Cross-protocol cooperative processing method and device for multi-source detection equipment, equipment, storage medium and program product

Abstract

The application relates to a cross-protocol collaborative processing method and device for multi-source detection equipment, equipment, a storage medium and a program product. The method comprises the following steps: collecting original messages of different communication protocol equipment first; then converting the original messages into a multi-dimensional tensor according to the number of equipment, the number of sampling periods and the number of fields; obtaining a sub-tensor by setting a multi-scale preset time window to slide and slice the multi-dimensional tensor; calculating single-equipment data confidence weights based on the sub-tensor; obtaining corresponding fusion output values by fusing multi-equipment structured data under the same time window; and finally, carrying out quantitative scoring on the fusion output values of each scale time window, and selecting the highest comprehensive score as the cross-protocol collaborative processing result. The method can improve the accuracy and effectiveness of cross-protocol collaborative processing of multi-source detection equipment.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a cross-protocol collaborative processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for multi-source detection equipment. Background Technology

[0002] With the development of data processing technology and the continuous advancement of the construction of smart metrology laboratories, the number of multi-source testing devices deployed in the system is constantly increasing. The large amount of data generated by multi-source testing devices exhibits characteristics such as diverse protocols, different data formats, and inconsistent transmission methods.

[0003] Traditional technologies typically involve deploying fixed-format protocol parsers to interface with various testing devices. Raw data parsing and acquisition are completed manually using field mapping rules. Then, statically preset fusion rules are used to perform simple integration processing on the acquired multi-source data. Finally, the data is incorporated into a laboratory data management system for centralized storage and basic fusion applications of multi-source testing data. However, statically preset rules lack a comprehensive evaluation mechanism for fusion quality and automatic evolution capabilities. They cannot dynamically adjust the fusion window size, participating device range, or fusion method based on task type, data characteristics, or device status. This leads to system performance degradation in complex testing environments and makes it difficult to meet the high-precision metrological data fusion requirements. Summary of the Invention

[0004] Based on this, it is necessary to provide a cross-protocol collaborative processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for multi-source detection equipment that can take into account the stability, temporal variation characteristics, and systematic bias of observation data from different devices.

[0005] Firstly, this application provides a cross-protocol collaborative processing method for multi-source detection devices, including:

[0006] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0007] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0008] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0009] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0010] In one embodiment, the communication protocols include Modbus, OPC UA, MQTT, and CAN protocols; the multi-source detection devices include temperature detection devices, voltage detection devices, current detection devices, and frequency detection devices.

[0011] In one embodiment, the original message is transformed into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields, including:

[0012] Based on the field mapping rules in the protocol semantic abstraction engine of the runtime domain modeling language, the original message is parsed into detection device identifier, timestamp, field name, value and unit to obtain a structured data stream. The structured data stream is semantically aligned and the units are unified. According to the number of detection devices, the number of sampling periods and the number of fields, the structured data stream is transformed into a multidimensional tensor.

[0013] In one embodiment, the confidence weight of a single detection device as a data source is obtained based on the subtensor, including:

[0014] For subtensors, calculate the sum of squares of the differences between the observed values ​​and the mean values ​​of each field of a single detection device within the corresponding preset time window, as well as the product of the number of fields and the length of the corresponding preset time window;

[0015] The ratio between the sum of squared differences and the product is calculated as the observed fluctuation of a single detection device within the corresponding preset time window;

[0016] Based on the observed volatility, the confidence weight of each individual detection device as a data source is determined.

[0017] In one embodiment, based on the structured data of multiple detection devices within the same preset time window, the fusion output value for the corresponding preset time window is calculated, including:

[0018] For the structured data of multiple testing devices within the same preset time window, a weighted sum is calculated according to the confidence weight of each testing device to obtain a preliminary fusion value.

[0019] Calculate the sum of squared differences between the observed values ​​of all fields and the field mean values ​​of all detection devices within the corresponding preset time window; calculate the product of the sum of squared differences and the residual adjustment factor to obtain the residual adjustment term;

[0020] The residual adjustment term is added to the initial fusion value to obtain the fusion output value for the corresponding preset time window.

[0021] In one embodiment, for the fusion output value corresponding to a preset time window at each scale, a quantization score is performed, including:

[0022] For each scale, the cosine similarity between the fusion output value and the historical best observation value is calculated for the corresponding preset time window. The first ratio between the number of detection devices participating in the fusion and the total number of detection devices is calculated. The variance of the fusion output value at a consecutive preset number of time points is calculated. The second ratio between the fusion time and the preset maximum tolerable delay is calculated.

[0023] The cosine similarity, the first ratio, the fluctuation variance, and the second ratio are weighted and summed to obtain a comprehensive score.

