protocol compatibility in service discovery, improving the flexibility of service discovery, and improving development and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their description are used to explain the present invention and do not constitute an improper limitation of the present invention. Some specific embodiments of the present invention will be described in detail in an exemplary and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings:

[0016] Figure 1 It is a flow chart of a service discovery method based on machine learning in BMS communication according to an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram of communication between a host computer and a slave computer according to an embodiment of the present invention;

[0018] Figure 3 It is a schematic diagram of the architecture of the host computer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only an embodiment of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0020] Embodiment 1

[0021] See also Figure 1-Figure 3, this embodiment provides a service discovery method based on machine learning in BMS communication, including step S101, step S102, step S103 and step S104. Among them, step S101, step S102, step S103 and step S104 can be run on the upper computer to implement the service discovery method based on machine learning in BMS communication, by creating a protocol description library that supports updates, the protocol description library includes protocol descriptions of different BMS communication protocols used when different BMS boards are used as lower computers and functional interface descriptions of the different BMS communication protocols, and a built-in dynamic detection module, the dynamic detection module sends a detection packet to the lower computer, the detection packet has basic support inquiries for the different BMS communication protocols, the lower computer returns response information according to the basic support inquiries, and the The response information is input into a trained machine learning model, and the machine learning model performs intelligent analysis on the response information to identify the target protocol required for communication with the lower computer, and matches the target protocol with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol, and generates configuration parameters and interface codes required for communication with the lower computer based on the target protocol description and the functional interface description adapted to the target protocol, thereby reducing the difficulty and cost of system integration, improving the protocol compatibility in service discovery, improving the flexibility of service discovery, and improving development and maintenance efficiency.

[0022] See also Figure 3 , Figure 3 It is shown in a schematic diagram of the architecture of the host computer in this embodiment. It should be noted that the host computer, as the execution subject of all or part of the steps in the service discovery method based on machine learning in BMS communication, can execute steps S101, S102, S103 and S104 in this embodiment, and can also run part or all of the steps of the method involved below. Figure 3 It can be known that the host computer can be an information processing device including a memory, a processor and a network interface that are interconnected through a system bus. Those skilled in the art can understand that the host computer here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.

[0023] See also Figure 1-Figure 3, a service discovery method based on machine learning in BMS communication, comprising:

[0024] Step S101, creating a protocol description library supporting updates, wherein the protocol description library includes protocol descriptions of different BMS communication protocols used when different BMS boards are used as slave computers and functional interface descriptions of the different BMS communication protocols; wherein different BMS boards communicate with the host computer as slave computers, for example, the slave computers include BMS board 1-BMS board N, etc.

[0025] Step S102, a dynamic detection module is built in, the dynamic detection module sends a detection packet to the lower computer, the detection packet has basic support inquiries for the different BMS communication protocols, and the lower computer returns response information according to the basic support inquiries;

[0026] Step S103, inputting the response information into a trained machine learning model, and the machine learning model performs intelligent analysis on the response information to identify the target protocol required for communication with the lower computer;

[0027] Step S104, matching the target protocol with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol, and generating configuration parameters and interface codes required for communicating with the lower computer based on the target protocol description and the functional interface description adapted to the target protocol.

[0028] It should be noted that in the prior art, due to the presence of a variety of different BMS board protocols and functional interfaces between the upper and lower computers, the upper computer needs to adapt each BMS board protocol separately. When integrating multiple protocols, it is necessary to invest a large amount of manpower and material resources to maintain the adaptation logic and code development between different devices and protocols. In order to achieve protocol adaptation communication, a project inquiry method is proposed in the prior art. However, the traditional project inquiry method is sensitive to protocol changes. Once the lower computer protocol is updated or a new model of equipment appears, the upper computer needs to redevelop or modify the adaptation code, and the compatibility is not high. In addition, it is difficult for the upper computer to adaptively perform flexible configuration and page function display according to the protocol characteristics of the lower computer, but it is highly dependent on the project code mapping relationship hard-coded in advance. When the adaptation logic changes, a large amount of manual updating of the adaptation logic is required to maintain the protocol description and functional characteristics of different models of equipment, resulting in a reduction in the development efficiency of service discovery technology. In step S101, an extensible and updateable protocol description library is first established. The library may contain protocol descriptions of different BMS communication protocols, for example, the message format, data frame structure, verification method, instruction set, data encoding method, etc. of the communication protocol. The library may also contain functional interface descriptions, which correspond to the functional characteristics of specific protocols, such as equipment status query commands, parameter configuration instructions, fault diagnosis instructions, etc., as well as the corresponding presentation and interaction logic of these functional interfaces in the host computer UI. In this embodiment, by standardizing and abstracting the descriptions of protocols and functions, the need for hard-coded adaptation of a single protocol can be reduced. When a new model of BMS board appears or an existing protocol is updated, it is only necessary to incrementally add or modify its relevant description information to the description library without large-scale reconstruction of the host computer program, thereby reducing the difficulty of system integration and improving development and maintenance efficiency.

