Energy-saving transmission method of gateway data and related device

Abstract

The application relates to an energy-saving transmission method of gateway data and a related device. The method comprises the following steps: acquiring various types of gateway data uploaded by various sensing devices in a current building space; an AI model trained based on operation and maintenance characteristics of the current building space is deployed locally in a communication gateway; according to the AI model, the operation and maintenance levels of the various types of gateway data in the building operation and maintenance process of the current building space are determined, and the various types of gateway data corresponding to the same operation and maintenance level are integrated into corresponding gateway data sets; according to the operation and maintenance levels corresponding to the gateway data sets, the transmission strategies corresponding to the gateway data sets are determined, and the gateway data sets are uploaded to a preset cloud platform according to the corresponding transmission strategies. The method can avoid low-value data from occupying communication bandwidth and gateway energy consumption, and thus the data transmission efficiency and energy-saving demand are considered.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) data transmission, and in particular to an energy-efficient method and related apparatus for transmitting gateway data. Background Technology

[0002] In the field of IoT data transmission, the aggregation and uploading of data collected by various sensing devices are involved to achieve remote monitoring and equipment operation and maintenance management. Common data transmission methods involve processing gateway data according to a fixed collection cycle and uploading it to a cloud platform. However, this method fails to differentiate the importance of different data points in the operation and maintenance process, transmitting them in a uniform manner. This results in a large amount of low-value data consuming communication bandwidth and gateway power, making it difficult to balance data transmission efficiency with energy-saving requirements. Summary of the Invention

[0003] Therefore, it is necessary to provide an energy-saving method and related apparatus for transmitting gateway data in response to the above-mentioned technical problems.

[0004] In a first aspect, this application provides an energy-saving method for transmitting gateway data, applied to a communication gateway on the edge computing side, the method comprising: Acquire various gateway data uploaded by various sensing devices within the current building space; The communication gateway is locally deployed with an AI model trained based on the operation and maintenance characteristics of the current building space. According to the AI ​​model, the operation and maintenance level corresponding to various types of gateway data in the building operation and maintenance process of the current building space is determined, and various types of gateway data corresponding to the same operation and maintenance level are integrated into a corresponding gateway data set. Based on the operation and maintenance level corresponding to each gateway data set, determine the transmission strategy corresponding to each gateway data set, and upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

[0005] Secondly, this application also provides a communication gateway for use on the edge computing side, wherein the communication gateway has an AI model trained based on the operation and maintenance characteristics of the current building space deployed locally. The communication gateway includes: The acquisition module is used to acquire various gateway data uploaded by various sensing devices in the current building space; The determination module is used to determine the operation and maintenance level corresponding to various types of gateway data in the current building space operation and maintenance process according to the AI ​​model, and to integrate various types of gateway data corresponding to the same operation and maintenance level into a corresponding gateway data set. The transmission module is used to determine the transmission strategy corresponding to each gateway data set according to the operation and maintenance level of each gateway data set, and to upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

[0006] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the above steps.

[0007] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the above steps.

[0008] The aforementioned energy-saving transmission method and related devices for gateway data firstly involve the unified acquisition of gateway data uploaded by various sensing devices within the building space at the communication gateway, thereby centralizing data from different sensing devices into a single data source. Secondly, an AI model deployed locally on the communication gateway determines the operational level of various gateway data during building maintenance and integrates data corresponding to the same operational level, thus forming a structured gateway data set based on the usage methods in operational management. Thirdly, a corresponding transmission strategy is determined based on the operational level of each gateway data set, and data is uploaded to the cloud platform according to each strategy, allowing data from different operational levels to be processed separately during transmission. Based on this, the entire technical solution enables centralized acquisition, hierarchical classification, and categorized transmission of various gateway data generated within the building space at the communication gateway, thereby avoiding low-value data from consuming communication bandwidth and gateway energy, and thus balancing data transmission efficiency and energy-saving requirements. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating an energy-saving data transmission method for gateways in one embodiment. Figure 2 This is a schematic diagram of a gateway data acquisition and transmission structure for a building operation and maintenance management scenario in one embodiment; Figure 3 This is a structural block diagram of a communication gateway in one embodiment. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0012] In one embodiment, such as Figure 1 As shown, an energy-saving method for transmitting gateway data is provided. This embodiment illustrates the application of this method to a communication gateway on the edge computing side. The method includes the following steps S101 to S103.

[0013] Step S101: Obtain various gateway data uploaded by each sensing device in the current building space.

[0014] For example, in the current operation and maintenance of building spaces, various sensors are pre-deployed to sense environmental conditions and equipment operating status. These include temperature and humidity sensors to detect changes in ambient temperature and humidity, gas sensors to monitor air quality, infrared sensors to sense personnel activity, and current or power monitoring devices to record equipment operating status. These sensors continuously sense the corresponding environmental conditions and equipment operating status at their respective monitoring locations and generate corresponding data records according to a predetermined collection cycle. Examples include temperature and humidity data generated by temperature and humidity sensors, air quality data generated by gas sensors, personnel activity detection data generated by infrared sensors, and power consumption status data generated by current or power monitoring devices.

[0015] The aforementioned data is sent to the communication gateway through the internal communication link of the current building space, enabling the communication gateway to continuously receive data from different sensing devices, identify and record various types of data according to their sources, and organize them according to the order of arrival time, so that data from the same sensing device forms a continuous data sequence, thereby forming various types of gateway data uploaded by various sensing devices in the current building space at the communication gateway.

[0016] Optionally, the communication gateway on the edge computing side is a local data access and processing node deployed in a building operation and maintenance scenario. Located at the edge of the building network, it receives and initially processes data generated by various sensing devices within the building, allowing some data processing to be completed locally without relying entirely on a remote cloud platform. In a building operation and maintenance scenario, this communication gateway typically connects to environmental monitoring equipment, energy consumption monitoring equipment, and equipment operation monitoring equipment within the building, and undertakes the function of data aggregation and uploading. This communication gateway can be a centralized communication gateway located in a low-voltage electrical room, or a regional communication gateway located in a floor power distribution room or equipment room.

[0017] Step S102: The communication gateway has an AI model trained based on the operation and maintenance characteristics of the current building space. According to the AI ​​model, the operation and maintenance level corresponding to various types of gateway data in the building operation and maintenance process of the current building space is determined, and various types of gateway data corresponding to the same operation and maintenance level are integrated into the corresponding gateway data set.

