Data intelligent processing method and device applied to industrial communication intelligent platform

Abstract

The application discloses a data intelligent processing method and device applied to an industrial communication intelligent platform, and comprises the following steps: obtaining to-be-processed data corresponding to a communication device, determining a data calculation task through a preset heterogeneous computing resource scheduling algorithm and the to-be-processed data; generating a data time sequence corresponding to all to-be-processed data based on the data calculation task and all to-be-processed data, performing a data processing operation on all to-be-processed data based on the data calculation task and the data time sequence, and obtaining a data processing result; analyzing the data processing result to obtain a data analysis result, generating a data optimization instruction corresponding to the communication device according to the data analysis result, and performing an optimization operation matched with the data optimization instruction on the communication device. It can be seen that the application can improve the intelligence and efficiency of data processing, and improve the accuracy and reliability of data processing.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a data intelligent processing method and apparatus applied to an industrial communication intelligent platform. Background Technology

[0002] In the field of industrial communication and data processing, with the deepening application of technologies such as artificial intelligence, big data, and edge computing in industry, industrial scenarios are placing higher demands on communication equipment. Traditional industrial communication equipment has exposed a series of significant shortcomings. Its functions are limited, supporting only a limited number of communication protocols. When faced with diverse devices and protocols in industrial scenarios, heterogeneous communication problems become prominent, making it difficult to achieve efficient and smooth communication between different devices, severely hindering the integration and collaborative operation of industrial systems. Traditional equipment generally lacks intelligent analysis capabilities, merely completing data transmission tasks without the ability to deeply mine and process data, resulting in low intelligence and efficiency in data processing, as well as low accuracy and reliability.

[0003] Therefore, it is particularly important to provide a new data processing method to improve the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing. Summary of the Invention

[0004] This invention provides a data intelligent processing method and apparatus for use in industrial communication intelligent platforms, which can improve the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing.

[0005] To address the aforementioned technical problems, the first aspect of this invention discloses a data intelligent processing method applied to an industrial communication intelligent platform, the method comprising:

[0006] The data to be processed corresponding to the communication device is obtained, and the data computing task is determined by a preset heterogeneous computing resource scheduling algorithm and the data to be processed.

[0007] Based on the data computation task and all the data to be processed, a data time series corresponding to all the data to be processed is generated. Based on the data computation task and the data time series, a data processing operation is performed on all the data to be processed to obtain a data processing result. The data processing operation is used to achieve microsecond-level time synchronization and dynamic bandwidth allocation of the communication device.

[0008] The data processing results are analyzed to obtain data analysis results. Based on the data analysis results, a data optimization instruction corresponding to the communication device is generated, and an optimization operation matching the data optimization instruction is performed on the communication device.

[0009] As an optional implementation, in the first aspect of the present invention, before analyzing the data processing results to obtain the data analysis results, the method further includes:

[0010] The communication device is monitored in real time by a preset machine vision algorithm to obtain real-time monitoring results. Based on the real-time monitoring results, it is determined whether the communication device meets the preset device operating conditions.

[0011] When it is determined that the communication device meets the preset device operating conditions, based on the real-time monitoring results, a data fusion operation is performed on the real-time monitoring results and all the data to be processed to obtain the data fusion result;

[0012] Based on the data fusion result, an update operation is performed on the data processing result, and the operation of analyzing the data processing result to obtain the data analysis result is triggered.

[0013] As an optional implementation, in the first aspect of the present invention, generating the data optimization instruction corresponding to the communication device based on the data analysis result includes:

[0014] The data interaction information of the communication device is determined, and based on the data interaction information and the data analysis results, combined with a preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated.

[0015] Based on the data interaction parameters and the preset edge computing model, a calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, a target optimization instruction that matches the data interaction result is determined in a preset optimization instruction database.

[0016] Based on all the target optimization instructions, generate the data optimization instructions corresponding to the communication device;

[0017] The data optimization instructions are used to optimize seamless data interaction between heterogeneous devices.

[0018] As an optional implementation, in the first aspect of the present invention, before performing data processing operations on all the data to be processed to obtain the data processing result, the method further includes:

[0019] Obtain the device network status of the communication device, and determine whether the device network status is used to indicate that the communication device is in offline mode.

[0020] When it is determined that the device network status indicates that the communication device is in the offline mode state, based on all the data to be processed and the preset microsecond-level TSN time synchronization network, a target preset model is determined from the pre-determined pre-stored model through a preset device fault prediction algorithm; wherein, the target preset model is an intelligent model pre-stored in the communication device; the preset device fault prediction algorithm includes the LSTM time series analysis prediction algorithm;

[0021] The step of performing data processing operations on all the data to be processed to obtain data processing results includes:

[0022] Based on the target preset model, data processing operations are performed on all the data to be processed to obtain the data processing results.

[0023] As an optional implementation, in the first aspect of the present invention, the step of performing a data fusion operation on the real-time monitoring results and all the data to be processed to obtain a data fusion result includes:

[0024] The real-time monitoring results are preprocessed by extracting multi-scale defect feature information corresponding to the real-time monitoring results through a feature pyramid network. The preprocessing operation includes one or more of adaptive white balance operation and image enhancement operation. The feature pyramid network includes an FPN network.

[0025] Data alignment is performed on the multi-scale defect feature information and all the data to be processed to obtain the data alignment result. Then, combined with the pre-determined cross-model fusion model, data fusion is performed on the multi-scale defect feature information and all the data to be processed to obtain the data fusion result.

[0026] The pre-determined cross-model fusion model includes the Transformer model.

[0027] As an optional implementation, in the first aspect of the present invention, determining the data computing task using a preset heterogeneous computing resource scheduling algorithm and the data to be processed includes:

[0028] By using a preset heterogeneous computing resource scheduling algorithm, a spatiotemporal topology map of resources corresponding to the communication device is generated, and multi-dimensional feature encoding is performed on all the data to be processed to obtain multi-dimensional data feature information corresponding to all the data to be processed.

[0029] Based on the resource spatiotemporal topology map and the multidimensional data feature information, dynamic planning parameters corresponding to all the data to be processed are generated, and it is determined whether the planning accuracy corresponding to the dynamic planning parameters meets the preset calculation accuracy conditions.

[0030] When it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined according to the dynamic programming parameters.

[0031] As an optional implementation, in the first aspect of the present invention, the step of performing data processing operations on all the data to be processed based on the data computing task and the data time series to obtain data processing results includes:

[0032] Based on the data computing task, determine the task scheduling parameters corresponding to the data computing task, and determine the data computing resource parameters according to the task scheduling parameters;

[0033] Based on the data time series, multidimensional feature sequence parameters are determined, wherein the multidimensional feature sequence parameters include time domain feature parameters, frequency domain feature parameters, and time-frequency domain feature parameters;

[0034] Based on the data calculation resource parameters and the multidimensional feature sequence parameters, data calculation and processing parameters corresponding to all the data to be processed are generated, and data processing operations are performed on all the data to be processed based on the data calculation and processing parameters to obtain data processing results.

[0035] A second aspect of the present invention discloses a data intelligent processing device applied to an industrial communication intelligent platform, the device comprising:

[0036] The acquisition module is used to acquire the data to be processed corresponding to the communication device;

[0037] The determination module is used to determine the data computing task based on a preset heterogeneous computing resource scheduling algorithm and the data to be processed;

[0038] The generation module is used to generate a data time series corresponding to all the data to be processed based on the data calculation task and all the data to be processed;

[0039] The processing module is used to perform data processing operations on all the data to be processed based on the data calculation task and the data time series, and obtain data processing results; wherein, the data processing operations are used to achieve microsecond-level time synchronization and dynamic bandwidth allocation of the communication device;

[0040] The analysis module is used to analyze the data processing results and obtain data analysis results;

[0041] The generation module is also used to generate data optimization instructions corresponding to the communication device based on the data analysis results;

[0042] An optimization module is used to perform optimization operations on the communication device that match the data optimization instructions.

[0043] As an optional implementation, in a second aspect of the invention, the apparatus further includes:

[0044] The monitoring module is used to perform real-time monitoring operations on the communication device through a preset machine vision algorithm before the analysis module analyzes the data processing results and obtains the data analysis results, so as to obtain real-time monitoring results.

[0045] The judgment module is used to determine whether the communication device meets the preset device operating conditions based on the real-time monitoring results.

[0046] The fusion module is used to perform a data fusion operation on the real-time monitoring results and all the data to be processed based on the real-time monitoring results when the judgment module determines that the communication device meets the preset device operating conditions, so as to obtain the data fusion result.

[0047] The update module is used to perform an update operation on the data processing result based on the data fusion result, and trigger the analysis module to perform the operation of analyzing the data processing result to obtain the data analysis result.

[0048] As an optional implementation, in the second aspect of the present invention, the specific method by which the generation module generates the data optimization instruction corresponding to the communication device based on the data analysis result includes:

[0049] The data interaction information of the communication device is determined, and based on the data interaction information and the data analysis results, combined with a preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated.

[0050] Based on the data interaction parameters and the preset edge computing model, a calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, a target optimization instruction that matches the data interaction result is determined in a preset optimization instruction database.