[0024] Secondly, this application also provides a cross-protocol collaborative processing device for multi-source detection equipment, comprising:

[0025] The acquisition module is used to acquire raw messages detected by multi-source detection devices using different communication protocols;

[0026] The conversion module is used to convert the original message into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields;

[0027] The fusion module is used to set preset time windows at multiple scales; for each preset time window at each scale, the multidimensional tensor is sliced ​​by sliding according to the preset time window to obtain sub-tensors; based on the sub-tensors, the confidence weight of a single detection device as a data source is obtained; based on the structured data of multiple detection devices in the same preset time window, the fusion output value of the corresponding preset time window is calculated.

[0028] The scoring module is used to quantify and score the fusion output value corresponding to the preset time window for each scale; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0029] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0030] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0031] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0032] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0033] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0035] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0036] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0037] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0038] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0039] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0040] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0041] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0042] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0043] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0044] The aforementioned cross-protocol collaborative processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product for multi-source detection equipment first collects raw messages from devices using different communication protocols. Then, based on the number of devices, sampling periods, and fields, the raw messages are transformed into multi-dimensional tensors. The device, sampling, and field dimensional features of the multi-source data are organized in the form of multi-dimensional tensors, completely breaking down the data interoperability barriers between different communication protocols and improving the standardization of multi-source detection data. By setting multi-scale preset time windows, sliding slices are applied to the multi-dimensional tensors to obtain sub-tensors, taking into account the detection data features at different time granularities. Based on the sub-tensors, the confidence weight of single-device data is calculated, and the structured data of multiple devices under the same time window is fused to obtain the corresponding fused output value. Combined with the dynamic weighting of single-device confidence weights, the impact of single-device data bias is effectively avoided, improving the targeting and reliability of multi-source data fusion. Finally, the fused output values ​​of each time window are quantitatively scored, and the highest comprehensive score is selected as the cross-protocol collaborative processing result. By quantifying and scoring the multi-scale fusion output values ​​and selecting the optimal result, the system achieves the best decision-making for the fusion results across multiple time scales. This fully integrates detection information from different time dimensions, significantly improving the accuracy and effectiveness of cross-protocol collaborative processing of multi-source detection equipment, and providing high-quality fusion decision data for subsequent equipment status analysis and anomaly early warning. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is an application environment diagram of a cross-protocol collaborative processing method for multi-source detection devices in one embodiment.

[0047] Figure 2 This is a flowchart illustrating a cross-protocol collaborative processing method for multi-source detection devices in one embodiment;

[0048] Figure 3 This is a flowchart illustrating a cross-protocol collaborative processing method for multi-source detection devices in another embodiment;

[0049] Figure 4 This is a structural block diagram of a cross-protocol collaborative processing device for multi-source detection equipment in one embodiment;

[0050] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0053] The cross-protocol collaborative processing method for multi-source detection devices provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. First, raw messages from devices using different communication protocols are collected. Then, based on the number of devices, sampling periods, and fields, the raw messages are converted into multidimensional tensors. Sub-tensors are obtained by sliding slices of the multidimensional tensors through multi-scale preset time windows. The confidence weight of single-device data is calculated based on the sub-tensors, and the structured data from multiple devices within the same time window are fused to obtain the corresponding fused output value. Finally, the fused output values ​​for each scale time window are quantitatively scored, and the highest comprehensive score is selected as the cross-protocol collaborative processing result.

[0054] Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0055] In one exemplary embodiment, such as Figure 2 As shown, a cross-protocol collaborative processing method for multi-source detection devices is provided, which can be applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 202 to 208. Wherein:

[0056] Step 202: Collect the raw messages detected by multi-source detection devices using different communication protocols;

[0057] Multi-source detection equipment refers to various sensing / detection terminals deployed in detection scenarios to collect various status / parameter data. These hardware devices (such as PMUs, fault recorders, and temperature and humidity sensors in power systems) have different equipment types, manufacturers, and communication specifications, but possess data acquisition and external communication capabilities. Different communication protocols refer to the standardized communication rules and formats followed by multi-source detection equipment when transmitting data with external systems. Different devices use heterogeneous protocol types (such as Modbus, OPC UA, Profinet, MQTT, etc.), and the message formats, transmission rates, and command specifications of these protocols differ, forming a core barrier to multi-source data interoperability. Raw messages refer to the original data transmission units generated by the multi-source detection equipment according to its own communication protocol, without any parsing / processing / conversion. They are the original carriers of the equipment's detection data, containing original information such as the detection parameters collected by the equipment, equipment identification, timestamps, and protocol headers / footers. The format varies depending on the communication protocol.