[0018] For example, the AI ​​model represents a data analysis model trained on historical data accumulated during building operations and maintenance (O&M). It identifies and classifies input data based on the operational status reflected in the input data, using methods such as neural network models or classification models. This allows the model to determine how the input data is used during O&M. After receiving gateway data from different sensing devices, the communication gateway inputs this data into the AI ​​model for identification. The AI ​​model then determines the O&M level based on the usage of various gateway data during building operations. The O&M level represents the management level corresponding to different gateway data during building O&M, such as monitoring layer data for real-time device status monitoring, management layer data for statistical analysis of energy consumption changes, and statistical layer data for recording historical operational data. After determining the O&M level, the communication gateway organizes the various gateway data according to the determination results, merging data corresponding to the same O&M level into a single gateway data set, thus forming multiple gateway data sets corresponding to different O&M levels.

[0019] For example, personnel activity detection data generated by infrared sensors, as well as some data reflecting the real-time operating status of equipment, are typically used in building operations and maintenance to check whether the current environment or equipment is operating normally. Therefore, this type of data can be classified as monitoring layer data. Power consumption status data generated by current or power monitoring devices are mainly used to reflect changes in the energy consumption of building equipment. In building operations and maintenance, this type of data is typically used for energy consumption management and operation management. Therefore, this type of data can be classified as management layer data. Temperature and humidity data generated by temperature and humidity sensors, as well as air quality data generated by gas sensors, can reflect the changing patterns of the building environment over a longer period of time after continuous collection and long-term storage. Therefore, this type of data can be classified as statistical layer data.

[0020] Step S103: Based on the operation and maintenance level corresponding to each gateway data set, determine the transmission strategy corresponding to each gateway data set, and upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

[0021] For example, since data at different operation and maintenance levels are used differently in building operation and maintenance processes, data uploads need to be processed differently based on their operation and maintenance level. For instance, for gateway datasets identified as monitoring layer data, which are used to view device or environmental status in real time, the communication gateway sets its corresponding transmission strategy to an immediate upload mode, so that this type of data is continuously sent to the remote cloud platform in its original time sequence after generation. For gateway datasets identified as management layer data, which are mainly used to analyze device energy consumption or operating conditions, the communication gateway sets its corresponding transmission strategy to a periodic upload mode, that is, the data is processed and sent uniformly at preset time intervals. For gateway datasets identified as statistical layer data, which are mainly used for long-term operation records, the communication gateway sets its corresponding transmission strategy to a centralized upload mode, that is, the data is sent uniformly after it has accumulated to a certain scale.

[0022] After determining the transmission strategy corresponding to each gateway data set, the communication gateway establishes a communication connection with the preset cloud platform through the internal communication link of the current building space, and sends the corresponding gateway data sets to the cloud platform according to the transmission strategy. The cloud platform refers to a data management and analysis platform deployed in a remote computing environment, such as a data platform for building operation and maintenance management or a building management system platform. This cloud platform receives data from the communication gateways and stores and manages it, thereby completing the process of uploading each gateway data set to the cloud platform.

[0023] Optionally, Figure 2 A schematic diagram of a gateway data acquisition and transmission structure for building operation and maintenance management scenarios is shown. The structure consists of three parts: on-site electromechanical equipment, sensors and modules, and edge computing side. It interacts with a remote cloud platform through a communication network.

[0024] The on-site electromechanical equipment includes building electromechanical equipment such as power distribution control cabinets, power distribution cabinets, diesel generator sets, water supply equipment, fire protection pipeline equipment, and fire pump sets. These devices undertake basic operational functions such as power supply, water supply, and fire protection during building operation and maintenance, and their operating status, environmental conditions, and working parameters require continuous monitoring. Furthermore, sensors and modules monitor and collect the operating status, environmental conditions, and working parameters of the on-site electromechanical equipment. These include electrical monitoring modules, status monitoring modules, temperature and humidity sensors, liquid level sensors, water pressure sensors, water immersion sensors, and network cameras. For example, electrical monitoring modules collect electrical operating parameters of power distribution control cabinets or power distribution cabinets; status monitoring modules detect equipment operating status; temperature and humidity sensors monitor the ambient temperature and humidity; liquid level and water pressure sensors monitor the status of water supply equipment or pipelines; water immersion sensors detect water accumulation around the equipment; and network cameras collect on-site image information.

[0025] Furthermore, these sensors and modules transmit the collected gateway data to the edge computing side via LoRa wireless transmission or network cables. A communication gateway is deployed on the edge computing side, containing an AI recognition and processing module. The communication gateway uniformly receives and manages gateway data from different sensors and modules, and performs data analysis and processing logic locally through the AI ​​recognition and processing module. The processed data is then uploaded to the cloud platform via LAN, WiFi, or 5G communication methods, thereby enabling remote monitoring and maintenance management of the building's electromechanical equipment.

[0026] In the aforementioned energy-saving transmission method for gateway data, in step S101, gateway data uploaded by various sensing devices within the current building space is uniformly acquired at the communication gateway, thereby forming a centralized data source for data scattered across different sensing devices. In step S102, the operation and maintenance levels of various gateway data in the building operation and maintenance process are determined based on the AI ​​model deployed locally on the communication gateway, and data corresponding to the same operation and maintenance level are integrated, thereby forming a structured gateway data set according to the usage methods in operation and maintenance management. In step S103, corresponding transmission strategies are determined based on the operation and maintenance levels corresponding to each gateway data set, and data is uploaded to the cloud platform according to each transmission strategy, thereby distinguishing and processing data at different operation and maintenance levels during transmission. Based on this, in the entire technical solution, various gateway data generated within the current building space are centrally acquired, hierarchically divided, and classified for transmission at the communication gateway, thereby avoiding low-value data from occupying communication bandwidth and gateway energy consumption, and thus balancing data transmission efficiency and energy-saving requirements.

[0027] In an exemplary embodiment, the training method of the AI ​​model includes: constructing training samples in a cloud platform based on the operation and maintenance characteristics of general building spaces, filtering the training samples based on the operation and maintenance characteristics of the current building space, and extracting training features from the filtered training samples to train a preset machine learning model to obtain an AI model suitable for the current building space.