[0029] It should also be noted that in the traditional project inquiry method, the upper computer needs to make clear inquiries for known project codes. In step S102, a detection packet is sent through the built-in dynamic detection module to make basic support inquiries to the lower computer. For example, the basic support inquiry can include inquiries about several basic protocol features (for example: Is a specific message header format supported? Is there a specific data field? Does it respond to standard query instructions? Etc.), and the lower computer's response information to the detection packet will reflect the protocol features and functional interface characteristics implemented by it. The detection here does not rely on a specific project code, but is based on a general inquiry of protocol features. When new equipment appears or the protocol is updated, there is no need to modify the query method of the upper computer. It is only necessary to use the existing detection mechanism to improve the protocol compatibility and flexibility of service discovery, because the detection package can be designed to be universal and reusable, and there is no need to customize specific query logic for each BMS model.

[0030] It should also be noted that in step S103, the response information will be input into the pre-trained machine learning model after being acquired by the dynamic detection module. The model can be trained based on the historically collected lower computer response data and the corresponding protocol description association. Through the intelligent analysis of the machine learning model, the protocol type that the lower computer may use can be automatically identified according to the lower computer response characteristics, and the response content can be pattern matched, feature extracted and classified, so as to identify the most likely target protocol. Among them, the use of machine learning models can improve the automation and intelligence of adaptation. The advantage of the model is that when new protocol features are added or the protocol is upgraded, as long as the training data and model are updated, the upper computer can adapt to the new protocol more easily. This is more flexible and efficient than the traditional method of relying on manual coding to identify the protocol, thereby reducing maintenance costs and development difficulties. At the same time, when a new protocol appears, the adaptation performance to unknown protocols can be improved by continuously improving the training data of the model, and the protocol compatibility performance can also be enhanced.

[0031] It should also be noted that in step S104, once the machine learning model identifies the target protocol, the host computer can find the corresponding protocol description and functional interface description in the protocol description library. Based on these description information, the host computer can automatically generate configuration parameters, such as communication baud rate, data verification mode, data parsing rules, etc., which can automatically adapt to the target protocol without manual adjustment. The host computer can also automatically generate interface code, which can automatically generate call logic for specific functions (such as status query, parameter setting, fault diagnosis) according to the functional interface description of the target protocol. This allows the host computer to automatically generate a code structure that matches different BMS board protocols, thereby saving a lot of human programming costs. It should be noted that when the protocol description library is continuously updated and expanded, the host computer can quickly adapt to the addition or update of new protocols through step S104, greatly improving the flexibility of service discovery, and significantly reducing the workload of writing customized communication code for specific devices, thereby improving development and maintenance efficiency.

[0032] In some preferred embodiments, the communication protocols that may be used by the different BMS boards include Modbus, Ethernet / IP and OPC UA. It should be noted that Modbus is one of the most classic and common communication protocols in the field of industrial automation. It is simple, mature and stable, and is widely used in various PLCs, sensors, actuators and intelligent devices. Many old and new devices are compatible with Modbus communication when their functions are expanded. Therefore, the introduction of the Modbus protocol can adapt to the communication habits of a large number of existing BMS boards. Ethernet / IP is an open industrial protocol based on Ethernet and is widely used in modern industrial automation systems and Internet of Things architectures. The protocol has good support in the industrial field and can meet the requirements of high-speed communication, real-time control and strong scalability. OPC UA is a new generation of industrial interconnection standards. It emphasizes cross-platform, information modeling and security features, and is generally regarded by the industry as a key communication standard leading to the era of Industry 4.0 and the Internet of Things. By supporting OPC UA, higher-level data integration, information semanticization and secure encrypted transmission can be achieved, which is conducive to flexible and scalable data interaction with host computers and cloud systems.