[0058] Step 204: Convert the original message into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields;

[0059] The number of measuring devices refers to the total number of multi-source detection devices / channels participating in this data acquisition and completing the upload of the original message. This is the cardinality of the first dimension of the multidimensional tensor, representing the "device source dimension" of the data, used to distinguish detection data collected by different devices. The number of sampling periods refers to the time series length / number of samplings for continuous acquisition of the original messages from the multi-source detection devices, i.e., the number of time periods corresponding to the messages continuously acquired according to the device sampling frequency. This is the cardinality of the second dimension of the multidimensional tensor, representing the "time-series sampling dimension" of the data, used to reflect the changing characteristics of the detection data over time. The number of fields refers to the number of valid detection parameters (such as the number of independent detection indicators like voltage, current, temperature, and humidity) included in the original message generated by a single detection device in a single sampling. This is the cardinality of the third dimension of the multidimensional tensor, representing the "detection indicator dimension" of the data, used to distinguish different types of detection parameters collected by a single device.

[0060] Step 206: Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0061] Sliding slicing refers to the operation of continuously and sequentially truncating tensor data along the temporal sampling dimension of a multidimensional tensor, using a preset time window of a certain scale as a fixed truncation unit, according to a set step size (such as per sampling period). This operation only truncates the temporal dimension, preserving the integrity of the device and field dimensions, and is a key method for extracting local temporal data from a global multidimensional tensor. Confidence weight refers to a quantitative coefficient (usually ranging from 0 to 1) representing the reliability of data from a single detection device within a corresponding time window, calculated using algorithms such as data stability, bias, and consistency with similar devices, based on the sub-tensor data of that device as the analysis sample. The weight value is positively correlated with data reliability and serves as the basis for differentiated weighting during multi-source data fusion, providing reference value for distinguishing data from different devices.

[0062] Step 208: For the fusion output value corresponding to the preset time window of each scale, perform quantitative scoring; select the fusion output value with the highest comprehensive score as the collaborative processing result.

[0063] Among them, quantitative scoring refers to the process of quantitatively evaluating the fusion output value corresponding to each scale time window based on a preset quantitative evaluation index system (such as data stability, deviation from the benchmark value, consistency of multi-device data, and time series feature matching degree) and using standardized algorithms (such as weighted scoring method, membership function method, and comprehensive index method). The scoring result is a specific numerical value, and the value directly reflects the quality and effectiveness of the corresponding fusion output value, realizing the comparability judgment of fusion results at different scales.

[0064] In the aforementioned cross-protocol collaborative processing method for multi-source detection devices, the original messages from devices using different communication protocols are first collected. Then, based on the number of devices, the number of sampling periods, and the number of fields, the original messages are transformed into multi-dimensional tensors. The device, sampling, and field dimensional features of the multi-source data are organized in the form of multi-dimensional tensors, completely breaking down the data interoperability barriers between different communication protocols and improving the standardization of multi-source detection data. By setting multi-scale preset time windows, sliding slices are performed on the multi-dimensional tensors to obtain sub-tensors, taking into account the detection data features at different time granularities. The confidence weight of single-device data is calculated based on the sub-tensors, and the structured data of multiple devices under the same time window is fused to obtain the corresponding fusion output value. Combined with the dynamic weighting of single-device confidence weights, the impact of single-device data bias is effectively avoided, improving the targeting and reliability of multi-source data fusion. Finally, the fusion output values ​​of each scale time window are quantitatively scored, and the highest comprehensive score is selected as the cross-protocol collaborative processing result. By quantifying and scoring the multi-scale fusion output values ​​and selecting the optimal result, the system achieves the best decision-making for the fusion results across multiple time scales. This fully integrates detection information from different time dimensions, significantly improving the accuracy and effectiveness of cross-protocol collaborative processing of multi-source detection equipment, and providing high-quality fusion decision data for subsequent equipment status analysis and anomaly early warning.

[0065] In one embodiment, the communication protocols include Modbus, OPC UA, MQTT, and CAN protocols; the multi-source detection devices include temperature detection devices, voltage detection devices, current detection devices, and frequency detection devices.

[0066] In the above embodiments, by supporting multiple protocols and multi-source detection devices, and uniformly parsing and managing them on the same platform, the compatibility with heterogeneous detection devices is significantly improved. This solves the problems of inconsistent protocols and unparseable data structures between devices from different manufacturers in the existing smart metering system, breaks down communication barriers, builds a foundation for device-level protocol interoperability, and achieves true device access, identification, and fusion.

[0067] In one embodiment, the original message is transformed into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields, including:

[0068] Based on the field mapping rules in the protocol semantic abstraction engine of the runtime domain modeling language, the original message is parsed into detection device identifier, timestamp, field name, value and unit to obtain a structured data stream. The structured data stream is semantically aligned and the units are unified. According to the number of detection devices, the number of sampling periods and the number of fields, the structured data stream is transformed into a multidimensional tensor.