[0028] For example, a general building space refers to a building space with a common operational structure in building operation and maintenance management scenarios, such as office buildings, commercial complexes, teaching buildings, or public service buildings. Although these buildings have different uses, they share similar operation and maintenance management methods in areas such as environmental monitoring, equipment operation monitoring, and energy consumption management. The operation and maintenance characteristics of general building spaces represent the common data usage patterns in the operation and maintenance processes of the aforementioned buildings. For example, environmental data is used for environmental status monitoring, equipment operation data is used for equipment status management, and electricity consumption data is used for energy consumption statistics management. Based on these operation and maintenance characteristics, the cloud platform organizes historical gateway data from different general building spaces according to the correspondence between data types and operation and maintenance uses, so that each data record simultaneously includes data content and its operation and maintenance level in operation and maintenance management, thereby forming training samples.

[0029] After the training samples are constructed, they are then filtered based on the current building space's operational characteristics. These characteristics represent the data focus of the specific building in actual operation and maintenance management. For example, office buildings typically focus on environmental comfort and equipment operation, while commercial buildings typically focus on personnel activity and energy consumption changes. Therefore, during the filtering process, training samples that are consistent with the current building space's operational focus are retained, while training samples with low correlation to the operational focus are removed.

[0030] Furthermore, training features are extracted from the selected training samples. These features include data source type, data change frequency, and data change duration, which reflect the characteristics of data usage. At the same time, the operation and maintenance level corresponding to each data point is retained as a sample identifier in the training samples, thus ensuring that the training samples contain both training features and corresponding operation and maintenance level identifiers. The extracted training features are then input into a preset machine learning model for training. During the training process, the model gradually forms judgment rules based on the correspondence between the training features and the operation and maintenance level identifiers, thereby obtaining an AI model suitable for the current building space.

[0031] Specifically, during training, the machine learning model learns to correlate training features with operation and maintenance (O&M) level identifiers in the training samples, gradually mapping different feature combinations to corresponding O&M levels. For example, when a training sample contains the feature combination "data source type is equipment operation monitoring equipment, data change frequency is high and data change duration is short," the corresponding O&M level is usually the monitoring layer used to view the real-time equipment operation status. Therefore, the model will gradually establish the correspondence between this feature combination and the monitoring layer during training. When a training sample contains the feature combination "data source type is electricity consumption monitoring equipment, data change frequency is periodic and change duration is long," the corresponding O&M level is usually the management layer used to statistically analyze energy consumption changes. Therefore, the model will gradually establish the correspondence between this feature combination and the management layer. When a training sample contains the feature combination "data source type is environmental monitoring equipment, data change frequency is low and change duration is long," the corresponding O&M level is usually the statistical layer used for long-term recording and historical analysis. Therefore, the model will gradually establish the correspondence between this feature combination and the statistical layer during training.

[0032] In this embodiment, training samples are constructed in the cloud platform based on the general operation and maintenance characteristics of building spaces. The training samples are then filtered in conjunction with the current operation and maintenance characteristics of the building space. This allows the training data to simultaneously reflect the general data patterns in the general building operation and maintenance process as well as the actual operation and maintenance concerns of the current building. At the same time, by extracting training features from the filtered training samples and using them to train a preset machine learning model, the model can form judgment rules based on the correspondence between data features and operation and maintenance levels. This results in an AI model that is adaptively suitable for the current building space.

[0033] In an exemplary embodiment, the deployment of the AI ​​model includes: compressing and quantizing the model parameters of the trained AI model in the cloud platform according to the computing and storage resources of the communication gateway, and deploying the compressed AI model to the local storage of the communication gateway in the form of a model file, so as to realize the lightweight deployment of the AI ​​model on the edge computing side.

[0034] For example, a communication gateway, as a data processing device deployed at the edge of a building network, has different local operating conditions than a cloud platform. For instance, a communication gateway typically only has computing resources for local data processing and storage resources for local program and data storage. These resources are limited in scale compared to server resources in a cloud platform. Therefore, the processor computing power and local storage capacity of the communication gateway must be prioritized during model deployment. Based on this, the model parameters in the AI ​​model are compressed and quantized within the cloud platform according to the processor computing power and local storage capacity of the communication gateway. The compression process involves organizing and merging a large number of parameters in the model to reduce the number of parameters and thus lower the computational burden during model operation. The quantization process involves converting the model parameters from their original high-precision values ​​into a smaller numerical representation, thereby controlling the overall storage space occupied by the model.

[0035] For example, in building operation and maintenance management scenarios, data generated by different environmental monitoring devices often have similar uses in operation and maintenance. For instance, temperature and humidity data are typically used simultaneously for environmental status monitoring. Therefore, the parameters in the model that originally corresponded separately to temperature and humidity data can be merged and organized, allowing the model to represent this type of data with a unified set of rules for environmental monitoring data. This reduces the number of parameters in the model used to represent similar operation and maintenance purposes. Another example is that parameters representing the frequency or duration of data changes in the original model are usually expressed with high-precision values. During quantization, these parameters can be converted into preset numerical ranges. For instance, the frequency of change can be converted from continuous values ​​to ranges such as high-frequency, periodic, or low-frequency changes, and the duration of change can be converted from specific durations to ranges such as short-term, moderately continuous, or long-term continuous changes. This allows the model to directly make judgments based on these ranges, thereby reducing the data usage of parameters during storage and computation.

[0036] After completing the above processing, the processed AI model is organized into a model file format that can be read by the device, such as a binary file or a structured data file. The model file is then sent to the communication gateway through the internal communication link of the current building space, so that the model file is written into the local storage space of the communication gateway. The communication gateway can then read the model file from the local storage space and use it to analyze and process gateway data during subsequent operation.

[0037] In this embodiment, as summarized above, the trained AI model is compressed and quantized by combining the computing and storage resources of the communication gateway in the cloud platform, so that the model parameter scale and parameter representation form match the local operating conditions of the communication gateway. At the same time, by deploying the AI ​​model after model parameter compression and quantization as a model file to the local storage of the communication gateway, the communication gateway can read and call the model file locally for data processing, thereby realizing the lightweight deployment of the AI ​​model on the edge computing side.

[0038] In an exemplary embodiment, based on the AI ​​model, the operation and maintenance level corresponding to various types of gateway data in the current building space is determined, and various types of gateway data corresponding to the same operation and maintenance level are integrated into a corresponding gateway data set, including steps S201 to S203.

[0039] Step S201: Based on the operation and maintenance functions of the sensing devices corresponding to various types of gateway data in the building operation and maintenance process, determine the operation and maintenance response requirements corresponding to various types of gateway data.