[0051] Based on all the target optimization instructions, generate the data optimization instructions corresponding to the communication device;

[0052] The data optimization instructions are used to optimize seamless data interaction between heterogeneous devices.

[0053] As an optional implementation, in a second aspect of the present invention, the acquisition module is further configured to acquire the device network status of the communication device before the processing module performs data processing operations on all the data to be processed and obtains the data processing result;

[0054] The judgment module is also used to determine whether the device network status is used to indicate that the communication device is in offline mode.

[0055] The determining module is further configured to, when the judging module determines that the device network status indicates that the communication device is in the offline mode state, determine a target preset model from a pre-determined pre-stored model based on all the data to be processed and a preset microsecond-level TSN time synchronization network, using a preset device fault prediction algorithm; wherein, the target preset model is an intelligent model pre-stored in the communication device; the preset device fault prediction algorithm includes an LSTM time series analysis prediction algorithm;

[0056] The specific methods by which the processing module performs data processing operations on all the data to be processed to obtain the data processing results include:

[0057] Based on the target preset model, data processing operations are performed on all the data to be processed to obtain the data processing results.

[0058] As an optional implementation, in a second aspect of the present invention, the fusion module performs a data fusion operation on the real-time monitoring results and all the data to be processed based on the real-time monitoring results to obtain the data fusion result. The specific methods include:

[0059] The real-time monitoring results are preprocessed by extracting multi-scale defect feature information corresponding to the real-time monitoring results through a feature pyramid network. The preprocessing operation includes one or more of adaptive white balance operation and image enhancement operation. The feature pyramid network includes an FPN network.

[0060] Data alignment is performed on the multi-scale defect feature information and all the data to be processed to obtain the data alignment result. Then, combined with the pre-determined cross-model fusion model, data fusion is performed on the multi-scale defect feature information and all the data to be processed to obtain the data fusion result.

[0061] The pre-determined cross-model fusion model includes the Transformer model.

[0062] As an optional implementation, in the second aspect of the present invention, the determining module determines the specific method of the data computing task by using a preset heterogeneous computing resource scheduling algorithm and the data to be processed, including:

[0063] By using a preset heterogeneous computing resource scheduling algorithm, a spatiotemporal topology map of resources corresponding to the communication device is generated, and multi-dimensional feature encoding is performed on all the data to be processed to obtain multi-dimensional data feature information corresponding to all the data to be processed.

[0064] Based on the resource spatiotemporal topology map and the multidimensional data feature information, dynamic planning parameters corresponding to all the data to be processed are generated, and it is determined whether the planning accuracy corresponding to the dynamic planning parameters meets the preset calculation accuracy conditions.

[0065] When it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined according to the dynamic programming parameters.

[0066] As an optional implementation, in a second aspect of the present invention, the processing module performs data processing operations on all the data to be processed based on the data calculation task and the data time series, and obtains the data processing result in the following specific ways:

[0067] Based on the data computing task, determine the task scheduling parameters corresponding to the data computing task, and determine the data computing resource parameters according to the task scheduling parameters;

[0068] Based on the data time series, multidimensional feature sequence parameters are determined, wherein the multidimensional feature sequence parameters include time domain feature parameters, frequency domain feature parameters, and time-frequency domain feature parameters;

[0069] Based on the data calculation resource parameters and the multidimensional feature sequence parameters, data calculation and processing parameters corresponding to all the data to be processed are generated, and data processing operations are performed on all the data to be processed based on the data calculation and processing parameters to obtain data processing results.

[0070] A third aspect of the present invention discloses another data intelligent processing device applied to an industrial communication intelligent platform, the device comprising:

[0071] Memory containing executable program code;

[0072] A processor coupled to the memory;

[0073] The processor calls the executable program code stored in the memory to execute the data intelligent processing method for industrial communication intelligent platforms disclosed in the first aspect of the present invention.

[0074] The fourth aspect of the present invention discloses a computer-storable medium storing computer instructions, which, when invoked, are used to execute the data intelligent processing method for industrial communication intelligent platforms disclosed in the first aspect of the present invention.

[0075] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0076] In this embodiment of the invention, the data to be processed corresponding to the communication device is acquired. A data computation task is determined using a preset heterogeneous computing resource scheduling algorithm and the data to be processed. Based on the data computation task and all the data to be processed, a data time series corresponding to all the data to be processed is generated. Based on the data computation task and the data time series, data processing operations are performed on all the data to be processed to obtain data processing results. The data processing results are analyzed to obtain data analysis results. Based on the data analysis results, a data optimization instruction corresponding to the communication device is generated, and optimization operations matching the data optimization instruction are performed on the communication device. Therefore, implementing this invention can improve the intelligence and efficiency of data processing, as well as improve the accuracy and reliability of data processing. Attached Figure Description

[0077] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0078] Figure 1 This is a flowchart illustrating a data intelligent processing method applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention.

[0079] Figure 2 This is a flowchart illustrating another data intelligent processing method for an industrial communication intelligent platform disclosed in an embodiment of the present invention.

[0080] Figure 3 This is a schematic diagram of the structure of a data intelligent processing device applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention.

[0081] Figure 4 This is a schematic diagram of another data intelligent processing device applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention.

[0082] Figure 5 This is a schematic diagram of the structure of another data intelligent processing device applied to an industrial communication intelligent platform disclosed in an embodiment of the present invention. Detailed Implementation

[0083] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0084] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0085] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0086] This invention discloses a data intelligent processing method and apparatus applied to an industrial communication intelligent platform, which can improve the intelligence and efficiency of data processing, as well as improve the accuracy and reliability of data processing. These will be described in detail below.

[0087] Example 1

[0088] Please see Figure 1 , Figure 1 This is a flowchart illustrating a data intelligence processing method applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention. Figure 1 The described data intelligent processing method for industrial communication intelligent platforms can be applied to the data intelligent processing device of the industrial communication intelligent platform, or to the communication equipment itself. The data intelligent processing device applied to the industrial communication intelligent platform can be integrated into a local server or a cloud server; this embodiment of the invention does not impose limitations. Figure 1 As shown, the data intelligent processing method applied to the industrial communication intelligent platform may include the following operations:

[0089] 101. Obtain the data to be processed corresponding to the communication device, and determine the data computing task through the preset heterogeneous computing resource scheduling algorithm and the data to be processed.

[0090] In this embodiment of the invention, optionally, the communication device can be a new generation industrial communication intelligent platform (integrated sensor and intelligent value machine), which acquires device data in industrial scenarios through a multi-protocol communication module.

[0091] In this embodiment of the invention, the processed data may include, but is not limited to, equipment operating parameters, sensor data, and production process data.

[0092] In this embodiment of the invention, optionally, the heterogeneous computing resource scheduling algorithm may include an algorithm for dynamically allocating computing resources of a multi-core ARM processor, which allocates tasks to different processing units, such as CPU cores and AI acceleration units, according to task type and priority.

[0093] In this embodiment of the invention, optionally, the data parameters of the data to be processed include the type, size, and priority of the data to be processed, and the computing resources include the computing resources of a dynamically allocated multi-core ARM processor. For example, protocol conversion tasks can be assigned to CPU cores, AI inference tasks to integrated AI acceleration units (such as NPUs), and storage management tasks to memory management units, thereby achieving efficient utilization of computing resources and increasing the utilization rate to 90%.

[0094] 102. Based on the data computation task and all data to be processed, generate the data time series corresponding to all data to be processed. Based on the data computation task and the data time series, perform data processing operations on all data to be processed to obtain the data processing results.

[0095] In this embodiment of the invention, the data processing operation is used to achieve microsecond-level time synchronization and dynamic bandwidth allocation of communication devices.

[0096] In this embodiment of the invention, optionally, the above-mentioned generation of data time series corresponding to all data to be processed based on the data computing task and all data to be processed may include:

[0097] Based on the data computing task and all the data to be processed, the hardware timestamp engine of the intelligent Ethernet switching module is used to generate the data time series corresponding to all the data to be processed in chronological order.

[0098] In this embodiment of the invention, optional data processing operations include, but are not limited to, data cleaning, feature extraction, and model inference. The bandwidth is dynamically allocated through an intelligent Ethernet switching module to ensure low-latency response of data processing, with edge-side AI inference latency less than 100ms.

[0099] 103. Analyze the data processing results, obtain the data analysis results, generate the corresponding data optimization instructions for the communication equipment based on the data analysis results, and perform optimization operations on the communication equipment that match the data optimization instructions.

[0100] In this embodiment of the invention, optionally, the data processing results are analyzed to obtain data analysis results, which include information such as equipment operating status assessment, fault prediction, and production efficiency assessment.

[0101] In this embodiment of the invention, optionally, the data optimization instruction may include one or more of the following: equipment parameter adjustment instruction, production process optimization instruction, maintenance reminder instruction, etc.; the data optimization instruction is used to optimize the operation of the communication equipment.

[0102] In this embodiment of the invention, the optimized operations may optionally include one or more of the following: equipment parameter adjustment, production process optimization, maintenance reminders, etc.

