CN-121441683-B - Gateway processing method and device based on heterogeneous computation and dynamic energy efficiency

CN121441683BCN 121441683 BCN121441683 BCN 121441683BCN-121441683-B

Abstract

The invention relates to the technical field of data processing and discloses a gateway processing method and device based on heterogeneous computation and dynamic energy efficiency, wherein the method comprises the steps of executing data analysis operation on real-time data of a target gateway to obtain data analysis information, and determining a processing acceleration component according to a random forest decision tree model and task type information; the method comprises the steps of generating a gateway model based on task modeling parameters, determining parallel division points in the gateway model through a critical path analysis algorithm, determining at least one subtask set of a target gateway, determining each corresponding heterogeneous computing unit based on a processing acceleration component and each subtask set, determining computing processing parameters and further generating gateway processing parameters corresponding to the target gateway. Therefore, the invention can realize the intelligent processing of the industrial gateway data, is beneficial to improving the intelligence and the efficiency of the industrial gateway data processing, and is beneficial to improving the accuracy and the reliability of the industrial gateway data processing.

Inventors

CHENG JUN
ZHU PING
Tong Qigao

Assignees

杭州景唐通信技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. A gateway processing method based on heterogeneous computing and dynamic energy efficiency, the method comprising: collecting real-time data of a target gateway, and performing data analysis operation on the real-time data to obtain data analysis information, wherein the data analysis information comprises task type information, data quantity information and data time information corresponding to the real-time data, and the task type information comprises at least one of Modbus protocol analysis information, OPC UA data encryption information and AI abnormality detection information; determining a processing acceleration component corresponding to the target gateway according to a random forest decision tree model and the task type information, wherein the processing acceleration component comprises a hard component for accelerating the processing of a specific type of task; Generating a gateway model corresponding to the target gateway based on a task modeling parameter determined in advance, determining parallel segmentation points in the gateway model through a preset critical path analysis algorithm, and splitting a processing task corresponding to the target gateway into at least one subtask set capable of being executed in parallel based on the parallel segmentation points; selecting matched heterogeneous computing units from the processing acceleration component according to the task type of each subtask set, determining heterogeneous computing units corresponding to each subtask set, and determining computing processing parameters corresponding to each heterogeneous computing unit, wherein the heterogeneous computing units comprise at least one of a CPU, an FPGA and an NPU; and generating gateway processing parameters corresponding to the target gateway according to each heterogeneous computing unit and the computing processing parameters corresponding to each heterogeneous computing unit.
2. The heterogeneous computing and dynamic energy efficiency based gateway processing method of claim 1, further comprising: Determining gateway operation parameters of the target gateway in a preset future time period according to a predetermined target energy efficiency function and a predetermined predictive control model; Generating energy efficiency optimization parameters corresponding to the gateway processing parameters based on the gateway operation parameters and a predetermined depth deterministic strategy gradient reinforcement learning model; And according to the energy efficiency optimization parameters, updating the gateway processing parameters, and controlling the target gateway to execute gateway processing operations matched with the updated gateway processing parameters.
3. The heterogeneous computing and dynamic energy efficiency based gateway processing method according to claim 2, wherein after the collecting real-time data of the target gateway, the method further comprises: Performing data extraction operation on the real-time data based on a predetermined real-time message verification parameter to obtain a verification data extraction result, and generating a data verification parameter based on the predetermined real-time message verification parameter, wherein the verification data extraction result comprises an industrial Modbus field to an MQTT message body; Performing data verification operation on the verification data extraction result according to the data verification parameters to obtain a data verification result, wherein the data verification result comprises a protocol mapping verification result and an encrypted message verification result; judging whether the data verification result meets a preset data verification condition or not; when the data verification result is judged to meet the preset data verification condition, triggering and executing the operation of executing data analysis on the real-time data to obtain data analysis information.
4. The gateway processing method based on heterogeneous computing and dynamic energy efficiency according to claim 3, wherein after determining a heterogeneous computing unit corresponding to each subtask set based on the processing acceleration component and each subtask set, and determining a computing processing parameter corresponding to each heterogeneous computing unit, the method further comprises: according to the real-time data, determining real-time sequence data corresponding to the real-time data, and determining target feature data corresponding to the real-time sequence data, wherein the target feature data comprises one or more of abnormal feature vector data and compressed alarm message data corresponding to the real-time sequence data; Determining task queue depth information corresponding to the target gateway based on a predetermined time window polling parameter, and determining an activation state corresponding to each heterogeneous computing unit based on the target feature data and the task queue depth information; and updating the calculation processing parameters corresponding to each heterogeneous calculation unit according to the activation state corresponding to each heterogeneous calculation unit.