[0040] For example, in the building operation and maintenance process, different sensing devices undertake different operation and maintenance functions. For instance, the operation and maintenance function of current or power sensing devices is to continuously monitor the operating status of the devices so that building managers can understand changes in the operating status of the devices; the operation and maintenance function of temperature and humidity sensing devices used for environmental monitoring is to sense and record the building environment status so that building managers can understand changes in the building environment status; and the operation and maintenance function of infrared sensing devices used to detect personnel activities is to reflect the status of personnel activities so that building managers can understand changes in personnel activities.

[0041] Specifically, after the communication gateway obtains gateway data from the aforementioned sensing devices, it identifies the corresponding sensing device type based on the data source and uses the operation and maintenance functions of this type of sensing device as the basic information for data analysis. Then, it statistically analyzes the changes in this type of gateway data within a preset time range to obtain the change characteristics of this type of gateway data, including data change frequency, change duration, and change magnitude. Based on this, it performs correlation analysis between the above change characteristics and the corresponding operation and maintenance functions of the sensing devices, establishing a logical correspondence between the data change characteristics and the operation and maintenance functions. Even data from the same type of sensing device can differ in terms of change characteristics. Therefore, the operation and maintenance response requirement is not solely determined by the sensing device type, but rather by the combined operation and maintenance functions of the sensing devices and the data change characteristics (data change frequency, change duration, and change magnitude).

[0042] In other words, in a building operation and maintenance management scenario, multiple sensing devices of the same type can collect corresponding gateway data. Each sensing device corresponds to a type of gateway data, and the operation and maintenance response requirements corresponding to each type of gateway data are jointly determined by the operation and maintenance functions of the corresponding sensing device and the change characteristics of that type of gateway data. The operation and maintenance functions are used to limit the judgment range of the change characteristics of that type of gateway data, and the change characteristics are used to distinguish the specific operation and maintenance response requirements within the judgment range.

[0043] For example, under the operation and maintenance function of monitoring equipment operating status, the main focus is on the stability of equipment operation. Therefore, when the equipment operating status fluctuates significantly or changes abnormally in a short period of time, the operation and maintenance response requirement is determined to be "monitoring of abnormal changes in equipment operating status"; when the equipment operating status fluctuates normally within the operating cycle, the operation and maintenance response requirement is determined to be "monitoring of periodic changes in equipment operating status"; and when the equipment operating status remains stable over a longer period of time, the operation and maintenance response requirement is determined to be "monitoring of long-term stable changes in equipment operating status".

[0044] For example, under the operation and maintenance function of environmental status monitoring, the main focus is on the trend of environmental change. Therefore, when environmental data exceeds the preset environmental range in a short period of time, the operation and maintenance response requirement is determined to be "monitoring of abnormal changes in environmental status"; when environmental data shows diurnal or periodic changes, the operation and maintenance response requirement is determined to be "monitoring of periodic operational changes in environmental status"; and when environmental data shows a stable trend over a long period of time, the operation and maintenance response requirement is determined to be "monitoring of long-term stable changes in environmental status".

[0045] For example, under the operation and maintenance function of personnel activity monitoring, the main focus is on the space usage status. Therefore, when personnel activity data shows a concentrated change in personnel within a short period of time, the operation and maintenance response requirement is determined to be "monitoring of abnormal changes in personnel activity"; when personnel activity data shows regular changes within the working period, the operation and maintenance response requirement is determined to be "monitoring of periodic changes in personnel activity"; and when personnel activity data maintains a low level of change over a long period of time, the operation and maintenance response requirement is determined to be "monitoring of long-term stable changes in personnel activity".

[0046] Step S202: Based on the operational semantics reflected by the operational response requirements corresponding to various types of gateway data, determine the degree of operational attention for each type of gateway data in the building operation and maintenance process.

[0047] For example, the operation and maintenance response requirements corresponding to various types of gateway data are intended to indicate how this type of gateway data is used in the building operation and maintenance process. Therefore, corresponding operation and maintenance semantics can be extracted based on the operation and maintenance response requirements to represent the monitoring objects and monitoring methods corresponding to this type of gateway data in building operation and maintenance management. For example, when the operation and maintenance response requirement for a certain type of gateway data is "monitoring of abnormal changes in equipment operating status", the corresponding operation and maintenance semantics for this type of gateway data is "real-time response to equipment operating status"; when the operation and maintenance response requirement for a certain type of gateway data is "monitoring of periodic changes in equipment operating status", the corresponding operation and maintenance semantics for this type of data is "periodic statistics on equipment operating status"; when the operation and maintenance response requirement for a certain type of gateway data is "monitoring of long-term stable changes in equipment operating status", the corresponding operation and maintenance semantics for this type of gateway data is "long-term statistics on equipment operating status".

[0048] After obtaining the operational semantics corresponding to various gateway data types, the operational attention level for each type of gateway data is determined based on its usage frequency and response time requirements in building operations and maintenance management. The operational attention level indicates the level of attention given to this type of gateway data in building operations and maintenance management. For example, data corresponding to the operational semantic of "real-time response to device operating status" typically has a high usage frequency and requires a short response time in building operations and maintenance. For instance, it needs to be detected promptly when abnormal changes occur in device operating status; therefore, this type of gateway data corresponds to a high level of operational attention. Data corresponding to the operational semantic of "long-term statistics on device operating status" is typically viewed over a longer period in building operations and maintenance, with a relatively low usage frequency and more lenient response time requirements; therefore, this type of data corresponds to a low level of operational attention. Data corresponding to the operational semantic of "periodic statistics on device operating status" is typically viewed periodically in building space usage management or security management. Its usage frequency is between real-time response and long-term statistics, and its response time requirements are also at a medium level; therefore, this type of gateway data corresponds to a medium level of operational attention.

[0049] Step S203: Based on the level of operational and maintenance attention corresponding to each type of gateway data, determine the operational and maintenance level corresponding to each type of gateway data, and integrate the various types of gateway data corresponding to the same operational and maintenance level into the corresponding gateway data set.