[0103] In this embodiment of the invention, optionally, the application to the industrial communication intelligent platform may include an input interface and an output interface, and the application to the industrial communication intelligent platform can be applied to a sensor-intelligent value integrated machine. The input interface module is used to receive different types of data and communication signals. The input interface includes a data interface (485, 232, DIDO, CAN) and a communication interface (one or more of wired communication such as RJ45, fiber optic, and wireless communication such as Bluetooth, WiFi, 4G, 5G, LoRa, NB-IoT). The sensor-intelligent value integrated machine module is connected to the input interface module, and its functional modules include: a storage unit using an M.2 interface for storing the system, platform, and data; a programmable platform adaptable to different industry needs; an AI model unit with intelligent functions and the ability to continuously learn and evolve; a switch unit with management functions at layer two or higher, supporting ERPS, TSN, and other functions; a protocol conversion unit supporting conversion between different protocols; and GNSS. The positioning unit is used to implement positioning functions; the output interface module, connected to the integrated sensor and intelligent value module, is used to output processed data or signals in different ways. The output interface also includes wired communication interfaces (RJ45, fiber optic) and wireless communication interfaces (Bluetooth, WiFi, 4G, 5G, LoRa, NB-IoT). Further, for example, on a production line, equipment operation data collected by various sensors, such as temperature, pressure, and speed, are transmitted to the input interface module via CAN bus, 485, or 232 data interfaces. Simultaneously, image data from some monitoring devices in the workshop is accessed via a wired RJ45 network port. After this data enters the integrated sensor and intelligent value module, it is first categorized and stored in the M.2 storage unit for subsequent retrieval and analysis. The programmable platform flexibly configures the data processing flow and rules according to the needs of different production tasks, such as the difference between the precision inspection requirements in automotive parts production and the energy consumption monitoring requirements in home appliance production. The model unit learns from stored historical and real-time data. For example, by learning the data characteristics before production equipment failures, it gradually builds a fault prediction model to identify potential faults in advance. The switch unit utilizes its Layer 2 and higher management functions to rationally allocate network bandwidth and other resources, ensuring the timely transmission of critical data. Its supported ERPS function ensures rapid switching to backup lines when fiber optic links fail, while the TSN function ensures accurate transmission of time-sensitive control signals. The protocol conversion unit plays a crucial role when different systems need to exchange data, such as converting data from different communication protocols between the PLC control system and the MES system to achieve information interoperability. The GNSS positioning unit can be used for real-time positioning of mobile production and transportation equipment, facilitating scheduling and management.The processed data and instructions are then sent out through the output interface module. For example, equipment fault warning information is sent to the terminal equipment of maintenance personnel via wireless 4G or WiFi signals, and production statistics are transmitted to the factory's data center server via wired fiber optic cable for further aggregation and decision analysis. This enables efficient information flow and collaborative work throughout the smart factory, thereby improving production efficiency and management level.

[0104] It is evident that implementation Figure 1 The described method generates a data time series corresponding to all data to be processed based on the data computing task and all data to be processed. Based on the data computing task and the data time series, it performs data processing operations on all data to be processed to obtain data processing results. It analyzes the data processing results to obtain data analysis results, and generates data optimization instructions corresponding to the communication device based on the data analysis results. It then performs optimization operations on the communication device that match the data optimization instructions. Through a preset heterogeneous computing resource scheduling algorithm, it can dynamically allocate computing resources of a multi-core ARM processor according to the type, size, and priority of the data to be processed, which helps improve data processing efficiency and reduce data latency. It also utilizes the hardware timestamp engine of the intelligent Ethernet switching module to... The system generates time series data corresponding to all data to be processed in chronological order, ensuring data accuracy and real-time performance. By analyzing the data processing results, it generates data optimization instructions for communication devices and executes optimization operations that match these instructions. This enables intelligent diagnosis and adaptive optimization of the devices, improving their reliability and safety. By integrating multi-protocol communication, edge computing, AI large models, and machine vision capabilities, it connects the entire "device-edge-cloud" link, enabling communication perception collaboration, intelligent diagnosis, and adaptive optimization. This allows the system to automatically adapt to different industrial scenarios and needs, improving its intelligence and flexibility. Consequently, it enhances the intelligence and efficiency of data processing, as well as its accuracy and reliability.

[0105] Example 2

[0106] Please see Figure 2 , Figure 2 This is a flowchart illustrating another data intelligence processing method applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention. Figure 2 The described data intelligent processing method for industrial communication intelligent platforms can be applied to the data intelligent processing device of the industrial communication intelligent platform, or to the communication equipment itself. The data intelligent processing device applied to the industrial communication intelligent platform can be integrated into a local server or a cloud server; this embodiment of the invention does not impose limitations. Figure 2As shown, the data intelligent processing method applied to the industrial communication intelligent platform may include the following operations:

[0107] 201. Obtain the data to be processed corresponding to the communication device, and determine the data computing task through the preset heterogeneous computing resource scheduling algorithm and the data to be processed.

[0108] 202. Based on the data computation task and all data to be processed, generate the data time series corresponding to all data to be processed. Based on the data computation task and the data time series, perform data processing operations on all data to be processed to obtain the data processing results.

[0109] 203. Perform real-time monitoring on the communication equipment using a preset machine vision algorithm to obtain real-time monitoring results. Based on the real-time monitoring results, determine whether the communication equipment meets the preset equipment operating conditions.

[0110] In this embodiment of the invention, optionally, the machine vision algorithm is used to detect the operating status, appearance defects, etc. of the equipment, and the real-time monitoring results include one or more of the equipment's operating status data, appearance image data, etc.

[0111] In this embodiment of the invention, optionally, the above-mentioned determination of whether the communication device meets the preset device operating conditions based on real-time monitoring results may include:

[0112] The real-time security level of the communication equipment is determined based on the real-time monitoring results, and it is determined whether the real-time security level is greater than or equal to the preset security level threshold corresponding to the equipment operating conditions.

[0113] When the real-time security level is determined to be greater than or equal to the preset security level threshold corresponding to the device operating conditions, the communication device is determined to meet the preset device operating conditions; when the real-time security level is determined to be less than the preset security level threshold corresponding to the device operating conditions, the communication device is determined to not meet the preset device operating conditions.

[0114] 204. When it is determined that the communication equipment meets the preset equipment operating conditions, based on the real-time monitoring results, a data fusion operation is performed on the real-time monitoring results and all data to be processed to obtain the data fusion result.

[0115] In this embodiment of the invention, optionally, the process can be terminated when it is determined that the communication device does not meet the preset device operating conditions.

[0116] In this embodiment of the invention, optionally, the data fusion operation includes integrating the running status data and appearance image data from the real-time monitoring results with the data to be processed to form a more comprehensive dataset.

[0117] In this embodiment of the invention, optionally, the data fusion result includes at least the fused dataset, which contains real-time monitoring results and data to be processed.

[0118] 205. Based on the data fusion results, perform an update operation on the data processing results.

[0119] In this embodiment of the invention, the update operation may optionally include reprocessing the fused data, such as data cleaning, feature extraction, and model inference, to generate a more accurate data processing result.

[0120] In this embodiment of the invention, optionally, the above-mentioned update operation on the data processing result based on the data fusion result may include: adding the fused dataset corresponding to the data fusion result to the data processing result to perform the update operation on the data processing result.

[0121] 206. Analyze the data processing results, obtain the data analysis results, generate the corresponding data optimization instructions for the communication equipment based on the data analysis results, and perform optimization operations on the communication equipment that match the data optimization instructions.

[0122] In this embodiment of the invention, for detailed descriptions of steps 201-202 and 206, please refer to the other descriptions of steps 101-103 in Embodiment 1. These descriptions will not be repeated in this embodiment of the invention.

[0123] It is evident that implementation Figure 2The described method enables real-time monitoring of the communication device using a preset machine vision algorithm to obtain real-time monitoring results. Based on these results, it determines whether the communication device meets preset operating conditions. If so, it performs a data fusion operation with the real-time monitoring results and all the data to be processed to obtain a data fusion result. Based on the data fusion result, it updates the data processing result. This method, through a preset machine vision algorithm, enables real-time monitoring of the communication device and can acquire real-time operating status and appearance image data of the device. This not only improves the real-time performance of device monitoring but also enhances the accuracy of monitoring through image recognition technology. Based on the real-time monitoring results, it determines whether the communication device meets preset operating conditions. This system helps improve the reliability of equipment operation. When the communication equipment meets the preset operating conditions, it performs a data fusion operation with the real-time monitoring results and all data to be processed. The resulting data fusion can integrate multi-source data to form a more comprehensive dataset, which helps improve the accuracy and comprehensiveness of subsequent data processing and analysis. It also ensures the real-time updating of data processing results, allowing data analysis to be based on the latest data, thus providing more accurate analysis results. Through real-time monitoring and data fusion, potential problems of the equipment can be detected in a timely manner, and optimization instructions can be generated through data analysis to optimize and adjust the equipment, which helps improve the operating efficiency of the equipment. In turn, it helps improve the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing.

[0124] In an optional embodiment, a data optimization instruction corresponding to the communication device is generated based on the data analysis results, including:

[0125] The data interaction information of the communication equipment is determined. Based on the data interaction information and data analysis results, and combined with the preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated.