[0033] In some preferred embodiments, in the protocol description of the different BMS communication protocols, each protocol description includes a protocol description element and a parsing rule and an interaction process corresponding to the protocol description element. Further, the protocol description element includes a protocol identifier, version information, data format, command set, and error code. It should be noted that in this embodiment, the description elements of the communication protocol are refined into protocol identifiers, version information, data formats, command sets, and error codes, etc., which can achieve more efficient, scalable, and maintainable protocol management and adaptation. In a system that supports a variety of different BMS communication protocols, there must be a feature that quickly distinguishes the protocol type. By introducing a protocol identifier, the host computer can quickly determine which protocol is currently used by the lower device during preliminary identification to avoid confusion and misinterpretation. The protocol identifier can be used as a key index field for machine learning models, dynamic detection modules, and protocol description library retrieval, so that the host computer can efficiently retrieve the corresponding protocol description and parsing rules based on the identifier. The same type of protocol may have multiple version differences in different periods or under different manufacturer customization. Through clear version information, the host computer can select the correct data format, command set, or error code parsing strategy for different protocol versions. The protocol version information can be automatically adapted according to the firmware upgrade of the device in the later stage. There is no need to modify the code on a large scale. You only need to update the version-related information and corresponding rules in the protocol description library. The data format is the basis for unified data parsing. Different protocols often have differences in data packaging and encoding methods (such as byte order, data type, floating point format), etc. Using the data format as a descriptive element allows the host computer to have a clear definition when parsing the data, and automatically decode the data encoding of a specific protocol correctly. Moreover, when a new protocol or an existing protocol upgrade uses a different encoding method, you only need to update the data format definition in the description, and the parsing engine can automatically adapt. In addition, the command set is the core of the protocol interaction, including commands such as query status, set parameters, execute control, and obtain diagnostic information. Using the command set as an independent descriptive element can form a general "command-response" template on the host computer side. When you need to adapt a new command or modify the meaning of an existing command, you only need to update the corresponding command set definition in the protocol description library, so that the host computer can automatically generate or match the corresponding interaction process and interface code, reducing hard-coded logic. In addition, each protocol may have different definitions for errors and abnormal states. Through the error code description elements, the host computer can uniformly implement the analysis, classification and processing strategy of error information. When the lower computer returns a new error type or code, it only needs to update the corresponding error code and processing instructions in the protocol description library, so that the host computer can respond flexibly and enhance the system's fault diagnosis and adaptability.

[0034] In some preferred embodiments, the dynamic detection module sends the detection packet to the lower computer by broadcasting or actively requesting. It should be noted that the dynamic detection module sends the detection packet to the lower computer by broadcasting or actively requesting, which can adapt to diverse network topologies and device distributions. In the industrial automation and Internet of Things environment, the lower computers (such as BMS boards of different models) may be distributed in different network nodes, subnets or field buses. When the system is not completely clear about the distribution location and network address information of the lower computers, the detection information can be sent to multiple potential lower computers through broadcasting without configuring a specific target address in advance, and the devices present in the network can be found. If the upper computer knows the IP address or address range of some lower computers, it can also take the active request method to send the detection packet to a single point, and communicate with a specific lower computer more efficiently. This flexible detection method enables the system to quickly realize service discovery regardless of whether it is facing unknown device distribution or known device positioning.

[0035] In some preferred embodiments, the response information includes a protocol identifier and version information. It should be noted that the protocol identifier and version information are included in the response information, so that the protocol type and version can be quickly locked. When the upper computer receives the response from the lower computer, it is necessary to quickly determine the communication protocol and version used so as to retrieve the corresponding parsing rules and interaction processes from the protocol description library. If the response information directly includes the protocol identifier and version information, the upper computer can directly know the target protocol type and its version in the initial identification stage without the need for additional inference steps or multiple rounds of interaction.