[0050] For example, when a certain type of gateway data corresponds to a high level of operational and maintenance attention, it indicates that this type of gateway data needs to be prioritized and processed promptly during the operation and maintenance process, and therefore it is identified as monitoring layer data. When a certain type of gateway data corresponds to a medium level of operational and maintenance attention, it indicates that this type of gateway data is usually viewed periodically during the operation and maintenance management process, and therefore it is identified as management layer data. When a certain type of gateway data corresponds to a low level of operational and maintenance attention, it indicates that this type of gateway data is mainly used to record long-term changes in the building's operation and maintenance process, and therefore it is identified as statistical layer data. After completing the operational and maintenance layer determination, the communication gateway organizes the various types of gateway data according to the determination results, merging gateway data corresponding to the same operational and maintenance layer. This results in monitoring layer data being integrated into a monitoring layer data set, management layer data being integrated into a management layer data set, and statistical layer data being integrated into a statistical layer data set, thereby forming multiple gateway data sets corresponding to different operational and maintenance layers.

[0051] For example, for equipment operating status data, in the monitoring layer, there is data used to reflect changes in equipment operating status in real time, such as equipment current sudden change data, equipment power abnormal fluctuation data, or equipment start-stop status change data; in the management layer, there is data used to periodically statistically analyze equipment operating conditions, such as equipment runtime statistics, equipment load change cycle data, or equipment energy consumption cycle change data; and in the statistics layer, there is data used to record equipment operating trends over a long period of time, such as equipment long-term operating stability statistics, equipment annual operating load change data, or equipment historical cumulative energy consumption data.

[0052] For example, environmental data can be categorized as follows: in the monitoring layer, it corresponds to data reflecting abnormal environmental changes, such as data on sudden changes in indoor temperature, abnormal increases in humidity, or abnormal changes in air quality index; in the management layer, it corresponds to data used for periodic management of environmental conditions, such as daily variation statistics of indoor temperature and humidity or periodic variation data of air quality; and in the statistical layer, it corresponds to data used for long-term analysis of environmental change trends, such as seasonal variation data of indoor environment or long-term trend data of air quality.

[0053] For example, regarding the number of personnel activities, in the monitoring layer, it corresponds to data that reflects real-time personnel activity, such as data on sudden gatherings of people in a certain area or abnormal entry of people; in the management layer, it corresponds to data that periodically counts personnel usage, such as daily statistics on personnel flow in building areas or data on the frequency of personnel activities at different times; and in the statistics layer, it corresponds to data that is used for long-term analysis of space usage, such as long-term trend data on personnel flow in a region or statistical data on building space utilization.

[0054] In this embodiment, in step S201, the corresponding operation and maintenance response requirements are determined according to the operation and maintenance functions of the sensing devices corresponding to various types of gateway data, so that gateway data from different sources can form a clear response method according to their operation and maintenance purpose; in step S202, the corresponding operation and maintenance semantics are extracted according to the operation and maintenance response requirements corresponding to various types of gateway data, and the operation and maintenance attention level is determined according to the operation and maintenance semantics, so that the attention level of different gateway data in the operation and maintenance process is distinguished; in step S203, the operation and maintenance level is determined according to the operation and maintenance attention level corresponding to various types of gateway data, and data at the same operation and maintenance level are integrated to form a corresponding gateway data set; based on this, in the entire technical solution, different types of gateway data can be hierarchically divided and integrated according to their operation and maintenance response requirements and operation and maintenance attention levels.

[0055] In an exemplary embodiment, the degree of operational attention of various gateway data in the building operation and maintenance process is determined based on the operational and maintenance semantics reflected by the operational and maintenance response requirements corresponding to various types of gateway data, including steps S301 to S302.

[0056] Step S301: Based on the monitoring objects and change types reflected in the operation and maintenance semantics of each operation and maintenance response requirement, perform semantic classification on various types of gateway data to determine the operation and maintenance semantic type corresponding to each type of gateway data.

[0057] For example, the communication gateway parses the semantic structure of the operation and maintenance semantics, that is, it identifies the monitoring objects and change types contained therein. The monitoring objects refer to the building operation objects corresponding to this type of gateway data, such as equipment operation status, environmental status, or personnel activity status. The change type refers to the way the monitoring object changes over time, such as abnormal fluctuations, periodic changes, or long-term stable changes. After completing the above parsing, the operation and maintenance semantics are then categorized according to the change type, so that semantics with the same change type but different monitoring objects are grouped into the same category. For example, semantics representing abnormal fluctuations in equipment operation status, abnormal fluctuations in environmental status, and abnormal changes in personnel activity are uniformly categorized into the abnormal change type of operation and maintenance semantics. Semantics representing periodic changes in equipment operation status, periodic changes in environmental status, and periodic changes in personnel activity are uniformly categorized into the periodic change type of operation and maintenance semantics. Semantics representing long-term stable changes in equipment operation status, long-term changes in environmental status, and long-term changes in personnel activity are uniformly categorized into the long-term change type of operation and maintenance semantics.

[0058] Step S302: Based on the preset correspondence between operation and maintenance semantic types and operation and maintenance attention levels, convert the operation and maintenance semantic types corresponding to various gateway data into the corresponding operation and maintenance attention levels.

[0059] For example, to convert operational semantic types into operational concern levels, a correspondence between operational semantic types and operational concern levels needs to be established in advance. This correspondence describes the processing priority of data with different change types during building operations and maintenance. For instance, when the operational semantic type is "abnormal change," it indicates that this type of data reflects significant changes in the monitored object within a short period of time, requiring priority processing during building operations and maintenance. Therefore, in the correspondence, the abnormal change type is assigned a high level of operational concern. When the operational semantic type is "periodic change," it indicates that this type of data reflects changes in the monitored object within a certain time period. These changes are typically viewed at predetermined time intervals, so in the correspondence, the periodic change type is assigned a medium level of operational concern. When the operational semantic type is "long-term change," it indicates that this type of data reflects the change trend of the monitored object over a longer period. These changes are mainly used for long-term statistical analysis, so in the correspondence, the long-term change type is assigned a low level of operational concern.

[0060] Furthermore, instead of directly determining the level of operational attention from operational semantics, the approach first determines the operational semantic type and then converts it into the level of operational attention. The purpose of this is to structure and organize semantic information. Specifically, operational semantics typically includes two parts: the monitored object and the type of change. If the level of operational attention is directly determined based on operational semantics, a corresponding relationship needs to be established for each specific semantic type, which easily leads to scattered rules and difficulty in unified management. By first extracting the type of change and forming an operational semantic type, operational semantics with different monitored objects but the same type of change can be grouped into the same operational semantic type. Then, the level of operational attention is uniformly determined based on this operational semantic type, concentrating the judgment criteria at the change type level. This makes the semantic judgment process clearer and reduces redundant mapping relationships.

