An industrial internet of things system and method for transmitter line maintenance

Abstract

The application provides an industrial Internet of Things system and method for transmitter online maintenance, the system comprising an industrial Internet of Things management platform configured to: acquire monitoring data of a transmitter and working condition state data of a monitored device within a first preset time period; determine a cross-correlation coefficient according to the monitoring data and the working condition state data; in response to the cross-correlation coefficient being less than a coefficient threshold, identify an abnormal fluctuation mode corresponding to the monitoring data; determine a reliability index of the transmitter according to data features of the monitoring data; in response to the reliability index being less than a first reliability threshold, generate a filtering parameter according to a signal-to-noise ratio and a spectral distribution feature of the monitoring data, and control the transmitter to adjust a filtering window and / or a cutoff frequency based on the filtering parameter; and in response to the reliability index being less than or equal to a second reliability threshold, control the monitored device to shut down, trigger an alarm and generate a manual maintenance work order.

Description

Technical Field

[0001] This specification relates to the field of transmission maintenance technology, and in particular to an industrial Internet of Things system and method for online maintenance of transmitters. Background Technology

[0002] In industrial settings (such as chemical plants and power plants), transmitters (such as pressure transmitters, temperature transmitters, and vibration transmitters) are used to acquire real-time equipment operating status (such as pipeline pressure and temperature) and to convert the monitored physical quantities (such as pressure and temperature) into standard signals for remote transmission. However, industrial environments often present harsh conditions such as strong electromagnetic interference, high temperatures, and vibrations, which can cause abnormal fluctuations in transmitter data. Traditional data analysis methods struggle to accurately determine whether the problem stems from a problem with the monitored equipment (such as a sudden pressure surge due to pipeline blockage) or from performance degradation or malfunction of the transmitter itself (such as zero-point drift, clock drift, or communication delay).

[0003] Currently, there is a lack of effective fault tracing methods. Maintenance personnel often rely on experience to make judgments or frequently replace transmitters, resulting in high maintenance costs and a high risk of misjudgment.

[0004] Based on this, it is desired to provide an industrial IoT system, method, and storage medium for online maintenance of transmitters, which can accurately distinguish between actual equipment abnormalities and transmitter faults in high-noise environments, thereby improving the reliability and maintenance efficiency of industrial monitoring systems. Summary of the Invention

[0005] This specification provides one or more embodiments of an industrial IoT system for online maintenance of transmitters, including an industrial IoT management platform configured to perform an industrial IoT method for online maintenance of transmitters.

[0006] This specification provides one or more embodiments of an industrial IoT method for online maintenance of transmitters, implemented based on an IoT system. The method includes: acquiring monitoring data of the transmitter and operating status data of the monitored equipment within a first preset time period; determining a cross-correlation coefficient based on the monitoring data and the operating status data, wherein the cross-correlation coefficient is configured to characterize the correlation between the fluctuation trends of the monitoring data and the operating status data; identifying abnormal fluctuation patterns corresponding to the monitoring data in response to the cross-correlation coefficient being less than a coefficient threshold; determining a reliability index of the transmitter based on data characteristics of the monitoring data, wherein the data characteristics include at least one of packet loss rate, signal-to-noise ratio, and extreme value data; generating filtering parameters based on the signal-to-noise ratio and spectral distribution characteristics of the monitoring data in response to the reliability index being less than a first reliability threshold, and controlling the transmitter to adjust the filtering window and / or cutoff frequency based on the filtering parameters; and controlling the monitored equipment to shut down, triggering an alarm, and generating a manual maintenance work order in response to the reliability index being less than or equal to a second reliability threshold, wherein the first reliability threshold is greater than the second reliability threshold.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes any of the above-described methods for online maintenance of transmitters in the Industrial Internet of Things (IIoT).

[0008] One or more embodiments of the present invention have at least the following technical effects: (1) The industrial Internet of Things management platform distinguishes between real operating condition fluctuations and transmitter abnormalities by cross-correlation coefficients, fundamentally avoiding false alarms. According to the reliability index, hierarchical intelligent handling is performed: when the reliability index is less than the first reliability threshold, it indicates that the transmitter may have a minor repairable abnormality, and the online parameters are adaptively adjusted for the minor repairable abnormality; when the reliability index is less than the second reliability threshold, it indicates that the transmitter may have an irreparable serious fault, and the equipment is shut down safely and a manual maintenance work order is generated for the irreparable serious fault, maximizing the continuity of operation while ensuring safety. (2) By using onboard temperature data and power supply voltage data as inputs to the prediction model, the prediction model can directly perceive the working environment and status of the transmitter itself. This helps the prediction model distinguish between signal distortion caused by internal temperature drift or power fluctuations and faults caused by substantial damage to the sensing element, further refining the diagnostic granularity of the root cause of the fault. Through the internal state characteristics of the transmitter, the prediction model can better learn and adapt to the normal working mode of the transmitter under different conditions (such as high temperature and unstable voltage), so that it can still make a more stable and accurate reliability index assessment when facing complex conditions. Attached Figure Description