[0069] Specifically, the protocol semantic abstraction engine models the device communication process by introducing the existing communication behavior state machine model, abstracts the protocol interaction process into a state machine, identifies state nodes such as "initialization", "data reporting", and "abnormal response", and defines state transition rules according to the existing protocol frame structure. At the same time, it constructs a field mapping rule set by combining the existing key-value pair and semantic tag tree binding method, deconstructs the protocol messages of each monitoring device into a unified field expression form, and initially encapsulates them into a structured data stream.

[0070] Furthermore, the structured data stream undergoes semantic alignment and unit normalization using a pre-defined semantic label tree model. For example, semantically equivalent fields such as "Temp," "T," and "temperature value," but with different expressions, are uniformly mapped to the standard label "Temperature." Simultaneously, unit conversion rules (e.g., °F to °C, Pa to kPa) are employed to ensure logical comparability of all values, resulting in multi-source detection data with a unified data structure. After normalization using methods such as z-score, the data is integrated into a multi-dimensional standardized tensor. ,in, Represent real numbers; Indicates the number of testing devices; This indicates the number of sampling periods within a continuous time window; This indicates the number of normalized standard fields at each point in time, such as temperature, voltage, current, and frequency.

[0071] In the above embodiments, the original messages are parsed by the protocol semantic abstraction engine of the running domain modeling language and the exclusive field mapping rules, which achieves accurate and unified extraction of the core information of heterogeneous protocol messages, avoiding the errors and inefficiencies of manual parsing. The semantic alignment and unit unification of the structured data stream completely eliminate the semantic ambiguity and unit differences caused by different detection devices and communication protocols, laying a high-quality data foundation for subsequent tensor construction. The tensor transformation method with device, sampling period and field as dimensions enables the multi-source detection data to form a standardized representation with clear dimensions and a standardized structure. It not only retains the complete information such as the device source, time series changes and indicator characteristics of the data, but also improves the computability of the data and the efficiency of subsequent processing. It provides standardized and unified tensor data support for subsequent links such as multi-scale time window analysis and weighted fusion, which greatly improves the accuracy and standardization level of multi-source detection data structured processing.

[0072] In one embodiment, the confidence weight of a single detection device as a data source is obtained based on the subtensor, including:

[0073] For subtensors, calculate the sum of squares of the differences between the observed values ​​and the mean values ​​of each field of a single detection device within the corresponding preset time window, as well as the product of the number of fields and the length of the corresponding preset time window;

[0074] The ratio between the sum of squared differences and the product is calculated as the observed fluctuation of a single detection device within the corresponding preset time window;

[0075] Based on the observed volatility, the confidence weight of each individual detection device as a data source is determined.

[0076] Specifically, after completing the window slicing operation, based on the sub-tensor Perform local structural evaluation and confidence weight initialization. That is, quantify the confidence weights of each device. Output stability within the time window is used to adjust its participation weight in the fusion process. Specifically, for each sub-tensor... computing devices The observed volatility is:

[0077]

[0078] in, Indicates equipment In the window The local volatility (mean variance) within a window characterizes the stability of the device data within that window. Indicates a point-in-time index; Indicates device In time Upper The normalized values ​​of each field. This represents the mean of the field within this window. The formula above reflects the average variance of each dimension of the device's characteristics within a local time window; a smaller value indicates greater stability. Based on this value, the device is defined. In the window Confidence weights within :

[0079]

[0080] in, It is a very small positive number, used to avoid the denominator being zero, such as The confidence weight will directly affect the device's contribution to the result during fusion.

[0081] In the above embodiments, by combining the sum of squared differences with the ratio of the number of fields and the length of the time window, the precise and quantitative calculation of the data observation volatility of a single detection device within the corresponding time window is achieved. This calculation method can objectively reflect the stability and dispersion of the device data, providing a scientific and quantifiable basis for determining the confidence weight. Determining the confidence weight based on the observation volatility realizes a strong correlation between data quality and weight value, allowing devices with high data stability and low dispersion to obtain higher confidence weights, while devices with large data volatility have correspondingly lower weights, effectively avoiding the interference of low-stability device data on subsequent fusion results. This weight calculation method relies on the original sub-tensor data, without the need to introduce additional complex models. The calculation logic is simple and efficient, adapting to the batch processing needs of multi-scale time windows, and significantly improving the objectivity, accuracy, and calculation efficiency of confidence weight assignment.

[0082] In one embodiment, based on the structured data of multiple detection devices within the same preset time window, the fusion output value for the corresponding preset time window is calculated, including:

[0083] For the structured data of multiple testing devices within the same preset time window, a weighted sum is calculated according to the confidence weight of each testing device to obtain a preliminary fusion value.

[0084] Calculate the sum of squared differences between the observed values ​​of all fields and the field mean values ​​of all detection devices within the corresponding preset time window; calculate the product of the sum of squared differences and the residual adjustment factor to obtain the residual adjustment term;

[0085] The residual adjustment term is added to the initial fusion value to obtain the fusion output value for the corresponding preset time window.

