Computer data processing method and platform based on Internet of Things, and storage medium

Abstract

The invention relates to the technical field of industrial equipment data processing, in particular to a computer data processing method and platform based on the Internet of Things and a storage medium, and the method comprises the steps: collecting sensor waveform data, channel data and process criticality, and analyzing a data feature vector based on the sensor waveform data; analyzing an equipment health index and a channel quality index based on the data feature vector and the channel data; constructing a priority scoring model based on the equipment health index, the channel quality index and the process criticality to obtain a priority score; dynamic response grading is realized according to priority scores; and collecting device transmission data, determining a preemption weight based on the device transmission data and the priority score, and constructing a channel preemption decision based on the preemption weight. According to the invention, accurate analysis and processing of the data transmission priority are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment data processing, and in particular to a computer data processing method, platform and storage medium based on the Internet of Things. Background Art

[0002] In the Industrial Internet of Things (IIoT), massive amounts of equipment generate a wide range of data in real time, including production logs and status data. Traditional transmission methods can easily lead to network congestion and delays in critical data, impacting production. Therefore, data transmission priority management is necessary to rationally allocate resources, ensure timely processing of critical data, and improve production efficiency and stability.

[0003] Current computer data processing methods based on the Internet of Things generally have the following defects: traditional methods rely on device types or fixed rules to assign transmission priorities, and cannot dynamically respond to changes in the real-time health status of devices and fluctuations in channel quality, resulting in high-value data being delayed due to queue congestion or channel degradation; existing technologies ignore the topological correlation of equipment in the production chain and do not quantify the impact of failures on upstream and downstream, which may cause non-critical equipment to excessively occupy resources, while actual high-risk node data is not prioritized. Communication layer decisions are usually independent of business logic and do not take into account the urgency of data transmission. It is difficult to quickly seize resources in emergency scenarios, affecting the real-time nature of fault diagnosis and control. Sensor data analysis is mostly limited to time domain parameters, ignoring frequency domain characteristics and fluctuation characteristics, reducing anomaly detection coverage, and making it difficult to support accurate health assessments. Summary of the Invention

[0004] The object of the present invention is to provide a computer data processing method, platform and storage medium based on the Internet of Things to solve at least one of the problems existing in the prior art.

[0005] To achieve the above object, the present invention adopts the following technical solutions: A computer data processing method based on the Internet of Things, comprising: Collect sensor waveform data, channel data and process criticality, and analyze data feature vectors based on the sensor waveform data; Analyze device health index and channel quality index based on data feature vector and channel data; A priority scoring model is constructed based on the equipment health index, channel quality index, and process criticality to obtain a priority score; Dynamic response grading based on priority scores; Collect device transmission data, determine preemption weights based on device transmission data and priority scores, and build channel preemption decisions based on the preemption weights.

[0006] Furthermore, the peak amplitude parameter, kurtosis coefficient, frequency band energy ratio and amplitude fluctuation standard deviation are analyzed based on the sensor waveform data. The peak amplitude parameter is the difference between the maximum and minimum values of the sensor waveform data. The analysis method of the frequency band energy ratio is to perform fast Fourier transform on the sensor waveform data to obtain the data spectrum, and the energy proportion of 3kHz-5kHz in the data spectrum is used as the frequency band energy ratio. The analysis method of the amplitude fluctuation standard deviation is to divide the sensor waveform data into multiple groups of window sensor waveform data with a window size of 10ms, and calculate the peak amplitude parameter of each window sensor waveform data, and use the standard deviation of the peak amplitude parameter of each window sensor waveform data as the amplitude fluctuation standard deviation.

[0007] Furthermore, a data feature vector is constructed based on the peak amplitude parameter, kurtosis coefficient, frequency band energy ratio and amplitude fluctuation standard deviation, and the data feature vector is set to F, F=(A,K,E,DA), where F represents the data feature vector, A represents the peak amplitude parameter, K represents the kurtosis coefficient, E represents the frequency band energy ratio, and DA represents the amplitude fluctuation standard deviation.