[0009] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0010] Figure 1 This is a platform schematic diagram of an industrial Internet of Things (IoT) management platform shown in some embodiments of this specification; Figure 2 This is an exemplary flowchart of an industrial IoT method for online maintenance of transmitters, as shown in some embodiments of this specification; Figure 3 These are exemplary schematic diagrams of prediction models shown according to some embodiments of this specification; Figure 4 This is an exemplary schematic diagram illustrating the adjustment of the trigger frequency according to some embodiments of this specification. Detailed Implementation

[0011] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0012] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0013] Figure 1 This is an exemplary structural diagram of an industrial IoT system for online maintenance of transmitters, as shown in some embodiments of this specification.

[0014] In some embodiments, such as Figure 1 As shown, the industrial IoT system (hereinafter referred to as the system) 100 for online maintenance of transmitters may include an industrial IoT management platform 130.

[0015] The Industrial Internet of Things (IIoT) management platform 130 refers to a digital monitoring and management platform for supervising transmitters and monitored equipment. For details regarding transmitters and monitored equipment, please refer to [link to relevant documentation]. Figure 2 And its contents.

[0016] In some embodiments, the industrial IoT management platform 130 may be configured in a processor and / or server. The processor and / or server may process data and / or information acquired from other platforms. Based on this data, information, and / or processing results, the processor and / or server may execute program instructions to perform one or more functions described in this application.

[0017] In some embodiments, the industrial IoT management platform 130 includes interconnected sub-platforms and a data center. The sub-platforms can process data and / or information acquired from the data center.

[0018] In some embodiments, the sub-platform includes at least one of an anomaly assessment sub-platform, a data supervision sub-platform, a maintenance sub-platform, and an execution sub-platform.

[0019] The anomaly assessment sub-platform refers to a management platform for assessing and preventing abnormal fluctuations in monitoring data and operating condition data.

[0020] The data monitoring sub-platform refers to a platform used for monitoring, collecting, and analyzing the monitoring data from transmitters and the operating status data of the monitored equipment. For a description of the monitoring data from transmitters and the operating status data of the monitored equipment, please refer to [link to relevant documentation]. Figure 2 And its contents.

[0021] The maintenance sub-platform refers to a platform that generates maintenance strategies by identifying abnormal fluctuations and evaluating reliability metrics. For an explanation of reliability metrics, please refer to [link to relevant documentation]. Figure 2 And its contents.

[0022] The execution sub-platform refers to a platform used to coordinate, schedule, and execute maintenance strategies after identifying abnormal fluctuations.

[0023] In some embodiments, a data center includes a database, a data processing model library, and computing units.

[0024] Databases are used to collect, store, and manage large amounts of data related to transmitter regulation. Examples include MySQL, PostgreSQL, InfluxDB, and Prometheus.

[0025] A data processing model library refers to a collection of data processing models used for data processing. In some embodiments, data processing models may include predictive models, etc. For a description of predictive models, see [link to documentation]. Figure 3 And its contents.

[0026] A computing unit is a functional module used to perform arithmetic, logical, and other instruction operations. Computing units may include, but are not limited to, central processing units (CPUs).

[0027] In some embodiments, the system further includes an industrial IoT user platform 110, an industrial IoT service platform 120, an industrial IoT sensor network platform 140, and an industrial IoT sensing and control platform 150.

[0028] An Industrial Internet of Things (IIoT) user platform refers to an interactive platform for management personnel. In some embodiments, an IIoT user platform may include at least one human interaction device, such as a mobile phone or computer. Management personnel may be staff members of companies involved in transmitter manufacturing or maintenance management.

[0029] An Industrial Internet of Things (IIoT) service platform refers to a platform that provides communication services. In some embodiments, the IIoT service platform can be configured as a server that can interact with IIoT user platforms and IIoT management platforms (such as data centers).

[0030] An Industrial Internet of Things (IIoT) sensor network platform refers to a platform for sensing and communicating IIoT-based information. It facilitates bidirectional data exchange and transmission between the IIoT management platform (such as a data center) and the IIoT sensing and control platform. For example, an IIoT sensor network platform may include communication equipment, servers, and various gateway devices. The IIoT sensing and control information may include transmitter monitoring data, operating status data of monitored equipment, and the correlation of fluctuation trends. For an explanation of the correlation of fluctuation trends, please refer to [link to relevant documentation]. Figure 2 And its contents.

[0031] An industrial IoT sensing and control platform refers to a platform for controlling industrial IoT sensing and control information. In some embodiments, the industrial IoT sensing and control platform may include sensing and controlling various monitored devices and transmitters. The industrial IoT sensing and control platform may include the devices it senses and controls, as well as processing devices such as processors or servers.