[0126] Based on the data interaction parameters and the preset edge computing model, the calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, the target optimization instruction that matches the data interaction result is determined in the preset optimization instruction database.

[0127] Based on all target optimization instructions, generate data optimization instructions corresponding to the communication device;

[0128] Among them, the data optimization command is used to optimize seamless data interaction between heterogeneous devices.

[0129] In this embodiment of the invention, optionally, the data interaction information of the communication devices may include communication protocols, data transmission formats, interaction frequencies, etc. between the devices; this information is used to describe the data interaction methods and requirements of the communication devices in the industrial network.

[0130] In this embodiment of the invention, optionally, the above-mentioned generation of data interaction parameters corresponding to the data analysis results based on data interaction information and data analysis results, combined with a preset multi-protocol fusion model, may include:

[0131] The data interaction information and data analysis results are input into a preset multi-protocol fusion model, which performs analysis operations on the data interaction information and data analysis results to generate data interaction parameters corresponding to the data analysis results.

[0132] In this embodiment of the invention, optionally, a multi-protocol fusion model is used to standardize and fuse data from different protocols to ensure seamless data interaction between different devices.

[0133] In this embodiment of the invention, optional data interaction parameters include, but are not limited to, data format conversion parameters, protocol adaptation parameters, and transmission rate parameters.

[0134] In this embodiment of the invention, optionally, the above-mentioned calculation operation on the data interaction parameters based on the data interaction parameters and the preset edge computing model to obtain the data interaction result may include:

[0135] The data interaction parameters are input into a preset edge computing model, and the edge computing model performs calculation operations on the data interaction parameters to obtain the data interaction results.

[0136] In this embodiment of the invention, optionally, the edge computing model is used to perform data processing and analysis on edge devices, reduce data transmission latency, and improve response speed; the data interaction results include optimized data transmission format, protocol adaptation scheme, transmission rate, etc.

[0137] In this embodiment of the invention, optionally, the optimization instruction database stores a variety of optimization instructions and their corresponding applicable scenarios and parameters. By matching the data interaction results, the most suitable optimization instruction is selected to determine the target optimization instruction.

[0138] In this embodiment of the invention, optionally, the data optimization instructions are used to optimize seamless data interaction between heterogeneous devices, including but not limited to device parameter adjustment instructions, communication protocol conversion instructions, and data transmission optimization instructions.

[0139] As can be seen, implementing this optional embodiment can determine the data interaction information of communication devices. Based on the data interaction information and data analysis results, combined with a preset multi-protocol fusion model, data interaction parameters corresponding to the data analysis results are generated. According to the data interaction parameters and a preset edge computing model, calculation operations are performed on the data interaction parameters to obtain the data interaction results. Target optimization instructions matching the data interaction results are determined in the optimization instruction database. Based on all target optimization instructions, data optimization instructions corresponding to the communication devices are generated. By determining the data interaction information of the communication devices and generating data interaction parameters in combination with a preset multi-protocol fusion model, different types of devices can seamlessly interact with data in the same industrial network, improving the system's compatibility and flexibility. The edge computing model is used to calculate the data interaction parameters to obtain… The optimized data interaction results, through edge computing, can reduce data transmission latency, improve response speed, and ensure the efficiency and reliability of data interaction. Based on the data interaction results, target optimization instructions matching the data interaction results are determined from a pre-set optimization instruction database, which helps improve the accuracy of determining target optimization instructions and improve optimization effects. Based on data interaction information and data analysis results, combined with a pre-set multi-protocol fusion model, data interaction parameters corresponding to the data analysis results are generated, which can standardize and fuse data from different protocols, ensuring seamless data interaction between different devices. This not only improves the efficiency and reliability of data interaction but also enhances the intelligence level of device operation, thereby improving the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing.

[0140] In another optional embodiment, before performing data processing operations on all the data to be processed and obtaining the data processing result, the method further includes:

[0141] Obtain the device network status of the communication device and determine whether the device network status indicates that the communication device is in offline mode.

[0142] When the device network status is determined to indicate that the communication device is in offline mode, based on all data to be processed and the preset microsecond-level TSN time synchronization network, the target preset model is determined from the pre-determined pre-stored model through the preset device fault prediction algorithm; wherein, the target preset model is an intelligent model pre-stored in the communication device; the preset device fault prediction algorithm includes LSTM time series analysis prediction algorithm;

[0143] This involves performing data processing operations on all data to be processed, resulting in the following data processing results:

[0144] Based on the target pre-defined model, perform data processing operations on all data to be processed to obtain the data processing results.

[0145] In this optional embodiment, the device network status of the communication device can be obtained through the network interface of the communication device, including the current network connection status and network quality indicators. For example, it can be determined whether the device is in offline mode by detecting parameters such as network connection stability, latency, and bandwidth.

[0146] In this optional embodiment, it is further possible to terminate the process when it is determined that the device network status indicates that the communication device is not in offline mode.

[0147] In this optional embodiment, based on all the data to be processed and a preset microsecond-level TSN time synchronization network, a target preset model is determined from a pre-determined pre-stored model using a preset device fault prediction algorithm (such as the LSTM time series analysis prediction algorithm). For example, based on the time series characteristics of the data to be processed, the most suitable pre-stored model is selected for fault prediction. The LSTM (Long Short-Term Memory) time series analysis prediction algorithm is a special type of recurrent neural network (RNN) specifically designed for processing and predicting long-term dependencies in time series data. LSTM effectively solves the gradient vanishing and gradient exploding problems of traditional RNNs when processing long-sequence data by introducing a "gating" mechanism to control the flow of information.

[0148] In this optional embodiment, the pre-stored model may include an intelligent model pre-stored in the communication device for data processing and fault prediction in an offline state; the target preset model may include an intelligent model determined from the pre-stored model that is most suitable for the current data to be processed.

[0149] In this optional embodiment, the data processing operation may include operations such as data cleaning, feature extraction, and model inference to process the data to be processed and generate data processing results; the data processing results include the results obtained through the data processing operation, including information such as equipment operating status assessment, fault prediction, and production efficiency assessment.

[0150] In this optional embodiment, optionally, for example, data processing operations are performed on all the data to be processed according to the target pre-defined model, including data cleaning, feature extraction, and model inference. For example, an LSTM model is used to perform time series analysis on the equipment operation data to predict the equipment's failure trend and generate data processing results.

[0151] As can be seen, implementing this optional embodiment can obtain the device network status of the communication device and determine whether the device network status is used to indicate that the communication device is in an offline mode. If so, based on all the data to be processed and the preset microsecond-level TSN time synchronization network, the target preset model is determined in the pre-determined pre-stored model through the preset device fault prediction algorithm. Data processing operations are performed on all the data to be processed according to the target preset model to obtain the data processing results. When the communication device is in an offline mode, the device can continue to perform data processing and fault prediction through the pre-stored intelligent model and device fault prediction algorithm. This ensures that the equipment can continue to operate autonomously even when offline, maintaining the continuity of the production process and reducing production interruptions caused by network problems. Utilizing the LSTM time series analysis and prediction algorithm, it can effectively process time series data, capture long-term dependencies in the data, and make equipment fault prediction more accurate, which is conducive to improving equipment reliability and service life. Based on the microsecond-level TSN time synchronization network, it ensures high precision and real-time performance of data processing. Data processing is carried out through pre-stored models, which improves the efficiency of data processing and ensures that data processing results can still be generated quickly in offline mode. By predicting equipment faults in advance and making optimization adjustments, the frequency and cost of equipment maintenance are reduced. Even when the communication equipment is in offline mode, it can continue to perform data processing and fault prediction using pre-stored intelligent models and equipment fault prediction algorithms, ensuring that the equipment can still perform effective intelligent diagnosis and optimization operations even when offline.

[0152] In another optional embodiment, based on the real-time monitoring results, a data fusion operation is performed between the real-time monitoring results and all data to be processed to obtain a data fusion result, including:

[0153] Preprocessing operations are performed on the real-time monitoring results. Multi-scale defect feature information corresponding to the real-time monitoring results is extracted through a feature pyramid network. The preprocessing operations include one or more of adaptive white balance operations and image enhancement operations. The feature pyramid network includes an FPN network.

[0154] Data alignment is performed on the multi-scale defect feature information and all data to be processed to obtain the data alignment result. Then, combined with the pre-determined cross-model fusion model, data fusion is performed on the multi-scale defect feature information and all data to be processed to obtain the data fusion result.

[0155] Among them, the pre-determined cross-model fusion model includes the Transformer model.

[0156] In this optional embodiment, the preprocessing operation performed on the real-time monitoring results may include performing adaptive white balance operation and image enhancement operation on the real-time monitoring results to improve image quality and the accuracy of feature extraction. Adaptive white balance operation is an image processing technique used to adjust the color balance of the image, making the image colors closer to the real scene; image enhancement operation is an image processing technique used to improve the contrast and sharpness of the image, enhancing the visual effect of the image.