[0061] Optionally, the correspondence between the preset operation and maintenance semantic types and the level of operation and maintenance attention can be determined based on the response patterns of data in actual operation and maintenance activities during the building operation and maintenance process. Specifically, during the building operation and maintenance process, by combining historical operation and maintenance records and current operation and maintenance records, statistical analysis is performed on the triggering status and real-time processing requirements of data corresponding to different operation and maintenance semantic types in operation and maintenance management; where triggering status indicates the frequency of occurrence of this type of data used for status judgment or anomaly identification during the building operation and maintenance process, and real-time processing requirements indicate the response time limit required for operation and maintenance processing when this type of data occurs. By statistically analyzing the trigger events accumulated over a long period in historical operation and maintenance records, and combining this with the real-time processing needs reflected in current operation and maintenance records, a comprehensive evaluation is conducted on the data corresponding to each operation and maintenance semantic type. This results in operation and maintenance semantic types with high trigger frequency and short response time being associated with a higher level of operation and maintenance attention, operation and maintenance semantic types with moderate trigger frequency and response time within the normal range being associated with a medium level of operation and maintenance attention, and data with low trigger frequency and mainly used for trend analysis being associated with a lower level of operation and maintenance attention. This establishes a correspondence between operation and maintenance semantic types and the level of operation and maintenance attention.

[0062] In this embodiment, in step S301, semantic classification is performed based on the monitoring objects and change types reflected in the operation and maintenance response requirements corresponding to various types of gateway data, thereby obtaining the corresponding operation and maintenance semantic types. This allows operation and maintenance semantics with different monitoring objects but the same change type to be uniformly classified into the same category. In step S302, based on the preset correspondence between operation and maintenance semantic types and operation and maintenance attention levels, the operation and maintenance semantic types corresponding to various types of gateway data are converted into corresponding operation and maintenance attention levels. This determines the processing priority order of gateway data with different change types in the building operation and maintenance process. Based on this, in the entire technical solution, the gateway data can form a unified judgment basis through semantic classification when determining the operation and maintenance attention level, thereby completing the standardized conversion from operation and maintenance semantics to operation and maintenance attention level.

[0063] In an exemplary embodiment, the operation and maintenance level corresponding to each type of gateway data is determined according to the level of operation and maintenance attention corresponding to each type of gateway data, and the various types of gateway data corresponding to the same operation and maintenance level are integrated into a corresponding gateway data set, including steps S401 to S403.

[0064] Step S401: Based on the level of operational and maintenance attention corresponding to each type of gateway data and the data evolution stage of each type of gateway data, determine the timing of operational and maintenance intervention for each type of gateway data in the building operation and maintenance process.

[0065] For example, the communication gateway analyzes the continuous changes of various types of gateway data within a preset time range to identify their data evolution stages, where each stage represents the development status of the gateway data in the current operating cycle. For instance, when a certain type of gateway data just begins to change within the preset time range, it can be identified as the initial evolution stage; when this type of data continues to change over a period of time, it can be identified as the continuous evolution stage; when the data change reaches a stage high, it can be identified as the peak evolution stage; and when the data change gradually weakens and tends to stabilize, it is identified as the stable evolution stage.

[0066] After identifying the data evolution stages, a joint analysis is performed on each stage and its corresponding level of operational concern to determine the timing of operational intervention. This intervention timing is expressed as a time interval, such as 1 minute, 10 minutes, 30 minutes, or 12 hours, reflecting the urgency of processing or uploading this type of gateway data during building operations. Furthermore, the timing of operational intervention is not determined solely by the level of operational concern or the data evolution stage, but rather by both. Therefore, for the same level of operational concern, the timing of intervention differs across different data evolution stages; similarly, for the same data evolution stage, the timing of intervention also differs across different levels of operational concern.

[0067] For example, when a certain type of gateway data corresponds to a high level of operational and maintenance (O&M) concern, if it is in the initial evolution phase, the O&M intervention time can be set to 10 minutes to monitor changes in this type of data in advance. If it is in the continuous evolution phase, the O&M intervention time can be shortened to 1 minute to continuously track the changes. If it is in the peak evolution phase, the O&M intervention time can be set to 0 minutes, meaning immediate action is taken upon detecting a peak. As another example, when multiple types of gateway data are in the continuous evolution phase, their O&M intervention times will differ due to varying levels of O&M concern. Data with a high level of O&M concern can be intervened at 1 minute, data with a medium level of O&M concern at 30 minutes, and data with a low level of O&M concern at 12 hours. Thus, through these methods, the O&M intervention time can reflect both the evolutionary state of data changes and the level of O&M concern for different types of data during building O&M.

[0068] Step S402: Based on the timing of operation and maintenance intervention corresponding to each type of gateway data, perform hierarchical scoring on the degree of operation and maintenance attention corresponding to each type of gateway data to obtain the hierarchical score value corresponding to each type of gateway data.

[0069] For example, the timing of operational intervention can be used as a basis for refining and quantifying the level of operational attention. This involves establishing a correspondence between time intervals and scoring weights, transforming the original level of operational attention into comparable hierarchical scoring values. Specifically, different scoring weights can be assigned based on the time interval of the operational intervention. For instance, time intervals with intervention times of no more than 5 minutes can be assigned the highest scoring weight, 5 to 30 minutes a relatively high weight, 30 minutes to 2 hours a medium weight, and more than 2 hours a low weight. Based on this, the level of operational attention is used as the base scoring level, and then adjusted and calculated in conjunction with the scoring weights corresponding to the operational intervention timings to obtain the hierarchical scoring values.

[0070] For example, when multiple types of gateway data all correspond to a high level of operational concern, if the operational intervention time for one type of data is 1 minute, this time interval falls within the highest weighting range for scoring. Therefore, it is assigned the highest weight based on the high level of operational concern, resulting in a higher-level score, such as 90 points. Conversely, if the operational intervention time for another type of data, which also corresponds to a high level of operational concern, is 10 minutes, this time interval also falls within the high weighting range for scoring. Therefore, it is assigned a higher weight based on the high level of operational concern, resulting in a slightly lower-level score, such as 80 points. In this way, even with the same level of operational concern, different operational intervention times can still generate different-level score values, thus allowing for a more refined distinction in the processing priority of various types of gateway data in operational management.