[0032] In some embodiments of this specification, the industrial IoT system 100 for online maintenance of transmitters can form an information operation closed loop between various functional platforms and operate in a coordinated and regular manner under the unified management of the industrial IoT management platform, thereby realizing the informatization and intelligentization of online maintenance of transmitters.

[0033] Figure 2 This is a flowchart illustrating a method for online maintenance of a transmitter according to some embodiments of this specification. Figure 2 As shown, process 200 includes steps 210 to 260, and process 200 is executed by the industrial IoT management platform.

[0034] Step 210: Obtain the monitoring data of the transmitter and the operating status data of the monitored equipment within the first preset time period. The first preset time period can be preset based on manual experience.

[0035] A transmitter is a device that converts the raw, weak, non-standard electrical signals (such as mV-level thermoelectric potentials) sensed by sensing elements (such as piezoresistive diaphragms, thermocouples, etc.) into standard signals (such as 4-20mA analog current signals and other digital bus signals) and transmits them remotely.

[0036] In some embodiments, the transmitter includes a pressure transmitter, a temperature transmitter, etc.

[0037] Monitoring data refers to the data measured, collected, and uploaded by the transmitter from the monitored equipment.

[0038] In some embodiments, the monitoring data includes a data timestamp, a frame sequence number, and data from the monitored device. The data timestamp is high-precision time information used to accurately identify the actual moment the data point was collected, such as the number of seconds since 1970-01-01 00:00:00 UTC using a Unix timestamp. The frame sequence number is an identifier used to uniquely identify and sort data frames, such as a 16-bit unsigned integer sequence [0, 1, 2, ..., 65535, 0, 1, ...] that starts at 0, reaches a maximum value (e.g., 65535), and then increments cyclically. Data from the monitored device refers to the data measured and collected by the transmitter, such as the temperature, pressure, and on / off status of the monitored device (e.g., a pipeline).

[0039] In some embodiments, when the transmitter is operating normally, the data of the monitored device should be consistent with the operating condition data.

[0040] In some embodiments, the industrial IoT management platform can parse monitoring data from data frames sent by the transmitter.

[0041] The monitored equipment is the industrial physical equipment that the transmitter directly measures, such as pipes, motors, valves, etc.

[0042] Operating status data are control or status signals that describe the operating status of the monitored equipment.

[0043] In some embodiments, the operating condition data includes discrete switching data and continuous control data.

[0044] Discrete switch data are state quantities that characterize two or more defined states of a device (such as start / stop, on / off). For example, discrete switch data can be 0 and 1, where 0 represents the off state and 1 represents the off state.

[0045] Continuous control data is used to regulate or control continuously changing control quantities in equipment operation, such as pipeline temperature and pipeline pressure.

[0046] In some embodiments, the industrial IoT management platform can obtain operating status data from the monitored device (such as the control system of the monitored device).

[0047] Step 220: Determine the cross-correlation coefficient based on the monitoring data and operating condition data.

[0048] Cross-correlation coefficient is a parameter used to characterize the degree of correlation between the fluctuation trends of monitoring data and operating condition data.

[0049] In some embodiments, the industrial IoT management platform can use the monitored equipment data in the monitoring data within a first preset time period to form a measured data sequence, use the operating condition data within the first preset time period as an operating condition data sequence, and use the Pearson correlation coefficient between the measured data sequence and the operating condition data sequence as a cross-correlation coefficient.

[0050] Step 230: In response to the cross-correlation coefficient being less than the coefficient threshold, identify the abnormal fluctuation pattern corresponding to the monitoring data.

[0051] In some embodiments, the coefficient threshold is a statistical decision threshold used to determine whether the fluctuation trend of the transmitter monitoring data is significantly correlated with the fluctuation trend of the operating condition data of the monitored equipment.

[0052] There are two known scenarios: the cross-correlation coefficient is less than the coefficient threshold, and the cross-correlation coefficient is not less than the coefficient threshold. In some embodiments, when the cross-correlation coefficient is less than the coefficient threshold, the industrial IoT management platform can determine that the fluctuation trend of the transmitter monitoring data is inconsistent with the fluctuation trend of the monitored equipment's operating status data. In this case, it is necessary to identify the abnormal fluctuation pattern corresponding to the monitoring data. When the cross-correlation coefficient is not less than the coefficient threshold, the industrial IoT management platform can determine that the fluctuation trend of the transmitter monitoring data is consistent with the fluctuation trend of the monitored equipment's operating status data.

[0053] In some embodiments, the coefficient threshold is preset based on human experience, for example, preset to 0.7 or 0.8.

[0054] Abnormal fluctuation patterns can be characterized by characteristic waveforms that deviate from normal change patterns in monitoring data.

[0055] In some embodiments, abnormal fluctuation patterns include step anomalies, linear drift anomalies, etc. A step anomaly is a waveform in which the monitored data shows a vertical increase or decrease within a short period of time (e.g., within 1 second). A linear drift anomaly is a waveform in which the monitored data shows a linear increase or decrease over time.