[0086] Specifically, regarding the time point Top, window size is Bayesian fusion and residual correction were performed on all multi-source detection data with a unified data structure; the aim was to maintain high-quality fusion results even with observation biases and abrupt changes from different devices. A confidence-weighted average was used to construct the data for each time point. Time window Preliminary fusion value :

[0087]

[0088] in, It is the first Each device in time The above equation represents a typical confidence-weighted average, i.e., a Bayesian fusion process. To identify and correct systematic deviations between device observations and the overall mean, a residual adjustment mechanism is introduced to correct the initial fusion values, resulting in the corrected fusion output. :

[0089]

[0090] in, This is the corrected fused output. It is a residual adjustment term. This represents the average output of all devices within the current time window, i.e., the mean vector of all devices at the current time point. The residual adjustment coefficient represents the tolerance threshold for the degree of deviation. This formula comprehensively considers the variability weighting of equipment data and the correction for structural deviations, and is an extension of the standard Bayesian fusion model.

[0091] In the above embodiments, a preliminary fusion value is obtained by weighted summation of confidence weights, giving higher weight to data from high-reliability devices in the fusion result. This fully reflects the quality differences between different device data and ensures the basic reliability of the fusion result. At the same time, a residual adjustment term is introduced, which compensates and corrects the preliminary fusion value by combining the sum of squared differences with the residual adjustment coefficient. This effectively offsets the impact of overall bias and random error of multi-device data on the fusion result, further improving the fit between the fusion value and the actual detection value. This fusion calculation method takes into account the differentiated weights of device data and overall residual compensation. The calculation logic is clear and balances accuracy and efficiency. It can effectively integrate multi-source detection data under the same time window, outputting more accurate and robust fusion results, and providing high-quality fusion data support for subsequent quantitative scoring.

[0092] In one embodiment, for the fusion output value corresponding to a preset time window at each scale, a quantization score is performed, including:

[0093] For each scale, the cosine similarity between the fusion output value and the historical best observation value is calculated for the corresponding preset time window. The first ratio between the number of detection devices participating in the fusion and the total number of detection devices is calculated. The variance of the fusion output value at a consecutive preset number of time points is calculated. The second ratio between the fusion time and the preset maximum tolerable delay is calculated.

[0094] The cosine similarity, the first ratio, the fluctuation variance, and the second ratio are weighted and summed to obtain a comprehensive score.

[0095] Specifically, the comprehensive evaluation function is:

[0096]

[0097] in, It is in time Fusion Time Window The overall score for fusion quality is as follows; Indicates the accuracy weight, with a reference value range of [value missing]. ; This represents the coverage weight, with a reference value range of [value range missing]. ; This represents the stability weight, with a reference value range of [value missing]. ; All were determined using the entropy weight method; Cosine similarity is an accuracy indicator, representing the difference between the fusion result and the historical best observation. Cosine similarity between them; The first ratio is a coverage indicator, representing the proportion of participating devices under this window, which is determined through statistical analysis. The variance of fluctuation is a stability index, representing the degree of fluctuation of the fused output in the feature dimension, and is determined by the existing statistical variance calculation formula; The second ratio is the acceptable maximum fusion fluctuation threshold, set based on engineering experience; It is a delay indicator, on a scale The quantization time required to complete one fusion computation is obtained by running real-time measurements at the scale. The ratio of the time required to complete a fusion computation to the historical maximum or the set maximum tolerable latency.

[0098] In the above embodiments, the cosine similarity is used to measure the fit between the fused output value and the historical best value; the first ratio reflects the coverage of the participating fusion devices; the fluctuation variance characterizes the temporal stability of the fusion result; and the second ratio evaluates the timeliness of the fusion processing. This achieves a comprehensive and multi-dimensional quantitative evaluation of the fused output value. By integrating the four indicators through weighted summation, a comprehensive score is obtained. The weights of each indicator can be flexibly adjusted according to the actual detection scenario requirements, making the scoring results more in line with the actual analysis needs of the scenario and avoiding the one-sidedness of single-dimensional evaluation. This quantitative scoring method has clear logic and high computational efficiency. It can objectively and accurately reflect the comprehensive performance of the fused output value in terms of accuracy, coverage, stability, and timeliness at each scale time window. It provides a scientific and quantifiable basis for subsequent selection of the optimal collaborative processing result, greatly improving the rationality and reliability of the result selection decision.

[0099] In one embodiment, such as Figure 3 As shown, this is a cross-protocol collaborative processing method for multi-source detection devices in a specific embodiment, including:

[0100] S1. Obtain the original message, and after structured parsing, obtain the structured data stream. Then, perform semantic alignment and normalization processing, further integrate it into a standardized multidimensional tensor, and then perform scale sliding slicing processing to obtain a set of multi-scale sub-tensor sequences.