[0036] In some preferred embodiments, the machine learning model adopts a decision tree model or a support vector machine model. It should be noted that the decision tree model or the support vector machine (SVM) model is relatively easy to train on small and medium-sized data sets, the model complexity is relatively moderate, and the training speed can usually meet the real-time or near real-time requirements of industrial sites. For the initial deployment or rapid iteration of service discovery scenarios, this is conducive to quickly establishing a basic model and speeding up the launch. In addition,

[0037] When faced with multiple protocol feature dimensions, the decision tree model can form a clear decision path through layer-by-layer division of features, thereby more intuitively classifying and identifying the characteristics of different protocols. For scenarios with high feature dimensions and complex category distribution, SVM can also obtain better classification boundaries and improve recognition accuracy through appropriate kernel function selection.

[0038] In some preferred embodiments, the interface code required for communicating with the lower computer is encapsulated with the underlying details required for communicating with the lower computer, and the underlying details are used to provide the host computer application program to call, so as to simplify function calls and data interactions. It should be noted that by encapsulating the underlying communication details, the interface code can provide a high-level abstraction layer for the host computer application. The host computer does not need to understand the details such as specific communication protocols, data formats, error handling, etc., and only needs to call the standardized method provided by the interface to realize the required function. In the present embodiment, the communication logic is separated from the application logic to make the system structure clearer. As an intermediate layer, the interface code is independent of the host computer application and the lower computer communication protocol, and such modular design facilitates the independent development, testing and maintenance of each part. In addition, directly processing multiple communication protocols and complex data interaction logic will increase the development difficulty of the host computer application. By encapsulating the interface, the host computer developers only need to pay attention to the business logic, without having to deeply understand the underlying communication details, thereby greatly simplifying the development process.

[0039] In some preferred embodiments, the protocol description library supports users or developers to manually add new protocol descriptions and functional interface descriptions to the library, and supports obtaining new protocol descriptions and functional interface descriptions from the cloud and adding them to the library. It should be noted that in industrial sites and IoT environments, equipment types and protocol standards are constantly evolving. When a new model of BMS board appears or an existing protocol is upgraded, by allowing manual addition or obtaining updates from the cloud, the new protocol can be quickly adapted without large-scale reconstruction of the host computer program, so that the system can have good dynamic scalability and sustainable evolution capabilities. In addition, in this embodiment, users or developers are allowed to enrich the protocol description library by manual addition and cloud updates, making it an open, sustainable and perfect ecosystem. Different manufacturers can quickly incorporate their own protocol standards into the library to promote standardization and interoperability.

[0040] In some further embodiments, when adding new protocol descriptions and functional interface descriptions to the library, the compatibility status of the newly added protocol descriptions and functional interface descriptions with the existing system is checked through a protocol compatibility check tool to avoid communication failures. It should be noted that the newly added protocol descriptions and functional interface descriptions may conflict with existing protocol characteristics, command sets or data formats. If not checked, it may cause confusion in protocol parsing, instruction conflicts, misreading of data fields, etc., thereby causing communication failures. In this embodiment, through the compatibility check tool, potential conflicts or problems can be discovered and resolved before the new content is officially put into use, ensuring the stability and reliability of the overall operation of the system. It should be noted that in this embodiment, the compatibility check tool can use a verification tool based on a schema / architecture description file, such as JSON Schema or XMLSchema, combined with an automated testing framework to ensure the compatibility of the newly added protocol description and functional interface description with the existing system.

[0041] Embodiment 2

[0042] See also Figure 1-Figure 3 , this embodiment provides a service discovery system based on machine learning in BMS communication, including:

[0043] A host computer runs the service discovery method based on machine learning in BMS communication described in any of the above embodiments;

[0044] The lower computer communicates with the upper computer through the service discovery method based on machine learning in the BMS communication described in any one of the above embodiments.