[0056] In some embodiments, the industrial IoT management platform can plot a waveform with time on the horizontal axis and the monitored device data in the current monitoring data on the vertical axis. The plotted waveform is compared with waveforms of preset abnormal fluctuation patterns. The waveform of the preset abnormal fluctuation pattern that is most similar to the plotted waveform, corresponding to the preset abnormal fluctuation pattern, is taken as the abnormal fluctuation pattern. The preset abnormal fluctuation patterns and their waveforms can be preset based on experience.

[0057] Step 240: Determine the reliability index of the transmitter based on the data characteristics of the monitoring data.

[0058] Data characteristics are quantitative parameters that describe the integrity, authenticity, and quality status of monitoring data during transmission and generation.

[0059] In some embodiments, data features include at least one of the following: packet loss rate, signal-to-noise ratio, and extreme value data of the monitoring data.

[0060] Packet loss rate is used to characterize the transmission reliability of a communication link. In some embodiments, when receiving monitoring data, the industrial IoT management platform can use the ratio of the number of data frames that were actually not received to the number of frames that were expected to be received based on the frame sequence number as the packet loss rate, where the number of data frames that were actually not received is the difference between the number of frames that were expected to be received and the number of data frames that were actually received.

[0061] Signal-to-noise ratio (SNR) is a parameter used to quantify the quality and purity of a signal. In some embodiments, an industrial IoT management platform can obtain the SNR by performing spectral analysis on the measured data sequence.

[0062] Extreme value data refers to the maximum and / or minimum values ​​in the measured data sequence that exceed the normal range. In some embodiments, the industrial IoT management platform can extract the maximum and minimum values ​​from the measured data sequence and compare them with the preset safety upper limit and preset safety lower limit of the transmitter range. The maximum value exceeding the preset safety upper limit of the transmitter range, or the minimum value less than the preset safety lower limit of the transmitter range, is taken as extreme value data.

[0063] A reliability index is a quantitative value used to characterize the reliability of the current output data of a transmitter. That is, the more reliable the data output by the transmitter, the higher the reliability index.

[0064] In some embodiments, the more reliable the data output by the transmitter, the more reliable the transmitter's operating status and the less likely it is to have anomalies; conversely, the less reliable the data output by the transmitter, the less reliable the transmitter's operating status and the more likely it is to have anomalies.

[0065] In some embodiments, the industrial IoT management platform can evaluate data characteristics and use the evaluation results as a reliability indicator. For example, the industrial IoT management platform can normalize packet loss rate, signal-to-noise ratio, and extreme value data respectively, weight the normalization results, and use the weighted result as a reliability indicator. The weights are all negative values, and the weights can be preset based on human experience.

[0066] For more information on reliability metrics, please see [link to relevant documentation]. Figure 4 And related content.

[0067] There are two known cases: reliability index is less than the first reliability threshold and reliability index is not less than the first reliability threshold.

[0068] Step 250: In response to the reliability index being less than the first reliability threshold, filter parameters are generated based on the signal-to-noise ratio and spectral distribution characteristics of the monitoring data, and the transmitter is controlled to adjust the filter window and / or cutoff frequency based on the filter parameters.

[0069] The first reliability threshold can be preset based on human experience.

[0070] Spectral distribution characteristics are data used to characterize the distribution of frequency components in monitoring data. For example, spectral distribution characteristics may include the distribution of dominant frequency, harmonic components, and noise bands.

[0071] In some embodiments, the industrial IoT management platform can perform Fourier transform on the monitoring data to obtain spectral distribution characteristics.

[0072] Filter parameters are a set of parameters used to configure the transmitter's local filters.

[0073] In some embodiments, filtering parameters include filter window parameters and / or cutoff frequency. Filter window parameters are a set of parameters defining the data range and its movement rules covered by the filter in each operation (e.g., sampling by the filter function) when processing a data sequence (e.g., the measured data sequence). Filter window parameters may include window length, sliding step size, weighting coefficients, etc. Window length is the number of sampling points used to move the filter function (e.g., averaging filter) for sampling. Sliding step size refers to the step size of the window movement, typically 1 (i.e., point-by-point sliding). Weighting coefficients refer to the weight of each sampling point when moving the filter function (e.g., averaging filter) for sampling. Cutoff frequency refers to the cutoff frequency of the filter, including the cutoff frequency of a low-pass filter used to filter out high-frequency noise and the cutoff frequency of a high-pass filter used to filter out DC drift.

[0074] In some embodiments, the industrial IoT management platform can extract frequency bands with concentrated noise from the spectrum using a residual analysis algorithm based on the spectral distribution characteristics. Then, based on the frequency bands with concentrated noise and the signal-to-noise ratio (SNR), the platform determines the filtering parameters by querying a first preset table. This first preset table records the correspondence between the frequency bands with concentrated noise, the SNR, and the filtering parameters. For example, the higher the frequency band with concentrated noise, the higher the cutoff frequency of the low-pass filter, the higher the SNR, and the smaller the window length in the filtering window parameters. The first preset table can be preset based on manual experience.