[0101] Specifically, in a multi-source detection scenario within a smart metrology laboratory, the raw messages from multi-source detection devices are acquired. The smart metrology laboratory records a total of... Each of the monitoring devices reports observation data to the system using its own communication protocol. Since the communication protocol formats used by different devices are not uniform, including but not limited to Modbus, OPC UA, MQTT, and CAN, a protocol semantic abstraction engine based on a Domain-Specific Language (DSL) is first used to perform structured parsing of all raw messages. The protocol semantic abstraction engine models the device communication process by introducing an existing communication behavior state machine model, abstracting the protocol interaction process into a state machine, identifying state nodes such as "initialization," "data reporting," and "abnormal response," and defining state transition rules based on the existing protocol frame structure. Simultaneously, by combining existing key-value pair and semantic tag tree binding methods to construct a field mapping rule set, the protocol messages of each monitoring device are deconstructed into a unified field expression form and initially encapsulated into a structured data stream.

[0102] Furthermore, the structured data stream undergoes semantic alignment and unit normalization using a pre-defined semantic label tree model. For example, semantically equivalent fields such as "Temp," "T," and "temperature value," but with different expressions, are uniformly mapped to the standard label "Temperature." Simultaneously, unit conversion rules (e.g., °F to °C, Pa to kPa) are employed to ensure logical comparability of all values, resulting in multi-source detection data with a unified data structure. After normalization using methods such as z-score, the data is integrated into standardized multidimensional tensors. ,in, Represent real numbers; Indicates the number of testing devices; This indicates the number of sampling periods within a continuous time window; This indicates the number of normalized standard fields at each point in time, such as temperature, voltage, current, and frequency.

[0103] Furthermore, in the fusion and collaborative processing of multi-source detection equipment, based on the time scale set by the technical personnel... (in, Represents the total number of time scales, set Any element express, (e.g., 10 seconds, 30 seconds, 60 seconds, etc.) A multi-timescale window segmentation method is used to perform multi-scale sliding slicing operations on the multidimensional normalized tensor, generating a set of multi-scale sub-tensor sequences. Each sub-tensor is denoted as... , indicating at a point in time The upper slice of all device observation data at time window s, i.e., from time point s. All multi-source detection data with a unified data structure; among them, Representing a multidimensional normalized tensor The selected time point; the purpose of the above operations is to capture the changing trend of local structure at different time granularities, and to ensure that subsequent fusion processing is not limited to single-scale modeling, thereby enhancing the ability to adapt to time series and perceive data fluctuations.

[0104] S2. Based on the multi-scale sub-tensor sequence set, local structure evaluation and confidence weight initialization are performed to determine the confidence weights. Bayesian fusion and residual correction are then performed to obtain the corrected fusion output. An evolutionary control mechanism is introduced to evaluate the strategy and select the optimal option to obtain the final fusion output.

[0105] Specifically, after completing the window slicing operation, based on the sub-tensor Perform local structural evaluation and confidence weight initialization. That is, quantify the confidence weights of each device. Output stability within the time window is used to adjust its participation weight in the fusion process. Specifically, for each sub-tensor... computing devices The observed volatility is:

[0106] (1)

[0107] in, Indicates device In the time window The local volatility (mean variance) within a window characterizes the stability of the device data within that window. Indicates a point-in-time index; Indicates device In time Upper The normalized values ​​of each field. This represents the mean of the field within this window. The formula above reflects the average variance of each dimension of the device's characteristics within a local time window; a smaller value indicates greater stability. Based on this value, the device is defined. In the window Confidence weights within :

[0108] (2)

[0109] in, It is a very small positive number, used to avoid the denominator being zero, such as The confidence weights directly affect the device's contribution to the result during fusion. This completes the standardization of the input data structure and the initialization of the confidence weights, providing precise parameter inputs for the next step of fusion processing.

[0110] After completing the confidence weight initialization, at time point Top, window size is Bayesian fusion and residual correction were performed on all multi-source detection data with a unified data structure; the aim was to maintain high-quality fusion results even with observation biases and abrupt changes from different devices. A confidence-weighted average was used to construct the data for each time point. Time window Preliminary fusion value :

[0111] (3)

[0112] in, It is the first Each device in time The above equation represents a typical confidence-weighted average, i.e., a Bayesian fusion process. To identify and correct systematic deviations between device observations and the overall mean, a residual adjustment mechanism is introduced to correct the initial fusion values, resulting in the corrected fusion output. :

[0113] (4)

[0114] in, This is the corrected fused output. It is a residual adjustment term. This represents the average output of all devices within the current time window, i.e., the mean vector of all devices at the current time point. The residual adjustment coefficient represents the tolerance threshold for the degree of deviation. This formula comprehensively considers the variability weighting of equipment data and the correction for structural deviations, and is an extension of the standard Bayesian fusion model. Fusion Output It maintains both robustness and flexibility, and relies heavily on the computational results of the first two stages.