[0008] Furthermore, weights are assigned to the peak amplitude parameter, kurtosis coefficient, and frequency band energy ratio to obtain a device health index. The expression of the device health index is: H=w1×A / a+w2×|Kk|+w3×E, where H represents the device health index, w1 represents the first health weight, w2 represents the second health weight, w3 represents the third health weight, w1+w2+w3=1, a represents the peak amplitude threshold, and k represents the kurtosis threshold.

[0009] Furthermore, the channel quality index is analyzed based on the packet loss rate, signal strength and number of channel conflicts. The expression of the channel quality index is: Q=[e -λ×PL +R / r+1 / (1+0.2×NC)] / 3, where Q represents the channel quality index, λ represents the preset packet loss parameter, PL represents the packet loss rate, R represents the signal strength, r represents the preset signal strength parameter, and NC represents the number of channel conflicts.

[0010] Furthermore, dynamic weights are assigned to the equipment health index, channel quality index, and process criticality to obtain a priority score. The expression of the priority score is: Sc=α1×H+α2×Q+α3×C, where Sc represents the priority score, α1 represents the first priority weight, α2 represents the second priority weight, α3 represents the third priority weight, α1+α2+α3=1, and C represents the process criticality.

[0011] Furthermore, the response level and response action are analyzed based on the priority score. If the priority score is less than the first priority threshold, the response level is set to low, and the response action is conventional batch processing; if the priority score is greater than or equal to the first priority threshold and less than the second priority threshold, the response level is set to medium, and the response action is to move the batch queue forward to ensure that the transmission delay is less than 200ms; if the priority score is greater than or equal to the second priority threshold and less than the third priority threshold, the response level is set to high, and the response action is to compress the sensor waveform data to retain data frames greater than 3kHz for transmission; if the priority score is greater than or equal to the third priority threshold, the response level is set to urgent, and the response action is to seize a high-bandwidth channel for transmission.

[0012] Furthermore, the preemption weight is analyzed based on the transmission rate, the amount of data to be transmitted and the priority score. The expression of the preemption weight is: WP=Sc×e -G×1000 / (V×η) , where WP represents the preemption weight, G represents the amount of data to be transmitted, V represents the transmission rate, and η represents the preset time parameter; When the response level is emergency, if the preemption weight is greater than or equal to the preemption threshold, channel switching is triggered and the channel of the industrial device with the lowest preemption weight is selected for data transmission; When the response level is high, if the priority score is greater than or equal to [(second priority threshold + third priority threshold) / 2] and the channel quality index is less than or equal to the channel quality threshold, dual-path parallel transmission is started. Otherwise, the channel with the largest channel quality index is selected for transmission.

[0013] On the other hand, the present invention also provides a computer data processing platform based on the Internet of Things, comprising: A data acquisition unit, configured to collect sensor waveform data, channel data, and process criticality, and analyze data feature vectors based on the sensor waveform data; a parameter analysis unit, configured to analyze a device health index and a channel quality index based on a data feature vector and channel data; A model building unit, configured to build a priority scoring model based on a device health index, a channel quality index, and a process criticality to obtain a priority score; A hierarchical response unit, used to implement dynamic response grading based on priority scores; The decision analysis unit is used to collect device transmission data, determine the preemption weight based on the device transmission data and priority score, and build a channel preemption decision based on the preemption weight.

[0014] On the other hand, the present invention further provides a storage medium, characterized in that it stores instructions, which, when executed on a computer, enable the computer to execute any of the above methods.

[0015] The beneficial effects of the present invention are as follows: by dynamically coupling the equipment operating status, communication environment and production logic, a priority scoring model is constructed to solve the limitations of traditional methods that rely solely on equipment types or static rules, and to achieve accurate matching of data transmission strategies with real-time production needs. Through dynamic response classification and channel preemption mechanism, low-latency transmission of critical data is ensured under limited bandwidth. The data compression strategy specifically retains fault feature frequency bands, reduces invalid bandwidth occupancy, and improves overall network utilization. The process criticality calculation based on breadth-first search can identify core equipment that may cause chain shutdowns in advance. The health index's sensitive analysis of vibration characteristics supports early fault warning. The combination of the two greatly reduces the risk of unplanned production stoppages and ensures continuous and stable operation of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 Flowchart of the computer data processing method based on the Internet of Things in this embodiment.