[0075] In some embodiments, the industrial IoT management platform can use the filter window parameter in the filter parameters as the filter window of the transmitter, and the cutoff frequency in the filter parameters as the cutoff frequency of the transmitter, and control the transmitter to adjust the filter window and / or cutoff frequency based on the filter parameters.

[0076] There are two known cases: reliability index is less than or equal to the second reliability threshold, and reliability index is greater than the second reliability threshold.

[0077] Step 260: In response to the reliability index being less than or equal to the second reliability threshold, control the monitored equipment to shut down, trigger an alarm, and generate a manual maintenance work order.

[0078] The second reliability threshold is a parameter that is less than the first reliability threshold, and the second reliability threshold can be preset based on human experience.

[0079] Manual maintenance work orders are data used to guide workers in performing offline manual maintenance.

[0080] In some embodiments, a manual maintenance work order may include the monitored equipment number, transmitter number, and time of occurrence.

[0081] In some embodiments, when the reliability index is less than or equal to the second reliability threshold, the industrial IoT management platform can trigger an audible and visual alarm in the control room or on-site operation station, and automatically create and fill manual maintenance work orders. According to the preset maintenance personnel schedule, the platform can automatically dispatch manual maintenance work orders to the corresponding maintenance teams or personnel.

[0082] In some embodiments of this specification, the industrial IoT management platform distinguishes between fluctuations in actual operating conditions and transmitter malfunctions by using cross-correlation coefficients, fundamentally avoiding false alarms. Based on reliability indicators, tiered intelligent handling is implemented: when the reliability indicator is less than a first reliability threshold, it indicates that the transmitter may have a minor, repairable malfunction, and online parameter adaptive adjustments are made for such minor malfunctions; when the reliability indicator is less than a second reliability threshold, it indicates that the transmitter may have an irreparable serious fault, and for irreparable serious faults, the equipment is shut down safely and a manual maintenance work order is generated, maximizing operational continuity while ensuring safety.

[0083] In some embodiments, the industrial IoT management platform can further control the transmitter to change the sampling frequency to a non-multiplication value and perform resampling to obtain resampled data; in response to the frequency aliasing or morphological abrupt change in the waveform of the resampled data, it determines that the abnormal fluctuation of the monitoring data is a false signal caused by environmental noise; otherwise, it determines that the abnormal fluctuation of the monitoring data is a true signal caused by the fluctuation of the actual operating conditions.

[0084] Sampling frequency refers to the frequency at which the transmitter collects data from the monitored equipment.

[0085] Non-harmonic values ​​refer to new sampling frequencies that are not integer multiples of the industrial power supply frequency or its harmonic frequencies. The harmonic frequencies of the industrial power supply frequency are integer multiples of the fundamental frequency (50Hz or 60Hz) of the industrial power supply frequency. The industrial power supply frequency or its harmonic frequencies are preset based on human experience.

[0086] Non-harmonic values ​​can be selected based on human experience. For example, in an industrial power supply environment with a frequency of 50Hz, frequencies such as 63Hz or 47Hz can be selected.

[0087] In some embodiments, the industrial IoT management platform controls the transmitter to change the sampling frequency to a non-frequency doubling value, so that the transmitter re-collects data from the monitored device according to the changed sampling frequency, i.e., resampling.

[0088] In some embodiments, resampled data refers to monitoring data obtained after resampling.

[0089] In some embodiments, the waveform of the resampled data refers to the waveform plotted with the monitored device data in the resampled data as the vertical axis and the sampling time as the horizontal axis.

[0090] Frequency aliasing refers to the phenomenon where a high-frequency signal is incorrectly mapped to a low-frequency signal.

[0091] In some embodiments, the industrial IoT management platform can perform spectral analysis on the waveform of the resampled data. If frequency components that are not present in the original signal appear in a frequency band below the Nyquist frequency, it is determined that frequency aliasing has occurred.

[0092] A morphological abrupt change refers to an unexpected and drastic change in the basic shape, period, or amplitude of a waveform.

[0093] In some embodiments, the industrial IoT management platform can compare the waveforms of resampled data and monitoring data. If there are significant or discontinuous changes in the dominant frequency, amplitude envelope, or periodic characteristics, it is determined to be a morphological abrupt change. For example, if the difference between the dominant frequency in the waveform of the resampled data and the dominant frequency in the waveform of the monitoring data is greater than a preset abrupt change value, a morphological abrupt change is determined to exist. The industrial IoT management platform can obtain the dominant frequency in the waveform of the resampled data (or the dominant frequency in the waveform of the monitoring data) by performing a Fourier transform on the waveform of the resampled data (or the waveform of the monitoring data).

