A non-intrusive device edge computing method and terminal

Abstract

The application discloses a non-invasive device edge computing method and a terminal, a plurality of different containers are predefined, and different data processing functions are written for each container; different containers are combined according to a plurality of function requirements to obtain a plurality of function factories; real-time data of a device is collected in real time and preprocessed; the preprocessed real-time data is matched with corresponding function factories according to function requirements, and corresponding calculation output is carried out through data processing functions in the function factories; and the calculation output is collected into a database. Through the non-invasive mode, the data processing functions are made into components to directly call for the edge computing of the device data, the safety and complementary interference of the edge computing are greatly improved, and the effective utilization of resources is ensured.

Description

Technical Field

[0001] This invention relates to the field of edge computing technology, and in particular to a non-intrusive device edge computing method and terminal. Background Technology

[0002] With the development of network and digital technologies, users are increasingly demanding ultra-low latency and high-quality services. However, mobile devices have limited computing power and storage resources, leading to an increasingly prominent contradiction between compute-intensive applications and resource-constrained mobile devices. Cloud computing can provide end users with almost unlimited resources, but long-distance transmission can result in higher service latency.

[0003] The current approach of edge computing improves data quality by processing data closer to the data producer, eliminating the need to request responses from the cloud computing center via the network, thus greatly reducing system latency and enhancing service responsiveness.

[0004] However, it lacks a comprehensive and fair evaluation mechanism. Because edge computing deals with relatively new and wide-ranging business scenarios, a unified benchmark set to adapt to edge computing platforms across all scenarios is unlikely to emerge in the short term. Furthermore, excessive energy consumption in data centers is frequently observed during data collection: cloud computing and supercomputing rely on computing centers and data centers for support, and this over-concentration of resources consumes significant performance, resulting in substantial resource idleness due to low resource utilization. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a non-intrusive device edge computing method and terminal, which can greatly improve the security and non-interference of edge computing, while ensuring the effective utilization of resources.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A non-intrusive device edge computing method includes the following steps:

[0008] S1. Predefine multiple different containers and write different data processing functions for each container;

[0009] S2. Combining different containers according to various functional requirements yields multiple functional factories;

[0010] S3. Real-time data acquisition from the device and preprocessing;

[0011] S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory.

[0012] S5. Store the output calculation results in the database.

[0013] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0014] A non-intrusive edge computing terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0015] S1. Predefine multiple different containers and write different data processing functions for each container;

[0016] S2. Combining different containers according to various functional requirements yields multiple functional factories;

[0017] S3. Real-time data acquisition from the device and preprocessing;

[0018] S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory.

[0019] S5. Store the output calculation results in the database.

[0020] The beneficial effects of this invention are as follows: This invention provides a non-intrusive device edge computing method and terminal. By collecting real-time data from the device and performing preprocessing, accurate data is obtained through preliminary screening to improve subsequent performance. The preprocessed data is then matched with a variety of predefined data processing functions, and the matched data processing functions are directly called to perform data calculation optimization. This non-intrusive approach, by pre-compiling the data processing functions into components for direct invocation, greatly improves the security and non-interference resistance of edge computing while ensuring efficient resource utilization. Furthermore, the use of containers to encapsulate the data processing functions further ensures the security of edge computing. These containerized data processing functions are integrated into a functional factory based on commonly used functional requirements, allowing subsequent edge computing to be performed by directly calling the relevant functional factory, effectively improving the efficiency of edge computing. Attached Figure Description

[0021] Figure 1 This is an overall flowchart of a non-intrusive device edge computing method according to an embodiment of the present invention;

[0022] Figure 2 This is a flowchart illustrating a non-intrusive device edge computing method according to an embodiment of the present invention.

[0023] Figure 3 This is a schematic diagram of the structure of a non-intrusive edge computing terminal according to an embodiment of the present invention.

[0024] Label Explanation:

[0025] 1. A non-intrusive edge computing terminal; 2. Memory; 3. Processor. Detailed Implementation

[0026] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0027] Please refer to Figure 1 and Figure 2 A non-intrusive device edge computing method includes the following steps:

[0028] S1. Predefine multiple different containers and write different data processing functions for each container;

[0029] S2. Combining different containers according to various functional requirements yields multiple functional factories;

[0030] S3. Real-time data acquisition from the device and preprocessing;

[0031] S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory.

[0032] S5. Store the output calculation results in the database.