[0115] After the fusion results are generated, an evolutionary control mechanism is introduced to evaluate strategies and select the optimal solution for results at each scale. The evolutionary control mechanism is implemented through a multi-index comprehensive evaluation function to balance the accuracy, stability, and real-time performance of the fusion results. The core objective is to select the optimal output at multiple time window scales. This improves the overall fusion performance.

[0116] The multi-index comprehensive evaluation function is as follows:

[0117] (5)

[0118] in, It is in time Fusion Time Window The overall score for fusion quality is as follows; Indicates the accuracy weight, with a reference value range of [value missing]. ; This represents the coverage weight, with a reference value range of [value range missing]. ; This represents the stability weight, with a reference value range of [value missing]. ; All were determined using the entropy weight method; Cosine similarity is an accuracy indicator, representing the difference between the fusion result and the historical best observation. Cosine similarity between them; The first ratio is a coverage indicator, representing the proportion of participating devices under this window, which is determined through statistical analysis. The variance of fluctuation is a stability index, representing the degree of fluctuation of the fused output in the feature dimension, and is determined by the existing statistical variance calculation formula; The second ratio is the acceptable maximum fusion fluctuation threshold, set based on engineering experience; It is a delay indicator, within the time window. The quantization time required to complete one fusion computation is obtained by running real-time measurements within a time window. The ratio of the time required to complete a fusion computation to the historical maximum or the set maximum tolerable latency.

[0119] The final fusion output selection makes The result under the largest window, i.e.:

[0120] (6)

[0121] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0122] Based on the same inventive concept, this application also provides a cross-protocol collaborative processing apparatus for multi-source detection devices to implement the cross-protocol collaborative processing method for multi-source detection devices described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method. Therefore, the specific limitations in one or more embodiments of the cross-protocol collaborative processing apparatus for multi-source detection devices provided below can be found in the limitations of the cross-protocol collaborative processing method for multi-source detection devices described above, and will not be repeated here.

[0123] In one exemplary embodiment, such as Figure 4 As shown, a cross-protocol collaborative processing device for multi-source detection equipment is provided, including: an acquisition module 402, a conversion module 404, a fusion module 406, and a scoring module 408, wherein:

[0124] The acquisition module 402 is used to acquire the raw messages detected by multi-source detection devices using different communication protocols;

[0125] The conversion module 404 is used to convert the original message into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields;

[0126] The fusion module 406 is used to set preset time windows of multiple scales; for each preset time window, the multidimensional tensor is sliced ​​according to the preset time window to obtain sub-tensors; based on the sub-tensors, the confidence weight of a single detection device as a data source is obtained; based on the structured data of multiple detection devices in the same preset time window, the fusion output value of the corresponding preset time window is calculated.

[0127] The scoring module 408 is used to quantify and score the fusion output value corresponding to the preset time window for each scale; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0128] In one embodiment, the communication protocols include Modbus, OPC UA, MQTT, and CAN protocols; the multi-source detection devices include temperature detection devices, voltage detection devices, current detection devices, and frequency detection devices.

[0129] In one embodiment, the conversion module 404 is further configured to:

[0130] Based on the field mapping rules in the protocol semantic abstraction engine of the runtime domain modeling language, the original message is parsed into detection device identifier, timestamp, field name, value and unit to obtain a structured data stream. The structured data stream is semantically aligned and the units are unified. According to the number of detection devices, the number of sampling periods and the number of fields, the structured data stream is transformed into a multidimensional tensor.

[0131] In one embodiment, the fusion module 406 is further configured to:

[0132] For subtensors, calculate the sum of squares of the differences between the observed values ​​and the mean values ​​of each field of a single detection device within the corresponding preset time window, as well as the product of the number of fields and the length of the corresponding preset time window;

[0133] The ratio between the sum of squared differences and the product is calculated as the observed fluctuation of a single detection device within the corresponding preset time window;

[0134] Based on the observed volatility, the confidence weight of each individual detection device as a data source is determined.

[0135] In one embodiment, the fusion module 406 is further configured to:

[0136] For the structured data of multiple testing devices within the same preset time window, a weighted sum is calculated according to the confidence weight of each testing device to obtain a preliminary fusion value.

[0137] Calculate the sum of squared differences between the observed values ​​of all fields and the field mean values ​​of all detection devices within the corresponding preset time window; calculate the product of the sum of squared differences and the residual adjustment factor to obtain the residual adjustment term;

[0138] The residual adjustment term is added to the initial fusion value to obtain the fusion output value for the corresponding preset time window.

[0139] In one embodiment, the scoring module 408 is further configured to:

[0140] For each scale, the cosine similarity between the fusion output value and the historical best observation value is calculated for the corresponding preset time window. The first ratio between the number of detection devices participating in the fusion and the total number of detection devices is calculated. The variance of the fusion output value at a consecutive preset number of time points is calculated. The second ratio between the fusion time and the preset maximum tolerable delay is calculated.