[0094] Abnormal fluctuations in monitoring data refer to data that do not match the monitoring data normally collected by the transmitter.

[0095] In some embodiments, the presence of environmental noise or malfunctions in the monitored equipment can prevent the transmitter from collecting monitoring data normally. When the monitored equipment malfunctions (such as equipment failure), corresponding abnormal fluctuations will occur in the monitoring data. Similarly, the presence of environmental noise (such as electromagnetic interference) will also cause corresponding abnormal fluctuations in the monitoring data.

[0096] In some embodiments, when the abnormal fluctuations in the monitoring data are caused by environmental noise in the environment, the abnormal fluctuations in the monitoring data are false signals caused by environmental noise.

[0097] In some embodiments, when the abnormal fluctuations in the monitoring data are caused by abnormal conditions in the monitored equipment, the abnormal fluctuations in the monitoring data are the true signals caused by the actual operating condition fluctuations.

[0098] There are two scenarios: waveforms with resampled data exhibiting frequency aliasing or abrupt morphological changes, and waveforms with resampled data not exhibiting frequency aliasing or abrupt morphological changes.

[0099] In some embodiments, if the waveform of the resampled data exhibits frequency aliasing or abrupt changes in shape, the industrial IoT management platform will determine that the abnormal fluctuations in the monitoring data are false signals caused by environmental noise; if the waveform of the resampled data does not exhibit frequency aliasing or abrupt changes in shape, the industrial IoT management platform will determine that the abnormal fluctuations in the monitoring data are true signals caused by actual operating condition fluctuations.

[0100] In some embodiments of the specification, the industrial IoT management platform controls the transmitter to change the sampling frequency and resample, and analyzes the resampled waveform to determine the cause of abnormal fluctuations in the monitoring data. This can accurately distinguish between environmental interference and the actual faults of the monitored equipment, effectively avoid erroneous maintenance caused by external interference, and significantly enhance the pertinence and effectiveness of maintenance.

[0101] Figure 3These are exemplary schematic diagrams of prediction models shown according to some embodiments of this specification.

[0102] In some embodiments, the industrial IoT management platform is further configured to: determine a common-mode drift component 320 based on the median 310 of the changes in monitoring data from multiple transmitters; and determine a reliability index 360 using a prediction model 350, which is a machine learning model, based on the data characteristics 330 of the monitoring data, the operating status data 340 of the monitored equipment, and the common-mode drift component 320. For further details regarding transmitters, monitoring data, monitored equipment, data characteristics, and operating status data, please refer to [link to relevant documentation]. Figure 2 Related descriptions.

[0103] The change in monitoring data refers to the degree of change in the monitoring data. For example, it's the percentage change of monitoring data relative to a baseline value within a certain time window. See [link to explanation of time windows] for more information. Figure 2 The relevant description is as follows. In some embodiments, the benchmark value can be determined based on empirical presets. For example, the industrial IoT management platform can determine the benchmark value based on the average value of the transmitter's monitoring data under normal operating conditions, based on historical data. In some embodiments, the benchmark value can be determined based on the type of transmitter. For example, if the monitoring data of a temperature transmitter is a temperature value, the benchmark value can be 25°C.

[0104] The monitoring data from multiple transmitters on the same monitoring device may be affected by environmental interference (such as overall temperature changes, ground potential drift, and increased electromagnetic noise), resulting in reading drift in the same direction.

[0105] Reading drift refers to the change in the reading of a transmitter that is approximately a magnitude shift.

[0106] Common-mode drift component refers to the quantized value used to characterize the reading drift of multiple transmitters.

[0107] In some embodiments, the industrial IoT management platform can acquire monitoring data from multiple transmitters on the same monitoring device within the current time window, calculate the change in monitoring data for each transmitter, and take the median of the changes in monitoring data from multiple transmitters as the common-mode drift component.

[0108] In some embodiments, the industrial IoT management platform can determine the reliability index 360 based on the data characteristics 330 of the monitoring data, the operating status data 340 of the monitored equipment, and the common mode drift component 320, through a prediction model 350.

[0109] In some embodiments, the prediction model is a machine learning model. For example, the prediction model can be any one or a combination of deep neural networks (DNNs) or other custom model structures.

[0110] like Figure 3 The exemplary schematic diagram of the prediction model shown indicates that the inputs of the prediction model 350 may include data features 330 of the monitoring data, operating condition data 340 of the monitored equipment, and common mode drift component 320, and the output may include reliability index 360.

[0111] In some embodiments, the inputs to the prediction model also include onboard temperature data 370 and supply voltage data 380 from multiple transmitters.

[0112] Onboard temperature data refers to the temperature data of the circuit board or key components (such as the sensor head and the analog-to-digital converter (ADC) chip) inside the transmitter.

[0113] In some embodiments, the industrial IoT management platform can acquire onboard temperature data through the temperature sensor built into the temperature transmitter and upload the onboard temperature data along with the monitoring data to the data processing model library.