[0157] In this optional embodiment, the extraction of multi-scale defect feature information corresponding to the real-time monitoring results via the feature pyramid network may include: extracting feature monitoring information corresponding to each dimension in the real-time monitoring results via the feature pyramid network, and generating multi-scale defect feature information based on the feature monitoring information corresponding to each dimension. The FPN network can effectively identify defect features at different scales, improving the accuracy of defect detection.

[0158] In this optional embodiment, the above-mentioned data alignment operation based on multi-scale defect feature information and all data to be processed to obtain the data alignment result may include:

[0159] First spatiotemporal information corresponding to multi-scale feature information and second spatiotemporal information corresponding to all data to be processed are extracted. Based on the first and second spatiotemporal information, data alignment operations are performed between the multi-scale defect feature information and all data to be processed to obtain the data alignment result. For example, timestamp alignment and spatial coordinate alignment are used to ensure that the multi-scale defect feature information and the data to be processed can be correctly matched. The data alignment operation includes temporal alignment and spatial alignment to ensure the consistency of multi-scale defect feature information and the data to be processed in time and space.

[0160] In this optional embodiment, the above-mentioned combination of a pre-determined cross-model fusion model to perform data fusion operations on multi-scale defect feature information and all data to be processed to obtain a data fusion result may include: combining a pre-determined cross-model fusion model (such as the Transformer model) to perform data fusion operations on multi-scale defect feature information and all data to be processed to generate a data fusion result. The Transformer model processes and fuses data from different sources through a self-attention mechanism, generating a more comprehensive data fusion result. The Transformer model is a deep learning model based on a self-attention mechanism, widely used in natural language processing and computer vision tasks, and capable of processing and fusing data from different sources.

[0161] As can be seen, implementing this optional embodiment can perform preprocessing operations on real-time monitoring results. It extracts multi-scale defect feature information corresponding to the real-time monitoring results through a feature pyramid network, performs data alignment operations based on the multi-scale defect feature information and all data to be processed to obtain data alignment results, and combines a cross-model fusion model to perform data fusion operations on the multi-scale defect feature information and all data to be processed to obtain data fusion results. Extracting multi-scale defect feature information from real-time monitoring results through a feature pyramid network (FPN network) can effectively capture defect features at different scales, more accurately identify and locate defects, and improve the accuracy of defect detection. Preprocessing operations on real-time monitoring results, including adaptive white balance and image enhancement operations, can improve image quality and contrast. The FPN network can simultaneously utilize high-resolution shallow features and low-resolution... The system leverages deep features of high resolution to generate more comprehensive feature representations through feature fusion, improving the detection capability of multi-scale defects. By combining a pre-defined cross-model fusion model (such as the Transformer model) with multi-scale defect feature information and all data to be processed, it performs data fusion operations, generating more comprehensive and accurate data fusion results. It can effectively capture long-term dependencies in the data, improving the efficiency and accuracy of data fusion. Through data alignment operations, it ensures the consistency of multi-scale defect feature information and data to be processed in time and space, effectively addressing data inconsistencies and noise interference. Through preprocessing operations and multi-scale feature fusion, the system can more accurately identify and locate defects, reducing false alarm rates. It can effectively fuse real-time monitoring results with data to be processed, generating more comprehensive data fusion results and improving the accuracy and reliability of data processing.

[0162] In another optional embodiment, a data computing task is determined based on a preset heterogeneous computing resource scheduling algorithm and the data to be processed, including:

[0163] By using a pre-defined heterogeneous computing resource scheduling algorithm, a spatiotemporal topology map of resources corresponding to communication devices is generated, and multi-dimensional feature encoding is performed on all data to be processed to obtain multi-dimensional data feature information corresponding to all data to be processed.

[0164] Based on the spatiotemporal topology of resources and multidimensional data feature information, dynamic planning parameters corresponding to all data to be processed are generated, and it is determined whether the planning accuracy corresponding to the dynamic planning parameters meets the preset calculation accuracy conditions.

[0165] When it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined based on the dynamic programming parameters.

[0166] In this optional embodiment, the above-mentioned generation of the resource spatiotemporal topology map corresponding to the communication device using a preset heterogeneous computing resource scheduling algorithm may include: determining the device computing resources corresponding to the communication device using the preset heterogeneous computing resource scheduling algorithm, and determining the resource usage of the communication device based on the device computing resources corresponding to the communication device; generating the resource spatiotemporal topology map corresponding to the communication device based on the resource usage of all communication devices. For example, the usage of different computing resources (such as CPU cores, GPUs, NPUs, etc.) in the communication device is collected using the preset heterogeneous computing resource scheduling algorithm to generate the resource spatiotemporal topology map. This map describes the distribution and utilization of resources in time and space, providing a basis for subsequent task allocation.

[0167] In this optional embodiment, the multi-dimensional feature encoding may include encoding features such as data type, data size, and data priority. For example, image data may be labeled as type 1, text data as type 2, data size may be divided into three levels: small, medium, and large, and data priority may be divided into three levels: high, medium, and low.

[0168] In this optional embodiment, the above-mentioned generation of dynamic programming parameters corresponding to all data to be processed based on the resource spatiotemporal topology map and multidimensional data feature information may include:

[0169] Based on the spatiotemporal topology of resources and multidimensional data feature information, the data processing priority and available computing resource priority corresponding to each piece of data to be processed are determined. Based on the data processing priority and available computing resource priority of each piece of data to be processed, task allocation strategies and resource allocation strategies are generated. Then, dynamic programming parameters corresponding to all pieces of data to be processed are generated according to the task allocation strategies and resource allocation strategies. For example, dynamic programming parameters may include one or more of the following: execution order of computing tasks, execution time, and execution method.

[0170] In this optional embodiment, determining whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions can be done by checking whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions. For example, checking whether the task allocation delay is within an acceptable range and whether the resource utilization rate reaches a preset threshold. If the planning accuracy meets the conditions, proceed to the next step; otherwise, readjust the dynamic programming parameters.

[0171] In this optional embodiment, when it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined based on the dynamic programming parameters. For example, the protocol conversion task is assigned to the CPU core, the AI ​​inference task is assigned to the NPU, and the storage management task is assigned to the memory management unit.

[0172] In this optional embodiment, it is further possible to terminate the process when it is determined that the planning accuracy corresponding to the dynamic programming parameters does not meet the preset calculation accuracy conditions.

[0173] As can be seen, implementing this optional embodiment can generate a spatiotemporal topology map of resources corresponding to the communication equipment through a preset heterogeneous computing resource scheduling algorithm, and perform multi-dimensional feature encoding on all data to be processed to obtain multi-dimensional data feature information corresponding to all data to be processed; based on the spatiotemporal topology map and multi-dimensional data feature information, dynamic programming parameters corresponding to all data to be processed are generated, and it is determined whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions. If it does, the data calculation task is determined according to the dynamic programming parameters. By generating a spatiotemporal topology map, the distribution and utilization of different computing resources in the communication equipment in time and space can be fully understood, making resource allocation more reasonable and improving the overall resource utilization efficiency. Multi-dimensional feature encoding on all data to be processed can more comprehensively describe the attributes of the data. Based on the specific characteristics of the data, the system can more accurately allocate tasks and resources to meet processing needs, thereby improving the accuracy of data processing. It generates dynamic programming parameters based on the spatiotemporal topology of resources and multidimensional data feature information, enabling dynamic adjustment of task and resource allocation strategies. It can also improve task execution efficiency based on real-time resource status and data characteristics, and determine whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions. This ensures the accuracy of task and resource allocation and also improves system reliability and stability. Optimizing task scheduling through dynamic programming parameters reduces task waiting time and resource allocation delays, improving system response speed. Furthermore, it can dynamically generate and adjust data computation tasks based on the resource status of communication equipment and the characteristics of the data to be processed, improving resource utilization efficiency and data processing accuracy.

[0174] In another optional embodiment, based on the data computation task and the data time series, data processing operations are performed on all the data to be processed to obtain data processing results, including:

[0175] Based on the data computing task, determine the task scheduling parameters corresponding to the data computing task, and determine the data computing resource parameters according to the task scheduling parameters;

[0176] Based on the time series data, multidimensional feature sequence parameters are determined, including time domain feature parameters, frequency domain feature parameters, and time-frequency domain feature parameters.

[0177] Based on the data computation resource parameters and multidimensional feature sequence parameters, data computation processing parameters corresponding to all data to be processed are generated, and data processing operations are performed on all data to be processed based on the data computation processing parameters to obtain the data processing results.

[0178] In this optional embodiment, the data computing task may include computing tasks generated based on the data to be processed, including protocol conversion, AI inference, storage management, etc.; the task scheduling parameters may include parameters used to guide task execution and resource allocation, including task priority, task execution order, task allocation strategy, etc.; the data computing resource parameters may include parameters used to guide resource allocation and management, including resource type (such as CPU cores, GPU, NPU, etc.), resource quantity, resource allocation strategy, etc.

[0179] In this optional embodiment, the task scheduling parameters can be determined based on the data calculation task by generating a task scheduling table according to the task priority and execution order to guide the task execution order and resource allocation strategy.