5. The gateway processing method based on heterogeneous computing and dynamic energy efficiency according to claim 1, wherein the determining, based on the processing acceleration component and each subtask set, a heterogeneous computing unit corresponding to each subtask set, and determining a computing processing parameter corresponding to each heterogeneous computing unit, includes: For each subtask set, determining a task type parameter corresponding to the subtask set based on the processing acceleration component and the subtask set, and determining a heterogeneous computing unit corresponding to the subtask set according to the task type parameter corresponding to the subtask set and the subtask and the corresponding task type parameter; For each subtask set, determining a processing requirement parameter of the subtask set according to a heterogeneous computing unit corresponding to the subtask set and a task type parameter corresponding to the subtask set, and determining a computing processing parameter corresponding to the heterogeneous computing unit based on the processing requirement parameter of the subtask set; When the task type parameter corresponding to the subtask set comprises one or more of an encryption task type parameter, a decryption task type parameter and a protocol conversion parameter, the heterogeneous computing unit corresponding to the subtask set comprises an FPGA hardware acceleration unit, and when the task type parameter corresponding to the subtask set comprises a time sequence data reasoning task type parameter, the heterogeneous computing unit corresponding to the subtask set comprises an NPU neural network computing unit.
6. The gateway processing method based on heterogeneous computing and dynamic energy efficiency according to claim 3, wherein the determining whether the data verification result meets a preset data verification condition comprises: Determining a data environment parameter corresponding to a data verification result according to the data verification result, determining data key information of the data verification result according to the data environment parameter, and determining a safety data verification parameter according to the data environment parameter, wherein the safety data verification parameter comprises a safety starting chain verification parameter and a communication data encryption subkey parameter; generating a data verification parameter corresponding to the data verification result according to the data key information and the safety data verification parameter; Judging whether the data security degree corresponding to the data verification parameters is greater than or equal to a data security degree threshold corresponding to a preset data verification condition; when the data security degree corresponding to the data verification parameters is larger than or equal to a data security degree threshold corresponding to the preset data verification conditions, determining that the data verification result meets the preset data verification conditions; When the data security degree corresponding to the data verification parameters is smaller than the data security degree threshold corresponding to the preset data verification conditions, determining that the data verification result meets the preset data verification conditions.
7. The heterogeneous computing and dynamic energy efficiency based gateway processing method according to claim 2, wherein the performing an update operation on the gateway processing parameters according to the energy efficiency optimization parameters comprises: acquiring real-time gateway information of the target gateway, and generating dynamic adjustment energy efficiency parameters of the energy efficiency optimization parameters according to the real-time gateway information and the energy efficiency optimization parameters and combining a predetermined deep learning model; when the real-time gateway information is used for indicating that the target gateway is in an abnormal gateway state, determining abnormal gateway information corresponding to the target gateway, generating an abnormal processing parameter based on the dynamic adjustment energy efficiency parameter and the abnormal gateway information, and executing updating operation on the gateway processing parameter based on the abnormal processing parameter.
8. A gateway processing apparatus based on heterogeneous computing and dynamic energy efficiency, the apparatus comprising: The acquisition module is used for acquiring real-time data of the target gateway; The system comprises an analysis module, a data analysis module and an AI (advanced technology attachment) module, wherein the analysis module is used for executing data analysis operation on the real-time data to obtain data analysis information, wherein the data analysis information comprises task type information, data quantity information and data time information corresponding to the real-time data, and the task type information comprises at least one of Modbus protocol analysis information, OPC UA data encryption information and AI abnormality detection information; The processing acceleration component is used for accelerating the processing of the specific type of task, and comprises a hard component used for accelerating the processing of the specific type of task; the generation module is used for generating a gateway model corresponding to the target gateway based on the task modeling parameters determined in advance; The determining module is further configured to determine parallel segmentation points in the gateway model through a preset critical path analysis algorithm, and split a processing task corresponding to the target gateway into at least one subtask set that can be executed in parallel based on the parallel segmentation points; selecting matched heterogeneous computing units from the processing acceleration component according to the task type of each subtask set, determining heterogeneous computing units corresponding to each subtask set, and determining computing processing parameters corresponding to each heterogeneous computing unit, wherein the heterogeneous computing units comprise at least one of a CPU, an FPGA and an NPU; the generation module is further configured to generate gateway processing parameters corresponding to the target gateway according to each heterogeneous computing unit and the computing processing parameters corresponding to each heterogeneous computing unit.
9. A gateway processing apparatus based on heterogeneous computing and dynamic energy efficiency, the apparatus comprising: A memory storing executable program code; A processor coupled to the memory; The processor invokes the executable program code stored in the memory to perform the heterogeneous computing and dynamic energy efficiency based gateway processing method of any of claims 1-7.
10. A computer storage medium having stored thereon computer instructions which when executed by a processor implement the heterogeneous computing and dynamic energy efficiency based gateway processing method of any of claims 1-7.