[0033] As described above, the beneficial effects of this invention are as follows: By collecting real-time data from the device and performing preprocessing, accurate data is obtained through preliminary screening to improve subsequent performance. The preprocessed data is then matched with various predefined data processing functions, allowing direct invocation of the matched functions for data optimization. This non-intrusive approach, where data processing functions are pre-built into components for direct invocation of device data edge computing, significantly enhances the security and non-interference resistance of edge computing while ensuring efficient resource utilization. Furthermore, the use of containers to encapsulate the data processing functions further ensures the security of edge computing. These containerized functions are integrated into a functional factory based on common functional requirements, enabling subsequent edge computing by directly invoking the relevant functional factory, thus effectively improving the efficiency of edge computing.

[0034] Furthermore, step S1 also includes:

[0035] Each of the functional factories includes multiple containers, and the multiple containers are arranged into a processing queue according to the corresponding functional requirements.

[0036] As can be seen from the above description, combining multiple containers into a functional factory can make edge computing more flexible, allowing for different computing methods to be combined as needed.

[0037] Further, step S4 specifically includes:

[0038] S41. Match the corresponding functional factory to the preprocessed real-time data according to each of the current functional requirements of the equipment.

[0039] S42. Input the real-time data into the corresponding functional factory for layer-by-layer calculation, and according to the order of the containers in the functional factory, use the real-time data as the input of the first container, and use the output of the first container as the input of the second container, and so on, until all containers in the functional factory have calculated the real-time data.

[0040] S43. Output the calculation results of the functional factory corresponding to each of the current functional requirements of the device.

[0041] As described above, to ensure the accuracy of edge computing results for different functional requirements, there are also computational order requirements among the multiple containers encapsulated with different data processing functions within each functional factory. That is, based on functional requirements, data is first input into one container for computation and optimization, and the output is used as input to another container, and so on, achieving layer-by-layer computation and filtering of data. Through layer-by-layer stripping, the data is finally selected and used as the final result data for storage, ensuring performance improvement. Furthermore, the computation of multiple functional factories is performed synchronously, effectively improving the security and non-interference characteristics of edge computing, allowing each functional factory to operate independently, preventing a problem in one computation line from affecting other computations in the project.

[0042] Furthermore, the preprocessing includes:

[0043] Filter out redundant and invalid data, and calibrate data with deviations.

[0044] As described above, preprocessing performs initial screening of the data to ensure its validity and further improve the performance of subsequent edge computing.

[0045] Please refer to Figure 3A non-intrusive edge computing terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0046] S1. Predefine multiple different containers and write different data processing functions for each container;

[0047] S2. Combining different containers according to various functional requirements yields multiple functional factories;

[0048] S3. Real-time data acquisition from the device and preprocessing;

[0049] S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory.

[0050] S5. Store the output calculation results in the database.

[0051] As described above, the beneficial effects of this invention are as follows: Based on the same technical concept, and in conjunction with the aforementioned non-intrusive device edge computing method, a non-intrusive device edge computing terminal is provided. This terminal collects real-time data from the device and performs preprocessing to obtain accurate data through initial screening, thereby improving subsequent performance. The preprocessed data is then matched with various predefined data processing functions, allowing direct invocation of the matched functions for data optimization. This non-intrusive approach, by pre-compiling the data processing functions into components for direct invocation of device data edge computing, significantly enhances the security and non-interference resistance of edge computing while ensuring efficient resource utilization. Furthermore, the use of containers to encapsulate the data processing functions further ensures the security of edge computing. These encapsulated functions are integrated into a functional factory based on common functional requirements, enabling subsequent edge computing by directly invoking the relevant functional factory, effectively improving the efficiency of edge computing.

[0052] Furthermore, step S1 also includes:

[0053] Each of the functional factories includes multiple containers, and the multiple containers are arranged into a processing queue according to the corresponding functional requirements.

[0054] As can be seen from the above description, combining multiple containers into a functional factory can make edge computing more flexible, allowing for different computing methods to be combined as needed.

[0055] Further, step S4 specifically includes:

[0056] S41. Match the corresponding functional factory to the preprocessed real-time data according to each of the current functional requirements of the equipment.

[0057] S42. Input the real-time data into the corresponding functional factory for layer-by-layer calculation, and according to the order of the containers in the functional factory, use the real-time data as the input of the first container, and use the output of the first container as the input of the second container, and so on, until all containers in the functional factory have calculated the real-time data.

[0058] S43. Output the calculation results of the functional factory corresponding to each of the current functional requirements of the device.