[0018] Figure 2 This is a flow chart of the method for analyzing the device health index and channel quality index in this embodiment.

[0019] Figure 3 This is a flowchart of the method for constructing preemption decision in this embodiment.

[0020] Figure 4 Schematic diagram of the structure of the computer data processing platform based on the Internet of Things in this embodiment. DETAILED DESCRIPTION

[0021] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the following detailed description is illustrative rather than restrictive and should not be used to limit the scope of protection of the present invention.

[0022] It should be noted that, although the terms "first," "second," and "third" may be used to describe the embodiments of the present application, the description should not be limited to these terms. These terms are merely used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first."

[0023] See also Figure 1As shown, this is a computer data processing method based on the Internet of Things in this embodiment, including: Step S1: Collect sensor waveform data, channel data, and process criticality, and analyze data feature vectors based on the sensor waveform data. The sensor waveform data is the original waveform data of the industrial equipment vibration sensor. In this embodiment, the acquisition frequency of the sensor waveform data is 8kHz. The channel data includes packet loss rate, signal strength, and number of channel conflicts. The packet loss rate is the packet loss rate within the past 5 minutes. The number of channel conflicts is the number of times multiple industrial devices simultaneously attempt to use the same channel for communication within 10 seconds. The unit of the signal strength is dBm.

[0024] Specifically, the process criticality in this embodiment is the maximum value of the node impact value, which is calculated based on the process topology network. The analysis method of the node impact value is as follows: Each industrial device is abstracted as a network node, and the terminal node is labeled as the product output point. The industrial device flow direction is constructed. The industrial device flow direction is the working order of each industrial device during the manufacturing process. The coupling strength of industrial devices with adjacent working orders is labeled. The coupling strength is a parameter based on the proportion of downstream device efficiency reduction caused by industrial device failure. For example, if the equipment failure causes 100% shutdown of downstream equipment, the coupling strength is set to 1. If the equipment failure causes a 30% reduction in downstream equipment efficiency, the coupling strength is set to 0.3. Key parameters are assigned based on the attributes of industrial equipment. The key parameters include equipment redundancy parameters, fault repair parameters, alternative equipment existence parameters, and historical impact parameters. The equipment redundancy parameters are determined based on whether the equipment data is backed up, such as being set to 1 when there is no backup and to 0.3 when there are dual backups. The fault repair parameters are set based on the fault repair time, such as being set to 1 when the equipment fault repair time is greater than 4 hours and to 0.2 when it is less than 30 minutes. The alternative equipment existence parameter is set based on whether there is a backup device, such as being set to 1 when there is no alternative device and to 0.4 when there is an alternative device. The historical impact parameter is set based on the scope of equipment shutdown caused by the equipment failure, such as being set to 1 when the equipment failure causes a full line shutdown and to 0.3 when only a partial shutdown occurs. The basic key value of the industrial equipment is calculated based on the key parameters. The basic key value is the average value of the key parameters. Use breadth-first search to traverse the downstream nodes of each network node and calculate the node influence value, which is the sum of the product of the basic key value of the network node and the coupling strength × e (-0.2×路径长度) , the path length is an integer, which is the number of network nodes passed between the network node and the product output point.

[0025] Specifically, in step S1 of this embodiment, the peak amplitude parameter, kurtosis coefficient, band energy ratio, and amplitude fluctuation standard deviation are analyzed based on the sensor waveform data. The peak amplitude parameter is the difference between the maximum and minimum values of the sensor waveform data. The expression of the kurtosis coefficient is: , where K represents the kurtosis coefficient, s i Represents sensor waveform data, i represents the acquisition sequence number, Ni represents the number of acquisition sequence numbers, μ() represents the average value of the data in the brackets, σ() represents the standard deviation of the data in the brackets, the frequency band energy ratio analysis method is to perform fast Fourier transform on the sensor waveform data to obtain the data spectrum, and the energy proportion of 3kHz-5kHz in the data spectrum is used as the frequency band energy ratio, and the amplitude fluctuation standard deviation analysis method is to divide the sensor waveform data into multiple groups of window sensor waveform data with a window size of 10ms, and calculate the peak amplitude parameter of each window sensor waveform data, and use the standard deviation of the peak amplitude parameter of each window sensor waveform data as the amplitude fluctuation standard deviation.