Description

Gateway processing method and device based on heterogeneous computation and dynamic energy efficiency Technical Field The invention relates to the technical field of gateway intelligent processing, in particular to a gateway processing method and device based on heterogeneous computation and dynamic energy efficiency. Background The industrial gateway is key equipment in the fields of industrial automation and the Internet of things, is mainly used for efficiently acquiring, processing and transmitting various data of industrial field equipment, acquires and uploads the data such as temperature, pressure, vibration and the like of bottom equipment to a cloud or a control center through equipment such as a connecting sensor, a controller and a PLC, receives control instructions and feeds back execution conditions in real time, ensures the stability and the high efficiency of a production process, has strong instantaneity, can rapidly respond to equipment state change, timely processes fault early warning and production flow adjustment tasks, and meets the strict requirements of industrial production on the instantaneity. In the prior art, the industrial gateway system has significant problems in real-time processing and energy efficiency optimization, firstly, the traditional industrial gateway mostly adopts a single computing architecture, for example, only relies on a CPU to process data, which is easy to generate processing delay when facing complex tasks, and is difficult to meet strict requirements of industrial production on real-time performance, for example, in a high-precision manufacturing or automatic production line, rapid response is required for equipment state monitoring and fault early warning, the existing system can cause response delay due to insufficient processing speed, the production efficiency and equipment safety are influenced, secondly, the energy efficiency optimization aspect also has defects, the existing gateway has higher energy consumption when in high-load operation, the energy consumption cannot be effectively reduced when in low-load operation, and a dynamic adjustment mechanism is lacked, and the fixed energy consumption mode not only increases the operation cost, but also has adverse effects on long-term stable operation of equipment. It is important to provide a new industrial gateway processing method to improve the intelligence and reliability of gateway data processing. Disclosure of Invention The invention provides a gateway processing method and a gateway processing device based on heterogeneous computation and dynamic energy efficiency, which can realize intelligent processing of industrial gateway data, are beneficial to improving the intelligence and the efficiency of industrial gateway data processing and are beneficial to improving the accuracy and the reliability of industrial gateway data processing. The first aspect of the invention discloses a gateway processing method based on heterogeneous computation and dynamic energy efficiency, which comprises the following steps: collecting real-time data of a target gateway, and performing data analysis operation on the real-time data to obtain data analysis information, wherein the data analysis information comprises task type information, data quantity information and data time information corresponding to the real-time data, and the task type information comprises at least one of Modbus protocol analysis information, OPC UA data encryption information and AI abnormality detection information; Determining a processing acceleration component corresponding to the target gateway according to a random forest decision tree model and the task type information which are determined in advance; Generating a gateway model corresponding to the target gateway based on the task modeling parameters determined in advance, determining parallel division points in the gateway model through a preset critical path analysis algorithm, and determining at least one subtask set corresponding to the target gateway based on the parallel division points; Determining a heterogeneous computing unit corresponding to each subtask set based on the processing acceleration component and each subtask set, and determining a computing processing parameter corresponding to each heterogeneous computing unit; and generating gateway processing parameters corresponding to the target gateway according to each heterogeneous computing unit and the computing processing parameters corresponding to each heterogeneous computing unit. As an alternative embodiment, in the first aspect of the present invention, the method further includes: Determining gateway operation parameters of the target gateway in a preset future time period according to a predetermined target energy efficiency function and a predetermined predictive control model; Generating energy efficiency optimization parameters corresponding to the gateway processing parameters based on the gateway op