[0114] The power supply voltage data refers to the voltage value that provides the operating power to the internal circuitry of the transmitter.

[0115] In some embodiments, the industrial IoT management platform can acquire power supply voltage data through the voltage monitoring circuit built into the voltage transmitter, and upload the power supply voltage data along with the monitoring data to the data processing model library.

[0116] In some embodiments of this specification, by using onboard temperature data and power supply voltage data as inputs to the predictive model, the predictive model can directly perceive the transmitter's own operating environment and state. This helps the predictive model distinguish between signal distortion caused by internal temperature drift or power fluctuations and faults caused by substantial damage to the sensing element, further refining the diagnostic granularity of the root cause of the fault. Through the internal state characteristics of the transmitter, the predictive model can better learn and adapt to the normal operating mode of the transmitter under different conditions (such as high temperature and unstable voltage), thereby making a more stable and accurate reliability index assessment when facing complex conditions.

[0117] In some embodiments, the prediction model can be trained based on a large number of training samples with training labels. The industrial IoT management platform can input multiple training samples with training labels into the initial prediction model, construct a loss function using the training labels and the results of the initial prediction model, and iteratively update the parameters of the initial prediction model based on the loss function using methods such as gradient descent. When the loss function meets preset training conditions, a trained prediction model is obtained. These preset training conditions may include loss function convergence, the number of iterations reaching a threshold, etc.

[0118] Training samples and training labels can be obtained based on historical data. Training samples may include data characteristics of historical monitoring data, historical operating status data of monitored equipment, and historical common-mode drift components. Training labels may include reliability metrics corresponding to the training samples.

[0119] In some embodiments, the industrial IoT management platform can acquire historical monitoring data, including pre-transmission and post-transmission data, calculate the similarity between the pre-transmission and post-transmission monitoring data, and use the similarity as a training label. Transmission refers to the process of transferring historical monitoring data from the transmitter to the industrial IoT management platform. That is, the pre-transmission monitoring data is the monitoring data in the transmitter, and the post-transmission monitoring data is the monitoring data received by the industrial IoT management platform.

[0120] In some embodiments of this specification, by calculating the common-mode drift component, the reading offsets of multiple transmitters caused by overall environmental changes (such as day-night temperature variations) are quantified and isolated, preventing such environmental influences from being misjudged as hardware failures of a single transmitter, thereby significantly reducing the batch false alarm rate. By utilizing machine learning models to comprehensively consider multi-dimensional information such as data characteristics, operating condition data, and common-mode drift components, reliability index prediction is performed. Compared with traditional weighted calculations, this method can better capture complex nonlinear relationships, making reliability indexes closer to actual operating conditions and improving the accuracy of fault classification and reliability assessment.

[0121] Figure 4 This is an exemplary schematic diagram of a trigger frequency adjustment method according to some embodiments of this specification.

[0122] It is known that there are two scenarios: the reliability index is greater than the second preset threshold and the reliability index is not greater than the second preset threshold. In some embodiments, in response to the reliability index 360 being greater than the second preset threshold 410, the industrial IoT management platform determines the transmitter's delay index 430 based on the time difference sequence 420 between the platform's reception time and the actual acquisition time of the monitored data within the second preset time period. For more information on the reliability index 360, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0123] The second preset time period is a time window used for statistical analysis of network latency.

[0124] In some embodiments, the second preset time period includes multiple first preset time periods. The second preset time period is preset based on human experience.

[0125] The platform receiving time refers to the moment when the industrial IoT management platform receives the monitoring data.

[0126] In some embodiments, the industrial IoT management platform can automatically generate the platform reception time when it receives monitoring data.

[0127] The actual acquisition time refers to the moment when the transmitter acquires and monitors the data.

[0128] In some embodiments, the industrial IoT management platform can determine the timestamp in the monitoring data as the actual collection time.

[0129] The time difference sequence refers to the sequence of time differences between the transmission of monitoring data from the transmitter and the receipt of data by the industrial IoT management platform within a second preset time period.

[0130] In some embodiments, the industrial IoT management platform can use the difference between the actual acquisition time and the platform reception time of each frame of monitoring data as the time difference of that frame of monitoring data, and sort the time differences of multiple frames of monitoring data received within a second preset time period according to the platform reception time of the monitoring data to obtain a time difference sequence.

[0131] The latency index is a parameter that characterizes the network latency of transmitter data upload and its jitter quantification.

[0132] In some embodiments, the industrial IoT management platform can use the dimensionless value obtained by normalizing the variance or standard deviation of the time difference sequence as a delay indicator.

[0133] In some embodiments of the specification, by reflecting the time difference sequence that reflects the deviation between the actual acquisition time and the platform reception time, the deviation of the transmitter's internal clock relative to the system time base (synchronization index) can be directly and accurately measured, so as to facilitate subsequent clock calibration.