[0026] Specifically, in step S1 described in this embodiment, a data feature vector is constructed based on the peak amplitude parameter, kurtosis coefficient, frequency band energy ratio and amplitude fluctuation standard deviation, and the data feature vector is set to F, F=(A, K, E, DA), where F represents the data feature vector, A represents the peak amplitude parameter, E represents the frequency band energy ratio, and DA represents the amplitude fluctuation standard deviation.

[0027] Specifically, in step S1 described in this embodiment, the peak amplitude, kurtosis coefficient, frequency band energy ratio, and amplitude fluctuation standard deviation are analyzed through multi-dimensional sensor waveform analysis to achieve refined capture of the equipment operating status. The frequency band energy ratio focuses on a specific high-frequency band to effectively identify potential fault characteristics such as mechanical wear; the window calculation mechanism of the amplitude fluctuation standard deviation enhances the detection sensitivity of transient anomalies of the vibration signal; the process criticality is based on the dynamic calculation of the node impact value of the topological network, and comprehensively considers the equipment redundancy, repair difficulty and production coupling, avoiding the risk of misjudgment of key equipment by traditional static classification.

[0028] Please continue reading Figure 1 As shown, the computer data processing method based on the Internet of Things also includes: Step S2: Analyze the device health index and channel quality index based on the data feature vector and the channel data.

[0029] See also Figure 2 As shown in Figure 1, the analysis method for the device health index and channel quality index includes: Step S21: Analyze the device health index based on the data feature vector.

[0030] Specifically, in step S21 described in this embodiment, weights are assigned to the peak amplitude parameter, kurtosis coefficient, and frequency band energy ratio to obtain a device health index. The expression of the device health index is: H=w1×A / a+w2×|Kk|+w3×E, where H represents the device health index, w1 represents the first health weight, w2 represents the second health weight, w3 represents the third health weight, w1+w2+w3=1, a represents the peak amplitude threshold, and k represents the kurtosis threshold.

[0031] Specifically, in this embodiment, the first health weight is set to 0.5, the second health weight is set to 0.3, the third health weight is set to 0.2, the peak amplitude threshold is set to 30V, and the kurtosis threshold is set to 3. In this embodiment, there is no specific limitation on the setting of the health weight threshold, peak amplitude threshold and kurtosis threshold, and those skilled in the art can set them freely.

[0032] Please continue reading Figure 2 As shown, the device health index and channel quality index analysis method further includes: Step S22: Analyze the channel quality index based on the channel data.

[0033] Specifically, in step S22 of this embodiment, the channel quality index is analyzed based on the packet loss rate, signal strength and number of channel conflicts. The expression of the channel quality index is: Q=[e -λ×PL +R / r+1 / (1+0.2×NC)] / 3, where Q represents the channel quality index, λ represents the preset packet loss parameter, PL represents the packet loss rate, R represents the signal strength, r represents the preset signal strength parameter, and NC represents the number of channel conflicts.

[0034] Specifically, in this embodiment, the preset packet loss parameter is set to 0.05, and the preset signal strength parameter is set to -70dBm. In this embodiment, no specific limitation is imposed on the settings of the preset packet loss parameter and the preset signal strength parameter, and those skilled in the art may freely set them.

[0035] Specifically, in step S2 described in this embodiment, the device health index integrates physical characteristics and preset threshold differences, highlights the contribution of core parameters through weight distribution, and thus improves the accuracy of fault warning; the channel quality index innovatively integrates packet loss rate, signal strength, and number of conflicts, and uses an exponential decay function to weaken occasional packet loss interference, while strengthening the quality degradation representation of high-conflict channels, providing a reliable basis for dynamic channel selection.

[0036] Please continue reading Figure 1 As shown, the computer data processing method based on the Internet of Things also includes: Step S3: constructing a priority scoring model based on the device health index, the channel quality index, and the process criticality to obtain a priority score.