[0059] As described above, to ensure the accuracy of edge computing results for different functional requirements, there are also computational order requirements among the multiple containers encapsulated with different data processing functions within each functional factory. That is, based on functional requirements, data is first input into one container for computation and optimization, and the output is used as input to another container, and so on, achieving layer-by-layer computation and filtering of data. Through layer-by-layer stripping, the data is finally selected and used as the final result data for storage, ensuring performance improvement. Furthermore, the computation of multiple functional factories is performed synchronously, effectively improving the security and non-interference characteristics of edge computing, allowing each functional factory to operate independently, preventing a problem in one computation line from affecting other computations in the project.

[0060] Furthermore, the preprocessing includes:

[0061] Filter out redundant and invalid data, and calibrate data with deviations.

[0062] As described above, preprocessing performs initial screening of the data to ensure its validity and further improve the performance of subsequent edge computing.

[0063] This invention provides a non-intrusive edge computing method and terminal, suitable for business scenarios requiring only edge computing. The following detailed description is provided in conjunction with embodiments.

[0064] Please refer to Figure 1 Embodiment 1 of the present invention is as follows:

[0065] A non-intrusive device edge computing method, such as Figure 1 As shown, the steps include:

[0066] S1. Predefine multiple different containers and write different data processing functions for each container.

[0067] The data processing functions can be defined according to existing common needs. For example, aggregation functions can be defined for data that needs to be aggregated, weighted averaging functions can be defined for data that needs to be weighted averaged, and time calibration functions can be defined for data from different collection points to ensure that the data sources are classified to the same time. In this embodiment, the corresponding data processing functions can directly use existing functions or be defined according to their actual needs, without any restrictions. At the same time, encapsulating the data processing functions with containers can further ensure the security of edge computing.

[0068] S2. Combine different containers according to various functional requirements to obtain multiple functional factories.

[0069] Meanwhile, based on different functions, such as monthly data statistics, we can combine relevant monthly data statistics functions into a factory. When the preprocessed data calls this factory for calculation, we can obtain the desired monthly report data. Data processing functions, which will be encapsulated into containers for direct invocation, are integrated into a functional factory based on commonly used functional requirements. This allows for subsequent edge computing of the data by directly calling the relevant functional factories, effectively improving the efficiency of edge computing.

[0070] S3. Real-time data from the acquisition device is preprocessed. This preprocessing includes filtering redundant and invalid data and calibrating data with deviations. For example, the current collected by the electricity meter may have accuracy differences, which can be corrected by setting an electricity ratio. After initial screening of the data through preprocessing, more accurate data can be retrieved, ensuring data validity and further improving the performance of subsequent edge computing.

[0071] S4. Match the corresponding data processing functions to the preprocessed real-time data according to the functional requirements, and perform corresponding calculations and outputs.

[0072] S5. Store the output calculation results in the database.

[0073] In this embodiment, real-time data from the device is collected and preprocessed to obtain accurate data through preliminary screening, thereby improving subsequent performance. The preprocessed data is then matched with a variety of predefined data processing functions, which are then directly called to perform data optimization calculations. This non-intrusive approach, where the data processing functions are pre-built into components for direct invocation of edge computing on device data, greatly enhances the security and non-interference resistance of edge computing while ensuring the effective utilization of resources.

[0074] Please refer to Figure 2Embodiment two of the present invention is as follows:

[0075] A non-intrusive device edge computing method, based on the above embodiment one, in this embodiment, as follows: Figure 2 As shown, step S1 also includes:

[0076] Each functional factory contains multiple containers, and these containers are arranged into a processing queue according to their respective functional requirements.

[0077] In this embodiment, the way multiple containers are combined to form a functional factory makes edge computing more flexible, allowing for different computing methods to be combined as needed.

[0078] At the same time, for example Figure 2 As shown, in this embodiment, step S4 specifically includes:

[0079] S41. Match the corresponding functional factory to the preprocessed real-time data according to each functional requirement of the equipment.

[0080] S42. Input the real-time data into the corresponding functional factory for layer-by-layer calculation. According to the order of the containers in the functional factory, the real-time data is used as the input of the first container. After the first container performs the calculation, the output is used as the input of the second container, and so on, until all containers in the functional factory have performed the calculation on the real-time data.

[0081] S43. Output the calculation results of the functional factory corresponding to each functional requirement of the current equipment.

[0082] In this embodiment, to ensure the accuracy of edge computing results for different functional requirements, there are also computation order requirements among the multiple containers encapsulated with different data processing functions within each functional factory. That is, according to the functional requirements, the data matching them is optimized layer by layer through the containers of the functional factory. That is, the data is first input into a container for computation and optimization, and the output is used as the input of another container. This process is repeated to achieve layer-by-layer computation and filtering of the data. Through the peeling of layers, the data is finally selected and used as the final result data for storage, ensuring performance improvement.