[0180] In this optional embodiment, the determination of data computing resource parameters based on task scheduling parameters may be based on the type and priority of the task, and the allocation of corresponding computing resources, such as allocating AI inference tasks to the NPU and protocol conversion tasks to CPU cores.

[0181] In this optional embodiment, the above-mentioned determination of multidimensional feature sequence parameters based on data time series may include: extracting time-domain feature parameters (such as mean, variance, peak value), frequency-domain feature parameters (such as spectral energy, dominant frequency), and time-frequency-domain feature parameters (such as wavelet transform coefficients) from the data time series, and determining multidimensional feature sequence parameters based on the extracted time-domain feature parameters, frequency-domain feature parameters, and time-frequency-domain feature parameters.

[0182] In this optional embodiment, the time-domain feature parameters may include features describing the data in the time domain, such as mean, variance, peak value, etc.; the frequency-domain feature parameters may include features describing the data in the frequency domain, such as spectral energy, dominant frequency, etc.; and the time-frequency domain feature parameters may include features describing the data in the time and frequency domains, such as wavelet transform coefficients, etc.

[0183] In this optional embodiment, the above-mentioned generation of data calculation and processing parameters corresponding to all data to be processed based on data calculation resource parameters and multidimensional feature sequence parameters may include:

[0184] The process involves determining resource types based on data computation resource parameters and feature parameters based on multidimensional feature sequence parameters. Then, based on the resource type and feature parameters, a suitable data processing algorithm (such as a deep learning model or a traditional machine learning model) is selected, and the algorithm parameters are optimized to generate data computation processing parameters corresponding to all data to be processed. Furthermore, these data computation processing parameters can include parameters to guide data processing operations, such as data processing algorithm selection and parameter optimization strategies.

[0185] In this optional embodiment, data processing operations can be performed on all data to be processed based on data calculation and processing parameters to obtain data processing results. For example, a deep learning model can be used to perform model inference on extracted features to generate data processing results such as equipment operating status assessment, fault prediction, and production efficiency assessment. The data processing operations can include data cleaning, feature extraction, and model inference to process the data to be processed and generate data processing results. The data processing results can include the results obtained through the data processing operations, including information such as equipment operating status assessment, fault prediction, and production efficiency assessment.

[0186] As can be seen, implementing this optional embodiment can determine the task scheduling parameters corresponding to the data computing task based on the data computing task, and determine the data computing resource parameters based on the task scheduling parameters; determine multi-dimensional feature sequence parameters based on the data time series; generate data computing processing parameters corresponding to all data to be processed based on the data computing resource parameters and multi-dimensional feature sequence parameters; and perform data processing operations on all data to be processed based on the data computing processing parameters to obtain the data processing results. By determining the task scheduling parameters and data computing resource parameters, computing resources can be rationally allocated according to the task priority and resource availability, improving the efficiency of data processing. Data processing based on multi-dimensional feature sequence parameters (including time domain, frequency domain, and time-frequency domain feature parameters) can capture more comprehensively. By leveraging data characteristics, the system improves the accuracy of data processing. It dynamically allocates computing resources based on task scheduling parameters to ensure efficient resource utilization. Through the extraction and analysis of multi-dimensional feature sequence parameters, it can process various types of data to meet diverse industrial needs. By optimizing task scheduling and resource allocation, the system can respond to data processing demands more quickly, improving the real-time performance of data processing. The integration of advanced technologies such as task scheduling, resource management, and multi-dimensional feature extraction enhances the level of intelligence and flexibility. Furthermore, the extraction and analysis of multi-dimensional feature sequence parameters improves the system's reliability and trustworthiness. It can dynamically generate and adjust data processing parameters based on data calculation tasks and data time series, improving data processing efficiency and accuracy, and ensuring the system's efficient operation and reliability.

[0187] Example 3

[0188] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a data intelligent processing device applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention. Figure 3 As shown, the data intelligence processing device applied to the industrial communication intelligent platform may include:

[0189] The acquisition module 301 is used to acquire the data to be processed corresponding to the communication device;

[0190] The determination module 302 is used to determine the data computing task based on a preset heterogeneous computing resource scheduling algorithm and the data to be processed;

[0191] The generation module 303 is used to generate a data time series corresponding to all data to be processed based on the data calculation task and all data to be processed.

[0192] The processing module 304 is used to perform data processing operations on all data to be processed based on the data computing task and the data time series, and obtain the data processing results; wherein, the data processing operations are used to realize microsecond-level time synchronization and dynamic bandwidth allocation of communication devices.

[0193] Analysis module 305 is used to analyze the data processing results and obtain data analysis results;

[0194] The generation module 303 is also used to generate data optimization instructions corresponding to the communication device based on the data analysis results;

[0195] The optimization module 306 is used to perform optimization operations on the communication device that match the data optimization instructions.

[0196] It is evident that implementation Figure 3The described device can generate data time series corresponding to all data to be processed based on data computing tasks and all data to be processed; perform data processing operations on all data to be processed based on data computing tasks and data time series to obtain data processing results; analyze the data processing results to obtain data analysis results; and generate data optimization instructions corresponding to communication devices based on the data analysis results, and execute optimization operations on communication devices that match the data optimization instructions. It can dynamically allocate computing resources of multi-core ARM processors according to the type, size, and priority of the data to be processed through a preset heterogeneous computing resource scheduling algorithm, which is beneficial to improving data processing efficiency and reducing data latency. It can also utilize the hardware timestamp engine of the intelligent Ethernet switching module to... The system generates time series data corresponding to all data to be processed in chronological order, ensuring data accuracy and real-time performance. By analyzing the data processing results, it generates data optimization instructions for communication devices and executes optimization operations that match these instructions. This enables intelligent diagnosis and adaptive optimization of the devices, improving their reliability and safety. By integrating multi-protocol communication, edge computing, AI large models, and machine vision capabilities, it connects the entire "device-edge-cloud" link, enabling communication perception collaboration, intelligent diagnosis, and adaptive optimization. This allows the system to automatically adapt to different industrial scenarios and needs, improving its intelligence and flexibility. Consequently, it enhances the intelligence and efficiency of data processing, as well as its accuracy and reliability.

[0197] In an optional embodiment, such as Figure 4 As shown, the device also includes:

[0198] The monitoring module 307 is used to perform real-time monitoring operations on the communication device through a preset machine vision algorithm before the analysis module 305 analyzes the data processing results and obtains the data analysis results, and obtains real-time monitoring results.

[0199] The judgment module 308 is used to determine whether the communication device meets the preset device operating conditions based on the real-time monitoring results;

[0200] The fusion module 309 is used to perform a data fusion operation on the real-time monitoring results and all data to be processed based on the real-time monitoring results when the judgment module 308 determines that the communication device meets the preset device operating conditions, so as to obtain the data fusion result.

[0201] The update module 310 is used to perform update operations on the data processing results based on the data fusion results, and trigger the analysis module 305 to analyze the data processing results to obtain the data analysis results.

[0202] It is evident that implementation Figure 4The described device can perform real-time monitoring of the communication device using a preset machine vision algorithm to obtain real-time monitoring results. Based on the real-time monitoring results, it determines whether the communication device meets preset device operating conditions. If it does, it performs a data fusion operation with all the data to be processed based on the real-time monitoring results to obtain a data fusion result. Based on the data fusion result, it performs an update operation on the data processing result. This device can perform real-time monitoring of the communication device using a preset machine vision algorithm, enabling real-time acquisition of the device's operating status and appearance image data. This not only improves the real-time performance of device monitoring but also enhances the accuracy of monitoring through image recognition technology. Based on the real-time monitoring results, it determines whether the communication device meets preset device operating conditions. This system helps improve the reliability of equipment operation. When the communication equipment meets the preset operating conditions, it performs a data fusion operation with the real-time monitoring results and all data to be processed. The resulting data fusion can integrate multi-source data to form a more comprehensive dataset, which helps improve the accuracy and comprehensiveness of subsequent data processing and analysis. It also ensures the real-time updating of data processing results, allowing data analysis to be based on the latest data, thus providing more accurate analysis results. Through real-time monitoring and data fusion, potential problems of the equipment can be detected in a timely manner, and optimization instructions can be generated through data analysis to optimize and adjust the equipment, which helps improve the operating efficiency of the equipment. In turn, it helps improve the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing.

[0203] In another alternative embodiment, such as Figure 4 As shown, the specific methods by which the generation module 303 generates data optimization instructions corresponding to the communication device based on the data analysis results include:

[0204] The data interaction information of the communication equipment is determined. Based on the data interaction information and data analysis results, and combined with the preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated.

[0205] Based on the data interaction parameters and the preset edge computing model, the calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, the target optimization instruction that matches the data interaction result is determined in the preset optimization instruction database.

[0206] Based on all target optimization instructions, generate data optimization instructions corresponding to the communication device;

[0207] Among them, the data optimization command is used to optimize seamless data interaction between heterogeneous devices.