[0037] Specifically, in step S3 described in this embodiment, dynamic weights are assigned to the equipment health index, channel quality index, and process criticality to obtain a priority score. The expression of the priority score is: Sc=α1×H+α2×Q+α3×C, where Sc represents the priority score, α1 represents the first priority weight, α2 represents the second priority weight, α3 represents the third priority weight, α1+α2+α3=1, and C represents the process criticality.

[0038] Specifically, the priority weights in this embodiment are priority weights preset according to different industrial equipment. For example, the first priority weight of the compressor in the industrial refrigeration equipment is set to 0.7, the second priority weight is set to 0.1, and the third priority weight is set to 0.2. The first priority weight of the cooling water pump is set to 0.5, the second priority weight is set to 0.3, and the third priority weight is set to 0.2, etc. In this embodiment, there is no specific limitation on the setting of the priority weights, and those skilled in the art can set them freely, as long as they meet the analysis of the priority scores of different industrial equipment.

[0039] Specifically, in step S3 described in this embodiment, a dynamic weight allocation model based on equipment health, channel quality and process criticality is used to break through the traditional fixed weight limitations. By differentially configuring weights, the actual importance of different equipment in the production chain is accurately adapted to ensure that high-value data is processed first.

[0040] Please continue reading Figure 1 As shown, the computer data processing method based on the Internet of Things also includes: Step S4: Implement dynamic response grading based on priority scores.

[0041] Specifically, in step S4 described in this embodiment, the response level and response action are analyzed based on the priority score. If the priority score is less than the first priority threshold, the response level is set to low, and the response action is conventional batch processing; if the priority score is greater than or equal to the first priority threshold and less than the second priority threshold, the response level is set to medium, and the response action is to move the batch queue forward to ensure that the transmission delay is less than 200ms; if the priority score is greater than or equal to the second priority threshold and less than the third priority threshold, the response level is set to high, and the response action is to compress the sensor waveform data to retain data frames greater than 3kHz for transmission; if the priority score is greater than or equal to the third priority threshold, the response level is set to urgent, and the response action is to seize a high-bandwidth channel for transmission.

[0042] Specifically, in this embodiment, the first priority threshold is set to 0.3, the second priority threshold is set to 0.5, and the third priority threshold is set to 0.8. In this embodiment, there is no specific limitation on the setting of the priority threshold, and those skilled in the art can set it freely as long as it meets the division of response levels.

[0043] Specifically, in step S4 described in this embodiment, the four-level response mechanism triggers differentiated actions based on the scoring threshold: the medium level realizes queue advancement and delay control, the high level uses data compression to retain key frequency bands, and the emergency level triggers channel preemption, thereby achieving a balance between resource efficiency and real-time requirements, and avoiding low-priority data blocking key transmission paths.

[0044] Please continue reading Figure 1 As shown, the computer data processing method based on the Internet of Things also includes: Step S5: collecting device transmission data, determining a preemption weight based on the device transmission data and the priority score, and constructing a channel preemption decision based on the preemption weight.

[0045] See also Figure 3 As shown in FIG, it is a method for constructing a preemption decision, including: Step S51 : collecting device transmission data, wherein the device transmission data includes a transmission rate and a volume of data to be transmitted, wherein the unit of the transmission rate is Kbps, and the unit of the volume of data to be transmitted is KB.

[0046] Please continue reading Figure 3 As shown, the method for constructing the preemption decision further includes: Step S52: analyzing the preemption weight according to the transmission rate, the amount of data to be transmitted, and the priority score.

[0047] Specifically, in step S52 of this embodiment, the preemption weight is analyzed based on the transmission rate, the amount of data to be transmitted and the priority score. The expression of the preemption weight is: WP=Sc×e -G×1000 / (V×η) , where WP represents the preemption weight, G represents the amount of data to be transmitted, V represents the transmission rate, and η represents the preset time parameter.

[0048] Specifically, the preset time parameter in this embodiment is set to 50ms. In this embodiment, no specific limitation is imposed on the setting of the preset time parameter, and those skilled in the art may freely set it.