[0083] For example, if we need to calculate the AC power efficiency of a power station, we need to use data from different devices, such as AC meters, power supply systems (PCS), and batteries. The first step is to select the data from each device that will be used in the final calculation. The second step is to segment and accumulate the PCS data, for example, by accumulating the collected real-time voltage, current, and time intervals to obtain the DC and AC input energy for the day. The third step is to compare the energy from the meters, PCS, and batteries. Following these three steps, the corresponding function containers are placed into a functional factory. After data collection, the calculations performed by this functional factory yield the desired energy efficiency. If we also need to classify this energy efficiency into levels, we can add a fourth step to this functional factory: preset a threshold. When the energy efficiency reaches the corresponding threshold, the corresponding level can be output.

[0084] Furthermore, calculations for multiple functional factories can be performed simultaneously, i.e. Figure 2 The two factories shown effectively enhance the security and complementary interference of edge computing, enabling each functional factory to operate independently, preventing a problem in one line's computing from causing problems in other parts of the project.

[0085] Please refer to Figure 3 Embodiment 3 of the present invention is as follows:

[0086] A non-intrusive edge computing terminal 1 includes a memory 2, a processor 3, and a computer program stored on the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, it implements the steps in Embodiment 1 or Embodiment 2 described above.

[0087] In summary, the non-intrusive device edge computing method and terminal provided by this invention have the following beneficial effects:

[0088] 1. It can greatly improve the security and complementary interference of edge computing, so that each functional factory can operate independently, and the failure of computing on one line will not cause problems in other computing in the project.

[0089] 2. Different data processing functions are encapsulated into containers, making the calculation more flexible and allowing different calculation methods to be combined as needed.

[0090] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A non-intrusive device edge computing method, characterized in that, Including the following steps: S1. Predefine multiple different containers and write different data processing functions for each container; the data processing functions include aggregation functions, weighted average functions, or time calibration functions; S2. Based on various functional requirements, different containers are combined to obtain multiple functional factories, and the calculations of multiple functional factories can be performed simultaneously. S3. Real-time data acquisition from the device and preprocessing; S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory. S5. Store the output calculation results in the database; Step S1 also includes: Each of the functional factories includes multiple containers, and the multiple containers are arranged into a processing queue according to the corresponding functional requirements, making edge computing more flexible and allowing different computing methods to be combined at any time according to the situation. Step S4 specifically involves: S41. Match the corresponding functional factory to the preprocessed real-time data according to each of the current functional requirements of the equipment. S42. Input the real-time data into the corresponding functional factory for layer-by-layer calculation, and according to the order of the containers in the functional factory, use the real-time data as the input of the first container, and use the output of the first container as the input of the second container, and so on, until all containers in the functional factory have calculated the real-time data. S43. Output the calculation results of the functional factory corresponding to each of the current functional requirements of the device.

2. The non-intrusive device edge computing method according to claim 1, characterized in that, The preprocessing includes: Filter out redundant and invalid data, and calibrate data with biases.

3. A non-intrusive device edge computing terminal, characterized in that, Includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the following steps: S1. Predefine multiple different containers and write different data processing functions for each container; the data processing functions include aggregation functions, weighted average functions, or time calibration functions; S2. Based on various functional requirements, different containers are combined to obtain multiple functional factories, and the calculations of multiple functional factories can be performed simultaneously. S3. Real-time data acquisition from the device and preprocessing; S4. Match the preprocessed real-time data with the corresponding functional factory according to the functional requirements, and perform corresponding calculations and outputs through the data processing function in the functional factory. S5. Store the output calculation results in the database; Step S1 also includes: Each of the functional factories includes multiple containers, and the multiple containers are arranged into a processing queue according to the corresponding functional requirements, making edge computing more flexible and allowing different computing methods to be combined at any time according to the situation. Step S4 specifically involves: S41. Match the corresponding functional factory to the preprocessed real-time data according to each of the current functional requirements of the equipment. S42. Input the real-time data into the corresponding functional factory for layer-by-layer calculation, and according to the order of the containers in the functional factory, use the real-time data as the input of the first container, and use the output of the first container as the input of the second container, and so on, until all containers in the functional factory have calculated the real-time data. S43. Output the calculation results of the functional factory corresponding to each of the current functional requirements of the device.

4. A non-intrusive edge computing terminal according to claim 3, characterized in that, The preprocessing includes: Filter out redundant and invalid data, and calibrate data with biases.

Images