[0045] It should be noted that, in this embodiment, by creating a protocol description library that supports updates, the protocol description library includes protocol descriptions of different BMS communication protocols used when different BMS boards are used as lower computers and functional interface descriptions of the different BMS communication protocols, and a built-in dynamic detection module, the dynamic detection module sends a detection package to the lower computer, the detection package has basic support inquiries for the different BMS communication protocols, the lower computer returns response information based on the basic support inquiries, the response information is input into a trained machine learning model, the machine learning model performs intelligent analysis on the response information to identify the target protocol required for the communication of the lower computer, the target protocol is matched with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol, and according to the target protocol description and the functional interface description adapted to the target protocol, the configuration parameters and interface code required for communication with the lower computer are generated, thereby reducing the difficulty and cost of system integration, improving the protocol compatibility in service discovery, improving the flexibility of service discovery, and improving development and maintenance efficiency.

[0046] It should be pointed out that the above embodiments are only preferred specific implementation modes of the present invention, and the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A service discovery method based on machine learning in BMS communication, characterized in that: include: Create a protocol description library that supports updates, the protocol description library includes protocol descriptions of different BMS communication protocols used when different BMS boards are used as slave computers and functional interface descriptions of the different BMS communication protocols; the protocol description includes the message format, data frame structure, verification method, instruction set and data encoding method of the communication protocol; the functional interface description corresponds to the functional characteristics of the specific protocol, including device status query commands, parameter configuration instructions, fault diagnosis instructions and the corresponding presentation and interaction logic of the functional interface in the host computer UI; the protocol description library supports users or developers to manually add new protocol descriptions and functional interface descriptions to the library, and supports obtaining new protocol descriptions and functional interface descriptions from the cloud and adding them to the library; when adding new protocol descriptions and functional interface descriptions to the library, the protocol compatibility check tool is used to check the compatibility status of the newly added protocol descriptions and functional interface descriptions with the existing system to avoid communication failures; A built-in dynamic detection module sends a detection package to the lower computer, the detection package has basic support inquiries for the different BMS communication protocols, and the lower computer returns response information based on the basic support inquiries; the detection in the detection package does not rely on a specific project code, but is based on a general inquiry of protocol features; When new devices appear or protocols are updated, there is no need to modify the host computer's query method, just use the existing detection mechanism; The detection package is designed to be universal and reusable, and there is no need to customize specific query logic for each BMS model; Inputting the response information into a trained machine learning model, wherein the machine learning model performs intelligent analysis on the response information to identify the target protocol required for communication with the lower computer; The target protocol is matched with the protocol description and the functional interface description in the protocol description library to obtain a target protocol description and a functional interface description adapted to the target protocol, and the configuration parameters and interface code required for communication with the lower computer are generated according to the target protocol description and the functional interface description adapted to the target protocol.

2. The method for service discovery based on machine learning in BMS communication according to claim 1, characterized in that: The communication protocols that may be used by the different BMS boards include Modbus, Ethernet / IP and OPC UA.

3. The service discovery method based on machine learning in BMS communication according to claim 1, characterized in that: In the protocol descriptions of the different BMS communication protocols, each protocol description includes a protocol description element and a parsing rule and an interaction process corresponding to the protocol description element.

4. The method for service discovery based on machine learning in BMS communication according to claim 3, characterized in that: The protocol description elements include protocol identification, version information, data format, command set and error code.

5. The method for service discovery based on machine learning in BMS communication according to claim 1, characterized in that: The dynamic detection module sends the detection packet to the lower computer by broadcasting or actively requesting.

6. The method for service discovery based on machine learning in BMS communication according to claim 1, characterized in that: The response information includes a protocol identifier and version information.

7. The method for service discovery based on machine learning in BMS communication according to claim 1, characterized in that: The machine learning model adopts a decision tree model or a support vector machine model.

8. The method for service discovery based on machine learning in BMS communication according to any one of claims 1 to 7, characterized in that: The interface code required for communicating with the lower computer encapsulates the underlying details required for communicating with the lower computer, and the underlying details are used to provide to the upper computer application program for calling, so as to simplify function calling and data interaction.

9. A machine learning-based service discovery system in BMS communication, characterized in that: include: A host computer, running the service discovery method based on machine learning in BMS communication as described in any one of claims 1 to 8; The lower computer communicates with the upper computer through the service discovery method based on machine learning in BMS communication as described in any one of claims 1 to 8.

Images