[0049] Please continue reading Figure 3 As shown, the method for constructing the preemption decision further includes: Step S53: construct a channel preemption decision based on the preemption weight.

[0050] Specifically, in step S53 of this embodiment, when the response level is emergency, if the preemption weight is greater than or equal to the preemption threshold, channel switching is triggered, and the channel of the industrial device with the lowest preemption weight is selected for data transmission; When the response level is high, if the priority score is greater than or equal to [(second priority threshold + third priority threshold) / 2] and the channel quality index is less than or equal to the channel quality threshold, dual-path parallel transmission is started. Otherwise, the channel with the largest channel quality index is selected for transmission.

[0051] Specifically, in this embodiment, the preemption threshold is set to 1.2, and the channel quality threshold is set to 0.6. In this embodiment, there is no specific limitation on the setting of the preemption threshold and the channel quality threshold. Those skilled in the art can freely set them as long as they meet the requirements for making preemption decisions.

[0052] Specifically, in step S5 described in this embodiment, the preemption weight introduces a time sensitivity factor for the amount and rate of transmitted data, and channel switching is implemented based on weight comparison in an emergency state, effectively solving the sudden high-load transmission needs; the intelligent trigger conditions of high-level dual-path transmission are used to significantly improve the robustness of data transmission.

[0053] See also Figure 4 As shown, it is a computer data processing platform based on the Internet of Things in this embodiment, including: A data acquisition unit, configured to collect sensor waveform data, channel data, and process criticality, and analyze data feature vectors based on the sensor waveform data; a parameter analysis unit, configured to analyze a device health index and a channel quality index based on a data feature vector and channel data; A model building unit, configured to build a priority scoring model based on a device health index, a channel quality index, and a process criticality to obtain a priority score; A hierarchical response unit, used to implement dynamic response grading based on priority scores; The decision analysis unit is used to collect device transmission data, determine the preemption weight based on the device transmission data and priority score, and build a channel preemption decision based on the preemption weight.

[0054] An embodiment of the present application also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enables the computer to execute the computer data processing method based on the Internet of Things as described in the above method embodiment.

[0055] Those skilled in the art will appreciate that all or some of the steps and systems in the method disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include computer storage media (or non-transitory media) and communication media (or temporary media). As known to those skilled in the art, the term computer storage media is included in any method or technology for storing information (such as a computer-readable program, a data structure, a program module, or other data) and is volatile and non-volatile, removable, and non-removable. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory, or other memory technology, CD-ROM, digital versatile disks (DVD), or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage, or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, as is well known to those skilled in the art, communication media typically embodies computer-readable programs, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0056] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in this field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A computer data processing method based on the Internet of Things, characterized in that: include: Collect sensor waveform data, channel data and process criticality, and analyze data feature vectors based on the sensor waveform data; Analyze device health index and channel quality index based on data feature vector and channel data; A priority scoring model is built based on the device health index, channel quality index, and process criticality to obtain a priority score; Dynamic response grading based on priority scores; Collect device transmission data, determine preemption weights based on device transmission data and priority scores, and build channel preemption decisions based on the preemption weights.

2. The computer data processing method based on the Internet of Things according to claim 1, characterized in that: Based on the sensor waveform data, the peak amplitude parameter, kurtosis coefficient, frequency band energy ratio and amplitude fluctuation standard deviation are analyzed. The peak amplitude parameter is the difference between the maximum and minimum values of the sensor waveform data. The frequency band energy ratio is analyzed by performing a fast Fourier transform on the sensor waveform data to obtain a data spectrum, and the energy proportion of 3kHz-5kHz in the data spectrum is used as the frequency band energy ratio. The amplitude fluctuation standard deviation is analyzed by dividing the sensor waveform data into multiple groups of window sensor waveform data with a window size of 10ms, and calculating the peak amplitude parameter of each window sensor waveform data, and using the standard deviation of the peak amplitude parameter of each window sensor waveform data as the amplitude fluctuation standard deviation.