[0208] It is evident that implementation Figure 4The described device can determine the data interaction information of communication devices, and based on the data interaction information and data analysis results, combined with a preset multi-protocol fusion model, generate data interaction parameters corresponding to the data analysis results. According to the data interaction parameters and a preset edge computing model, it performs calculations on the data interaction parameters to obtain the data interaction results, and determines the target optimization instructions matching the data interaction results in an optimization instruction database. Based on all target optimization instructions, it generates data optimization instructions corresponding to the communication devices. By determining the data interaction information of the communication devices and generating data interaction parameters in combination with the preset multi-protocol fusion model, different types of devices can seamlessly interact with data in the same industrial network, improving the system's compatibility and flexibility. The edge computing model is used to calculate the data interaction parameters to obtain optimized data. The data interaction results, through edge computing, can reduce data transmission latency, improve response speed, and ensure the efficiency and reliability of data interaction. Based on the data interaction results, target optimization instructions matching the data interaction results are determined in a pre-set optimization instruction database, which helps improve the accuracy of determining target optimization instructions and improve optimization effects. Based on data interaction information and data analysis results, combined with a pre-set multi-protocol fusion model, data interaction parameters corresponding to the data analysis results are generated, which can standardize and fuse data from different protocols, ensuring seamless data interaction between different devices. This not only improves the efficiency and reliability of data interaction but also enhances the intelligence level of device operation, thereby improving the intelligence and efficiency of data processing, as well as the accuracy and reliability of data processing.

[0209] In yet another alternative embodiment, such as Figure 4 As shown, the acquisition module 301 is also used to acquire the device network status of the communication device before the processing module 304 performs data processing operations on all data to be processed and obtains the data processing result.

[0210] The judgment module 308 is also used to determine whether the device network status is used to indicate that the communication device is in offline mode.

[0211] The determining module 302 is further configured to, when the judging module 308 determines that the device network status indicates that the communication device is in an offline mode, determine the target preset model from the pre-determined pre-stored model based on all data to be processed and the preset microsecond-level TSN time synchronization network, using a preset device fault prediction algorithm; wherein, the target preset model is an intelligent model pre-stored in the communication device; the preset device fault prediction algorithm includes the LSTM time series analysis prediction algorithm;

[0212] Specifically, the processing module 304 performs data processing operations on all data to be processed, and obtains the data processing results in the following ways:

[0213] Based on the target pre-defined model, perform data processing operations on all data to be processed to obtain the data processing results.

[0214] It is evident that implementation Figure 4 The described device can acquire the network status of a communication device and determine whether the network status indicates that the communication device is in an offline mode. If so, based on all the data to be processed and the preset microsecond-level TSN time synchronization network, the device determines the target preset model from the pre-determined pre-stored model through a preset device fault prediction algorithm. The device then performs data processing operations on all the data to be processed according to the target preset model to obtain the data processing results. Even when the communication device is in an offline mode, the device can continue to perform data processing and fault prediction through the pre-stored intelligent model and the device fault prediction algorithm. This ensures that the equipment can continue to operate autonomously even when offline, maintaining the continuity of the production process and reducing production interruptions caused by network problems. Utilizing the LSTM time series analysis and prediction algorithm, it can effectively process time series data, capture long-term dependencies in the data, and make equipment fault prediction more accurate, which is conducive to improving equipment reliability and service life. Based on the microsecond-level TSN time synchronization network, it ensures high precision and real-time performance of data processing. Data processing is carried out through pre-stored models, which improves the efficiency of data processing and ensures that data processing results can still be generated quickly in offline mode. By predicting equipment faults in advance and making optimization adjustments, the frequency and cost of equipment maintenance are reduced. Even when the communication equipment is in offline mode, it can continue to perform data processing and fault prediction using pre-stored intelligent models and equipment fault prediction algorithms, ensuring that the equipment can still perform effective intelligent diagnosis and optimization operations even when offline.

[0215] In yet another alternative embodiment, such as Figure 4 As shown, the fusion module 309 performs a data fusion operation based on the real-time monitoring results and all the data to be processed, and the specific methods for obtaining the data fusion results include:

[0216] Preprocessing operations are performed on the real-time monitoring results. Multi-scale defect feature information corresponding to the real-time monitoring results is extracted through a feature pyramid network. The preprocessing operations include one or more of adaptive white balance operations and image enhancement operations. The feature pyramid network includes an FPN network.

[0217] Data alignment is performed on the multi-scale defect feature information and all data to be processed to obtain the data alignment result. Then, combined with the pre-determined cross-model fusion model, data fusion is performed on the multi-scale defect feature information and all data to be processed to obtain the data fusion result.

[0218] Among them, the pre-determined cross-model fusion model includes the Transformer model.

[0219] It is evident that implementation Figure 4 The described device performs preprocessing operations on real-time monitoring results. It extracts multi-scale defect feature information corresponding to the real-time monitoring results through a feature pyramid network (FPN), performs data alignment operations based on the multi-scale defect feature information and all data to be processed to obtain data alignment results, and combines a cross-model fusion model to perform data fusion operations on the multi-scale defect feature information and all data to be processed to obtain data fusion results. By extracting multi-scale defect feature information from the real-time monitoring results through a feature pyramid network (FPN), it can effectively capture defect features at different scales, more accurately identify and locate defects, and improve the accuracy of defect detection. Preprocessing operations on the real-time monitoring results, including adaptive white balance and image enhancement operations, can improve image quality and contrast. The FPN network can simultaneously utilize high-resolution shallow features and low-resolution... Deep features, through feature fusion, generate more comprehensive feature representations, improving the detection capability of multi-scale defects. Combined with a pre-determined cross-model fusion model (such as the Transformer model), data fusion operations are performed on multi-scale defect feature information and all data to be processed, generating more comprehensive and accurate data fusion results. It can effectively capture long-term dependencies in the data, improving the efficiency and accuracy of data fusion. Through data alignment operations, it ensures the consistency of multi-scale defect feature information and data to be processed in time and space, effectively dealing with data inconsistencies and noise interference. Through preprocessing operations and multi-scale feature fusion, the system can more accurately identify and locate defects, reducing the false alarm rate. It can effectively fuse real-time monitoring results with data to be processed, generating more comprehensive data fusion results and improving the accuracy and reliability of data processing.

[0220] In yet another alternative embodiment, such as Figure 4 As shown, the determining module 302 determines the specific method of the data computing task through a preset heterogeneous computing resource scheduling algorithm and the data to be processed, including:

[0221] By using a pre-defined heterogeneous computing resource scheduling algorithm, a spatiotemporal topology map of resources corresponding to communication devices is generated, and multi-dimensional feature encoding is performed on all data to be processed to obtain multi-dimensional data feature information corresponding to all data to be processed.

[0222] Based on the spatiotemporal topology of resources and multidimensional data feature information, dynamic planning parameters are generated for all data to be processed, and it is determined whether the planning accuracy corresponding to the dynamic planning parameters meets the preset calculation accuracy conditions.

[0223] When it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined based on the dynamic programming parameters.

[0224] It is evident that implementation Figure 4 The described device can generate a spatiotemporal topology map of resources corresponding to communication equipment through a preset heterogeneous computing resource scheduling algorithm, and perform multi-dimensional feature encoding on all data to be processed to obtain multi-dimensional data feature information corresponding to all data to be processed. Based on the spatiotemporal topology map and multi-dimensional data feature information, dynamic programming parameters corresponding to all data to be processed are generated. It is then determined whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions. If it does, the data calculation task is determined according to the dynamic programming parameters. By generating the spatiotemporal topology map, the device can comprehensively understand the distribution and utilization of different computing resources in the communication equipment in time and space, making resource allocation more reasonable and improving overall resource utilization efficiency. Multi-dimensional feature encoding on all data to be processed can more comprehensively describe the attributes and processing of the data. The system aims to achieve more precise task and resource allocation based on the specific characteristics of the data, thereby improving the accuracy of data processing. It generates dynamic programming parameters based on resource spatiotemporal topology maps and multidimensional data feature information, enabling dynamic adjustment of task and resource allocation strategies. It can also improve task execution efficiency based on real-time resource status and data characteristics, and determine whether the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions. This ensures the accuracy of task and resource allocation and improves system reliability and stability. Furthermore, by optimizing task scheduling through dynamic programming parameters, it can reduce task waiting time and resource allocation delays, improving system response speed. Finally, it can dynamically generate and adjust data computation tasks based on the resource status of communication equipment and the characteristics of the data to be processed, improving resource utilization efficiency and data processing accuracy.

[0225] In yet another alternative embodiment, such as Figure 4 As shown, the processing module 304 performs data processing operations on all data to be processed based on the data calculation task and the data time series, and the specific methods for obtaining the data processing results include:

[0226] Based on the data computing task, determine the task scheduling parameters corresponding to the data computing task, and determine the data computing resource parameters according to the task scheduling parameters;

[0227] Based on the time series data, multidimensional feature sequence parameters are determined, including time domain feature parameters, frequency domain feature parameters, and time-frequency domain feature parameters.

[0228] Based on the data computation resource parameters and multidimensional feature sequence parameters, data computation processing parameters corresponding to all data to be processed are generated, and data processing operations are performed on all data to be processed based on the data computation processing parameters to obtain the data processing results.