[0141] The cosine similarity, the first ratio, the fluctuation variance, and the second ratio are weighted and summed to obtain a comprehensive score.

[0142] Each module in the aforementioned cross-protocol collaborative processing device for multi-source detection equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0143] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a cross-protocol collaborative processing method for multi-source detection devices. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0144] Those skilled in the art will understand that Figure 5The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0145] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0146] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0147] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0148] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0149] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0150] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0151] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0152] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0153] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0154] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0155] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0156] Collect raw messages detected by multi-source detection devices using different communication protocols;

[0157] Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor;

[0158] Set multiple preset time windows at various scales; for each preset time window, slide slice the multidimensional tensor according to the preset time window to obtain sub-tensors; based on the sub-tensors, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window.

[0159] For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

[0160] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0161] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0162] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0163] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A cross-protocol collaborative processing method for multi-source detection equipment, characterized in that, The method includes: Collect raw messages detected by multi-source detection devices using different communication protocols; Based on the number of detection devices, the number of sampling cycles, and the number of fields, the original message is converted into a multidimensional tensor; Set multiple preset time windows at various scales; for each preset time window at various scales, slide slice the multidimensional tensor according to the preset time window to obtain a sub-tensor; based on the sub-tensor, obtain the confidence weight of a single detection device as a data source; based on the structured data of multiple detection devices in the same preset time window, calculate the fusion output value of the corresponding preset time window. For each scale, the corresponding fusion output value within the preset time window is quantitatively scored; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

2. The method according to claim 1, characterized in that, The communication protocols include Modbus, OPC UA, MQTT, and CAN protocols; the multi-source detection devices include temperature detection devices, voltage detection devices, current detection devices, and frequency detection devices.

3. The method according to claim 1, characterized in that, The process of converting the original message into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields includes: Based on the field mapping rules in the protocol semantic abstraction engine of the runtime domain modeling language, the original message is parsed into detection device identifier, timestamp, field name, value and unit to obtain a structured data stream; the structured data stream is semantically aligned and the units are unified, and the structured data stream is transformed into a multidimensional tensor according to the number of detection devices, the number of sampling periods and the number of fields.

4. The method according to claim 1, characterized in that, The step of obtaining the confidence weight of a single detection device as a data source based on the sub-tensor includes: For the subtensor, calculate the sum of squares of the differences between the observed values ​​and the mean values ​​of each field of a single detection device within the corresponding preset time window, and the product of the number of fields and the length of the corresponding preset time window; The ratio between the sum of squared differences and the product is calculated as the observed volatility of a single detection device within a corresponding preset time window. Based on the observed volatility, the confidence weight of each individual detection device as a data source is determined.

5. The method according to claim 1, characterized in that, The calculation of the fused output value for the corresponding preset time window based on the structured data of multiple detection devices within the same preset time window includes: For the structured data of multiple testing devices within the same preset time window, a weighted sum is calculated according to the confidence weight of each testing device to obtain a preliminary fusion value. Calculate the sum of squared differences between the observed values ​​of all fields and the mean values ​​of the fields within the corresponding preset time window for all detection devices; calculate the product of the sum of squared differences and the residual adjustment coefficient to obtain the residual adjustment term; The residual adjustment term is added to the preliminary fusion value to obtain the fusion output value for the corresponding preset time window.

6. The method according to claim 1, characterized in that, The quantification and scoring of the fusion output value corresponding to the preset time window for each scale includes: For each scale, the corresponding fusion output value within a preset time window is calculated, along with the cosine similarity between the fusion output value and the historical best observation value. A first ratio is calculated between the number of detection devices participating in the fusion and the total number of detection devices. The variance of the fusion output value at a consecutive preset number of time points is calculated. A second ratio is calculated between the fusion time and the preset maximum tolerable delay. The cosine similarity, the first ratio, the fluctuation variance, and the second ratio are weighted and summed to obtain a comprehensive score.

7. A cross-protocol collaborative processing device for multi-source detection equipment, characterized in that, The device includes: The acquisition module is used to acquire raw messages detected by multi-source detection devices using different communication protocols; The conversion module is used to convert the original message into a multidimensional tensor according to the number of detection devices, the number of sampling cycles, and the number of fields; The fusion module is used to set preset time windows of multiple scales; for each preset time window, the multidimensional tensor is sliced ​​according to the preset time window to obtain a sub-tensor; based on the sub-tensor, the confidence weight of a single detection device as a data source is obtained; based on the structured data of multiple detection devices in the same preset time window, the fusion output value of the corresponding preset time window is calculated. The scoring module is used to quantify and score the fusion output value corresponding to the preset time window for each scale; the fusion output value with the highest comprehensive score is selected as the collaborative processing result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images