3. The computer data processing method based on the Internet of Things according to claim 2, characterized in that: The data feature vector is constructed based on the peak amplitude parameter, kurtosis coefficient, frequency band energy ratio and amplitude fluctuation standard deviation, and the data feature vector is set to F, F=(A,K,E,DA), where F represents the data feature vector, A represents the peak amplitude parameter, K represents the kurtosis coefficient, E represents the frequency band energy ratio, and DA represents the amplitude fluctuation standard deviation.

4. The computer data processing method based on the Internet of Things according to claim 3, characterized in that: Weights are assigned to the peak amplitude parameter, kurtosis coefficient, and frequency band energy ratio to obtain the device health index. The expression of the device health index is: H=w1×A / a+w2×|Kk|+w3×E, where H represents the device health index, w1 represents the first health weight, w2 represents the second health weight, w3 represents the third health weight, w1+w2+w3=1, a represents the peak amplitude threshold, and k represents the kurtosis threshold.

5. The computer data processing method based on the Internet of Things according to claim 4, characterized in that: The channel quality index is analyzed based on the packet loss rate, signal strength and number of channel conflicts. The expression of the channel quality index is: Q=[e -λ×PL +R / r+1 / (1+0.2×NC)] / 3, where Q represents the channel quality index, λ represents the preset packet loss parameter, PL represents the packet loss rate, R represents the signal strength, r represents the preset signal strength parameter, and NC represents the number of channel conflicts.

6. The computer data processing method based on the Internet of Things according to claim 5, characterized in that: Dynamically assign weights to the device health index, channel quality index, and process criticality to obtain a priority score. The expression for the priority score is: Sc=α1×H+α2×Q+α3×C, where Sc represents the priority score, α1 represents the first priority weight, α2 represents the second priority weight, α3 represents the third priority weight, α1+α2+α3=1, and C represents the process criticality.

7. The computer data processing method based on the Internet of Things according to claim 6, characterized in that: Analyze the response level and response action based on the priority score. If the priority score is less than the first priority threshold, set the response level to low and the response action to regular batch processing. If the priority score is greater than or equal to the first priority threshold and less than the second priority threshold, the response level is set to medium, and the response action is to move the batch queue forward to ensure that the transmission delay is less than 200ms; if the priority score is greater than or equal to the second priority threshold and less than the third priority threshold, the response level is set to high, and the response action is to compress the sensor waveform data to retain data frames larger than 3kHz for transmission; if the priority score is greater than or equal to the third priority threshold, the response level is set to urgent, and the response action is to seize a high-bandwidth channel for transmission.

8. The computer data processing method based on the Internet of Things according to claim 7, characterized in that: The preemption weight is analyzed based on the transmission rate, the amount of data to be transmitted and the priority score. The expression of the preemption weight is: WP=Sc×e -G×1000 / (V×η) , where WP represents the preemption weight, G represents the amount of data to be transmitted, V represents the transmission rate, and η represents the preset time parameter; When the response level is emergency, if the preemption weight is greater than or equal to the preemption threshold, channel switching is triggered and the channel of the industrial device with the lowest preemption weight is selected for data transmission; When the response level is high, if the priority score is greater than or equal to [(second priority threshold + third priority threshold) / 2] and the channel quality index is less than or equal to the channel quality threshold, dual-path parallel transmission is started. Otherwise, the channel with the largest channel quality index is selected for transmission.

9. A computer data processing platform based on the Internet of Things, applied to the computer data processing method based on the Internet of Things according to any one of claims 1 to 8, characterized in that: include: A data acquisition unit, configured to collect sensor waveform data, channel data, and process criticality, and analyze data feature vectors based on the sensor waveform data; a parameter analysis unit, configured to analyze a device health index and a channel quality index based on a data feature vector and channel data; A model building unit, configured to build a priority scoring model based on a device health index, a channel quality index, and a process criticality to obtain a priority score; A hierarchical response unit, used to implement dynamic response grading based on priority scores; The decision analysis unit is used to collect device transmission data, determine the preemption weight based on the device transmission data and priority score, and build a channel preemption decision based on the preemption weight.

10. A storage medium, characterized in that: Instructions are stored, which, when run on a computer, enable the computer to execute the computer data processing method based on the Internet of Things as described in any one of claims 1 to 8.

Images