[0071] Step S403: Map the hierarchical score values ​​corresponding to various types of gateway data to obtain the operation and maintenance levels corresponding to various types of gateway data, and integrate various types of gateway data corresponding to the same operation and maintenance level into the corresponding gateway data set.

[0072] For example, a hierarchical mapping relationship between scoring ranges and operational levels is pre-established. For instance, a scoring range of 85 points or higher can be mapped to the monitoring layer, a scoring range between 60 and 85 points to the management layer, and a scoring range below 60 points to the statistics layer. Based on this, the communication gateway determines its scoring range according to the corresponding hierarchical scoring value of each type of gateway data, and then determines the corresponding operational level according to the aforementioned hierarchical mapping relationship. For example, gateway data with a scoring value of 90 points is mapped to the monitoring layer, gateway data with a scoring value of 80 points is mapped to the management layer, and gateway data with a scoring value of 50 points is mapped to the statistics layer. This allows gateway data that initially have the same level of operational attention to be categorized into different operational levels based on the timing of operational intervention.

[0073] Therefore, the above technical solution does not directly determine the operation and maintenance (O&M) levels corresponding to different types of gateway data based on the level of O&M attention. Instead, it determines the O&M level by progressively processing the data evolution stage, the timing of O&M intervention, and the level score. This is because the level of O&M attention only represents an overall judgment of data characteristics and does not reflect the specific changes in the data at the current stage. By introducing data evolution stages, data with the same level of O&M attention can be distinguished according to their current stage. Furthermore, the timing of O&M intervention converts the above stage information into specific time intervals. Finally, the level score quantifies different time intervals uniformly, causing data with the same level of O&M attention to have different score results. Thus, the determination of the O&M level is not only based on the overall characteristics of the data but also reflects the changes in the data at the current stage, thereby making the hierarchical division of gateway data more refined and improving the accuracy of data hierarchical management.

[0074] In this embodiment, in step S401, the timing of maintenance intervention is determined based on the level of maintenance attention corresponding to each type of gateway data and the data evolution stage of each type of gateway data, so that the processing time of different gateway data in the building maintenance process can match the current state of the data; in step S402, the level of maintenance attention is hierarchically scored based on the timing of maintenance intervention corresponding to each type of gateway data, thereby transforming the level of maintenance attention, which originally only indicated the processing priority, into a hierarchical score value that combines the timing of maintenance intervention, so that the processing priority of different data can be compared through a unified value; in step S403, hierarchical mapping is performed based on the hierarchical score value corresponding to each type of gateway data, and the maintenance level is reliably determined, forming a corresponding gateway data set; based on this, in the entire technical solution, various types of gateway data can be hierarchically divided through the joint analysis of maintenance attention, data evolution stage, and timing of maintenance intervention, so that the gateway data has a clear hierarchical structure in the subsequent management and processing process.

[0075] In an exemplary embodiment, the transmission strategy corresponding to each gateway data set is determined according to the operation and maintenance level corresponding to each gateway data set, including step S501.

[0076] Step S501: Based on the preset matching relationship between the operation and maintenance level and the transmission strategy, perform strategy matching on the operation and maintenance level corresponding to each gateway data set to obtain the transmission strategy corresponding to each gateway data set; wherein, the matching relationship includes a static matching relationship based on the preset building operation and maintenance rules and a dynamic matching relationship based on the current communication status of the communication gateway.

[0077] For example, the static matching relationship preset based on building operation and maintenance rules represents the pre-established correspondence between operation and maintenance levels and transmission strategies during the daily operation and maintenance management of a building, according to the usage and time requirements of different types of data in operation and maintenance management. For instance, monitoring layer data used to view the real-time operating status of equipment or the activity status of personnel usually needs to maintain a high data update frequency, so in the static matching relationship, monitoring layer data corresponds to an immediate upload strategy; management layer data used to statistically analyze changes in equipment energy consumption or operating conditions is usually viewed according to a certain management cycle, so in the static matching relationship, management layer data corresponds to a periodic upload strategy; statistical layer data used to record environmental changes or long-term operating conditions of equipment is mainly used for subsequent historical analysis, so in the static matching relationship, statistical layer data corresponds to a centralized upload strategy. When performing strategy matching, the communication gateway reads the operation and maintenance level corresponding to each gateway data set, and then, based on the preset static matching relationship, directly maps the operation and maintenance level to the corresponding transmission strategy, thereby obtaining the data transmission method determined based on the building operation and maintenance rules.

[0078] For example, the dynamic matching relationship based on the current communication status of the communication gateway represents a matching method that adjusts the data transmission strategy according to the current network communication status of the communication gateway during its actual operation. Here, communication status refers to the communication conditions of the communication gateway during actual operation, such as network connection quality, network bandwidth usage, or current data transmission load. For instance, when the communication status is normal, the communication gateway uploads data according to the original transmission strategy; however, when the communication status changes, such as when network bandwidth usage is high or the communication load is heavy, the communication gateway adjusts the transmission strategy of some data sets according to the preset dynamic matching relationship. For example, it may extend the upload interval of some periodically uploaded data, or postpone the transmission of some centrally uploaded data to a period of lower communication load. Based on this, this matching method allows the data upload method to be adjusted accordingly as the current communication status of the communication gateway changes.

[0079] For example, in the actual policy matching process, the communication gateway performs initial policy matching on the operation and maintenance level corresponding to each gateway data set based on the static matching relationship established by the building operation and maintenance rules, thereby obtaining the basic transmission policy corresponding to each data set under normal operation and maintenance management conditions. Then, combined with the current communication status of the communication gateway, the above basic transmission policy is further adjusted according to the dynamic matching relationship. When the communication status is consistent with the conditions of normal operation and maintenance management (e.g., the network connection between the communication gateway and the cloud platform is stable, the current communication link bandwidth usage is at a normal level, and the data transmission load does not exceed the preset threshold), the communication gateway directly adopts the basic transmission policy obtained by the static matching relationship as the final transmission policy. When the communication status changes (e.g., the communication gateway detects that the current network bandwidth usage is high, the communication link is congested, or the data transmission load reaches the preset limit), the upload frequency or upload time of some data sets is adjusted according to the dynamic matching relationship, so that the final transmission policy reflects both the data transmission method in the building operation and maintenance rules and the communication conditions under the current communication status, thereby completing the policy matching process for each gateway data set.