[0229] It is evident that implementation Figure 4The described apparatus can determine task scheduling parameters corresponding to data computing tasks based on data computing tasks, and determine data computing resource parameters based on task scheduling parameters; it can determine multi-dimensional feature sequence parameters based on data time series, and generate data computing processing parameters corresponding to all data to be processed based on data computing resource parameters and multi-dimensional feature sequence parameters, and perform data processing operations on all data to be processed based on data computing processing parameters to obtain data processing results. By determining task scheduling parameters and data computing resource parameters, it can rationally allocate computing resources according to task priority and resource availability, thereby improving data processing efficiency. Data processing based on multi-dimensional feature sequence parameters (including time domain, frequency domain, and time-frequency domain feature parameters) can capture data more comprehensively. This system enhances the accuracy of data processing by dynamically allocating computing resources based on task scheduling parameters to ensure efficient resource utilization. Through the extraction and analysis of multi-dimensional feature sequence parameters, it can process various types of data to meet diverse industrial needs. Optimized task scheduling and resource allocation enable the system to respond more quickly to data processing demands, improving real-time performance. The integration of advanced technologies such as task scheduling, resource management, and multi-dimensional feature extraction enhances intelligence and flexibility. Furthermore, the extraction and analysis of multi-dimensional feature sequence parameters improves system reliability and trustworthiness. It can dynamically generate and adjust data processing parameters based on data computation tasks and data time series, improving data processing efficiency and accuracy, and ensuring efficient system operation and reliability.

[0230] Example 4

[0231] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another data intelligent processing device applied to an industrial communication intelligent platform, as disclosed in an embodiment of the present invention. Figure 5 As shown, the data intelligence processing device applied to the industrial communication intelligent platform may include:

[0232] Memory 401 storing executable program code;

[0233] Processor 402 coupled to memory 401;

[0234] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the data intelligent processing method applied to the industrial communication intelligent platform described in Embodiment 1 or Embodiment 2 of the present invention.

[0235] Example 5

[0236] This invention discloses a computer-storable medium storing computer instructions. When these computer instructions are invoked, they are used to execute the steps in the data intelligent processing method applied to an industrial communication intelligent platform as described in Embodiment 1 or Embodiment 2 of this invention.

[0237] Example 6

[0238] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the data intelligent processing method applied to an industrial communication intelligent platform as described in Embodiment 1 or Embodiment 2.

[0239] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0240] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0241] Finally, it should be noted that the data intelligent processing method and apparatus for industrial communication intelligent platforms disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data intelligent processing method applied to an industrial communication intelligent platform, characterized in that, The method includes: Obtain the data to be processed corresponding to the communication device, and determine the data computing task through a preset heterogeneous computing resource scheduling algorithm and the data to be processed; Based on the data computation task and all the data to be processed, a data time series corresponding to all the data to be processed is generated. Based on the data computation task and the data time series, a data processing operation is performed on all the data to be processed to obtain a data processing result. The data processing operation is used to achieve microsecond-level time synchronization and dynamic bandwidth allocation of the communication device. The data processing results are analyzed to obtain data analysis results. Based on the data analysis results, a data optimization instruction corresponding to the communication device is generated, and an optimization operation matching the data optimization instruction is performed on the communication device. The step of generating data optimization instructions corresponding to the communication device based on the data analysis results includes: The data interaction information of the communication device is determined, and based on the data interaction information and the data analysis results, combined with a preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated. Based on the data interaction parameters and the preset edge computing model, a calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, a target optimization instruction that matches the data interaction result is determined in a preset optimization instruction database. Based on all the target optimization instructions, generate the data optimization instructions corresponding to the communication device; The data optimization instructions are used to optimize seamless data interaction between heterogeneous devices.

2. The data intelligent processing method applied to an industrial communication intelligent platform according to claim 1, characterized in that, Before analyzing the data processing results and obtaining the data analysis results, the method further includes: The communication device is monitored in real time by a preset machine vision algorithm to obtain real-time monitoring results. Based on the real-time monitoring results, it is determined whether the communication device meets the preset device operating conditions. When it is determined that the communication device meets the preset device operating conditions, based on the real-time monitoring results, a data fusion operation is performed on the real-time monitoring results and all the data to be processed to obtain the data fusion result; Based on the data fusion result, an update operation is performed on the data processing result, and the operation of analyzing the data processing result to obtain the data analysis result is triggered.

3. The data intelligent processing method applied to an industrial communication intelligent platform according to claim 2, characterized in that, Before performing data processing operations on all the data to be processed to obtain the data processing result, the method further includes: Obtain the device network status of the communication device, and determine whether the device network status is used to indicate that the communication device is in offline mode. When it is determined that the device network status indicates that the communication device is in the offline mode state, based on all the data to be processed and the preset microsecond-level TSN time synchronization network, a target preset model is determined from the pre-determined pre-stored model through a preset device fault prediction algorithm; wherein, the target preset model is an intelligent model pre-stored in the communication device; the preset device fault prediction algorithm includes the LSTM time series analysis prediction algorithm; The step of performing data processing operations on all the data to be processed to obtain data processing results includes: Based on the target preset model, data processing operations are performed on all the data to be processed to obtain the data processing results.

4. The data intelligent processing method applied to an industrial communication intelligent platform according to claim 2, characterized in that, Based on the real-time monitoring results, a data fusion operation is performed between the real-time monitoring results and all the data to be processed to obtain a data fusion result, including: The real-time monitoring results are preprocessed by extracting multi-scale defect feature information corresponding to the real-time monitoring results through a feature pyramid network. The preprocessing operation includes one or more of adaptive white balance operation and image enhancement operation. The feature pyramid network includes an FPN network. Data alignment is performed on the multi-scale defect feature information and all the data to be processed to obtain the data alignment result. Then, combined with the pre-determined cross-model fusion model, data fusion is performed on the multi-scale defect feature information and all the data to be processed to obtain the data fusion result. The pre-determined cross-model fusion model includes the Transformer model.

5. The data intelligent processing method applied to an industrial communication intelligent platform according to claim 1, characterized in that, The step of determining the data computing task using a preset heterogeneous computing resource scheduling algorithm and the data to be processed includes: By using a preset heterogeneous computing resource scheduling algorithm, a spatiotemporal topology map of resources corresponding to the communication device is generated, and multi-dimensional feature encoding is performed on all the data to be processed to obtain multi-dimensional data feature information corresponding to all the data to be processed. Based on the resource spatiotemporal topology map and the multidimensional data feature information, dynamic planning parameters corresponding to all the data to be processed are generated, and it is determined whether the planning accuracy corresponding to the dynamic planning parameters meets the preset calculation accuracy conditions. When it is determined that the planning accuracy corresponding to the dynamic programming parameters meets the preset calculation accuracy conditions, the data calculation task is determined according to the dynamic programming parameters.

6. The data intelligent processing method applied to an industrial communication intelligent platform according to claim 1, characterized in that, The process involves performing data processing operations on all the data to be processed based on the data computation task and the data time series to obtain data processing results, including: Based on the data computing task, determine the task scheduling parameters corresponding to the data computing task, and determine the data computing resource parameters according to the task scheduling parameters; Based on the data time series, multidimensional feature sequence parameters are determined, wherein the multidimensional feature sequence parameters include time domain feature parameters, frequency domain feature parameters, and time-frequency domain feature parameters; Based on the data calculation resource parameters and the multidimensional feature sequence parameters, data calculation and processing parameters corresponding to all the data to be processed are generated, and data processing operations are performed on all the data to be processed based on the data calculation and processing parameters to obtain data processing results.

7. A data intelligent processing device applied to an industrial communication intelligent platform, characterized in that, The device includes: The acquisition module is used to acquire the data to be processed corresponding to the communication device; The determination module is used to determine the data computing task based on a preset heterogeneous computing resource scheduling algorithm and the data to be processed; The generation module is used to generate a data time series corresponding to all the data to be processed based on the data calculation task and all the data to be processed; The processing module is used to perform data processing operations on all the data to be processed based on the data calculation task and the data time series, and obtain data processing results; wherein, the data processing operations are used to achieve microsecond-level time synchronization and dynamic bandwidth allocation of the communication device; The analysis module is used to analyze the data processing results and obtain data analysis results; The generation module is also used to generate data optimization instructions corresponding to the communication device based on the data analysis results; An optimization module is used to perform optimization operations on the communication device that match the data optimization instructions; The specific methods by which the generation module generates the data optimization instructions corresponding to the communication device based on the data analysis results include: The data interaction information of the communication device is determined, and based on the data interaction information and the data analysis results, combined with a preset multi-protocol fusion model, the data interaction parameters corresponding to the data analysis results are generated. Based on the data interaction parameters and the preset edge computing model, a calculation operation is performed on the data interaction parameters to obtain the data interaction result. Based on the data interaction result, a target optimization instruction that matches the data interaction result is determined in a preset optimization instruction database. Based on all the target optimization instructions, generate the data optimization instructions corresponding to the communication device; The data optimization instructions are used to optimize seamless data interaction between heterogeneous devices.

8. A data intelligent processing device applied to an industrial communication intelligent platform, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the data intelligent processing method applied to the industrial communication intelligent platform as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the data intelligent processing method applied to an industrial communication intelligent platform as described in any one of claims 1-6.

Images