[0080] In this embodiment, in step S501, according to the preset matching relationship between the operation and maintenance level and the transmission strategy, the operation and maintenance level corresponding to each gateway data set is matched with the strategy to obtain the transmission strategy corresponding to each gateway data set. Based on this, the final transmission strategy can be matched with the data transmission method in the building operation and maintenance rules and / or the communication conditions in the current communication state, thereby completing the adaptive matching process of the transmission strategy for the building operation and maintenance management scenario.

[0081] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0082] Based on the same inventive concept, this application also provides a communication gateway for implementing the energy-saving transmission method for gateway data as described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more communication gateway embodiments provided below can be found in the limitations of the energy-saving transmission method for gateway data described above, and will not be repeated here.

[0083] In one exemplary embodiment, such as Figure 3 As shown, a communication gateway is provided, which is applied to the edge computing side and has an AI model trained based on the current building space's operational characteristics deployed locally. The communication gateway includes: an acquisition module 101, a judgment module 102, and a transmission module 103, wherein: The acquisition module 101 is used to acquire various gateway data uploaded by various sensing devices in the current building space; The determination module 102 is used to determine the operation and maintenance level of various gateway data in the current building space according to the AI ​​model, and to integrate various gateway data corresponding to the same operation and maintenance level into a corresponding gateway data set. The transmission module 103 is used to determine the transmission strategy corresponding to each gateway data set according to the operation and maintenance level corresponding to each gateway data set, and upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

[0084] The modules in the aforementioned communication gateway can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0085] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above embodiments.

[0086] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above embodiments.

[0087] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods.

[0088] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0089] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for energy-efficient transmission of gateway data, characterized in that, A communication gateway applied to the edge computing side, the method comprising: Acquire various gateway data uploaded by various sensing devices within the current building space; The communication gateway is locally deployed with an AI model trained based on the operation and maintenance characteristics of the current building space. According to the AI ​​model, the operation and maintenance level corresponding to various types of gateway data in the building operation and maintenance process of the current building space is determined, and various types of gateway data corresponding to the same operation and maintenance level are integrated into a corresponding gateway data set. Based on the operation and maintenance level corresponding to each gateway data set, determine the transmission strategy corresponding to each gateway data set, and upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

2. The method according to claim 1, characterized in that, The training methods for the AI ​​model include: Training samples are constructed in the cloud platform based on the operation and maintenance characteristics of general building spaces. The training samples are then filtered according to the operation and maintenance characteristics of the current building space, and training features are extracted from the filtered training samples to train a preset machine learning model, thereby obtaining an AI model suitable for the current building space.

3. The method according to claim 1, characterized in that, The deployment methods of the AI ​​model include: Based on the computing and storage resources of the communication gateway in the cloud platform, the trained AI model is compressed and quantized, and the compressed AI model is deployed to the local storage of the communication gateway in the form of a model file, so as to realize the lightweight deployment of the AI ​​model on the edge computing side.

4. The method according to claim 1, characterized in that, The AI ​​model is used to determine the corresponding operation and maintenance level of various gateway data in the current building space during the building operation and maintenance process, and to integrate various gateway data corresponding to the same operation and maintenance level into a corresponding gateway data set, including: Based on the operation and maintenance functions of the sensing devices corresponding to various types of gateway data in the building operation and maintenance process, determine the operation and maintenance response requirements corresponding to various types of gateway data. Based on the operational semantics reflected by the operational response requirements corresponding to various types of gateway data, determine the degree of operational attention for various types of gateway data in the building operation and maintenance process; Based on the level of operational and maintenance attention corresponding to each type of gateway data, the operational and maintenance levels corresponding to each type of gateway data are determined, and the various types of gateway data corresponding to the same operational and maintenance level are integrated into the corresponding gateway data set.

5. The method according to claim 4, characterized in that, The determination of the degree of operational attention for various gateway data in the building operation and maintenance process, based on the operational and maintenance response requirements reflected by each type of gateway data, includes: Based on the monitoring objects and change types reflected in the operation and maintenance semantics of each operation and maintenance response requirement, the semantics of various gateway data are classified to determine the operation and maintenance semantic types corresponding to each type of gateway data. Based on the pre-defined correspondence between operation and maintenance semantic types and operation and maintenance attention levels, the operation and maintenance semantic types corresponding to various types of gateway data are converted into the corresponding operation and maintenance attention levels.

6. The method according to claim 4, characterized in that, The process involves determining the operational and maintenance (O&M) levels corresponding to different types of gateway data based on their respective O&M focus levels, and then integrating the various gateway data corresponding to the same O&M level into a corresponding gateway data set. This includes: Based on the level of operational and maintenance attention corresponding to each type of gateway data and the data evolution stage of each type of gateway data, determine the timing of operational and maintenance intervention for each type of gateway data in the building operation and maintenance process; Based on the timing of operation and maintenance intervention for each type of gateway data, a hierarchical score is assigned to the degree of operation and maintenance attention for each type of gateway data, resulting in a hierarchical score value for each type of gateway data. The hierarchical score values ​​corresponding to various types of gateway data are mapped hierarchically to obtain the operation and maintenance level corresponding to each type of gateway data, and the various types of gateway data corresponding to the same operation and maintenance level are integrated into the corresponding gateway data set.

7. The method according to claim 1, characterized in that, The step of determining the transmission strategy corresponding to each gateway data set based on the corresponding operation and maintenance level includes: Based on the preset matching relationship between the operation and maintenance level and the transmission strategy, the operation and maintenance level corresponding to each gateway data set is matched with the strategy to obtain the transmission strategy corresponding to each gateway data set; wherein, the matching relationship includes a static matching relationship based on the preset building operation and maintenance rules and a dynamic matching relationship based on the current communication status of the communication gateway.

8. A communication gateway, characterized in that, Applied to the edge computing side, the communication gateway locally deploys an AI model trained based on the current building space's operational characteristics; The communication gateway includes: The acquisition module is used to acquire various gateway data uploaded by various sensing devices in the current building space; The determination module is used to determine the operation and maintenance level corresponding to various types of gateway data in the current building space operation and maintenance process according to the AI ​​model, and to integrate various types of gateway data corresponding to the same operation and maintenance level into a corresponding gateway data set. The transmission module is used to determine the transmission strategy corresponding to each gateway data set according to the operation and maintenance level of each gateway data set, and to upload each gateway data set to the preset cloud platform according to the corresponding transmission strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images