[0134] It is known that there are two scenarios: the latency index is greater than the threshold and the latency index is not greater than the threshold. In some embodiments, in response to the latency index being greater than the threshold, the industrial IoT management platform generates a frequency reduction parameter based on the latency index; controls the transmitter to adjust the trigger frequency of the transmission timer based on the frequency reduction parameter and extends the retransmission waiting time.

[0135] The indicator threshold is a threshold used to determine whether network latency has increased abnormally.

[0136] In some embodiments, the indicator threshold can be preset based on human experience.

[0137] Frequency reduction parameters are parameters used to reduce the communication frequency of a transmitter.

[0138] In some embodiments, the frequency reduction parameters include the trigger frequency and the retransmission waiting time. The trigger frequency is the rate at which periodic data reporting actions (such as uploading monitoring data to an industrial IoT management platform) are triggered.

[0139] In some embodiments, the industrial IoT management platform can first calculate the difference between the latency indicator and the indicator threshold, and then determine the frequency reduction parameter based on the difference according to a preset mapping relationship (such as a preset linear function). The preset mapping relationship can be pre-set based on human experience. For example, the larger the difference, the lower the trigger frequency and the longer the retransmission waiting time.

[0140] In some embodiments, the industrial IoT management platform can adjust the trigger frequency of the transmission timer to the trigger frequency in the down-frequency parameter and extend the retransmission waiting time to the retransmission waiting time in the down-frequency parameter.

[0141] In some embodiments of the specification, the industrial IoT management platform reduces the number of network data packets per unit time by adjusting the trigger frequency of the transmission timer, thereby reducing the probability of channel collisions and helping to alleviate network congestion overall. By extending the retransmission waiting time, the industrial IoT management platform avoids network collisions exacerbated by immediate retransmissions, increasing the likelihood of transmitters remaining online in harsh network environments.

[0142] In some embodiments, in response to the latency index being less than or equal to the index threshold, the industrial IoT management platform determines the transmitter's synchronization index based on the occurrence time of the operating condition event of the monitored equipment and the acquisition time of the operating condition event by the transmitter within a third preset time period; in response to the synchronization index meeting preset conditions, clock calibration parameters are generated based on the synchronization index; the transmitter is controlled to set the initial sampling time based on the clock calibration parameters and adjust the frequency division coefficient of the sampling timer.

[0143] The third preset time period is a time window used to calculate clock synchronization deviation.

[0144] In some embodiments, the third preset time period includes multiple second preset time periods. The third preset time period can be preset based on human experience.

[0145] Operating events refer to events in which the operating status of monitored equipment undergoes a clear and identifiable change during operation, such as the issuance of valve opening / closing commands, motor starting / stopping, pump starting / stopping, and equipment switching operating modes. For example, valve opening.

[0146] The occurrence time of a condition event refers to the moment when a condition event occurs in the monitored equipment. For example, the moment when a valve opening command is issued.

[0147] In some embodiments, the industrial IoT management platform can obtain operating events and the times when the operating events occur from the control system (such as a system that controls the opening and closing of valves).

[0148] The acquisition time of an operating condition event refers to the moment when the operating condition event detected by the transmitter occurs.

[0149] In some embodiments, the industrial IoT management platform can identify the monitoring data (such as the data frame corresponding to the slope change point in the waveform plot drawn from the measured data sequence) where the characteristic changes of the operating condition event are located from the monitored data (such as the data frame corresponding to the slope change point in the waveform plot drawn from the measured data sequence), and use the timestamp of the monitoring data where the characteristic changes of the operating condition event are located as the collection time of the operating condition event.

[0150] The industrial IoT management platform can determine the characteristic changes of operating conditions by querying a second preset table based on the operating condition events. The second preset table records the correspondence between operating conditions and their characteristic changes, and can be pre-set based on human experience. For example, when the transmitter monitors a valve on a pipeline, and the operating condition event is a valve closing command, the transmitter can monitor the air velocity at the valve as the measured data sequence, and a sudden drop in air velocity (e.g., a decrease in air velocity exceeding 50%) in the measured data sequence is taken as a characteristic change of the operating condition event.

[0151] Synchronization metrics are parameters used to evaluate the accuracy of time synchronization of transmitters.

[0152] In some embodiments, the industrial IoT management platform can calculate the time difference between the transmitter's acquisition time of an operating condition event and the occurrence time of the operating condition event, and use this time difference as the time difference of the operating condition event. The average of the time differences of multiple operating condition events within a third preset time period is used as a synchronization index.

[0153] For more information on synchronicity metrics, please see the relevant descriptions below.

[0154] It is known that there are two scenarios: the synchronization index meets the preset conditions and the synchronization index does not meet the preset conditions. In response to the synchronization index meeting the preset conditions, the processor can generate clock calibration parameters based on the synchronization index. In some embodiments, the preset conditions are used to determine whether the synchronization index is abnormal. The preset conditions can be pre-set based on human experience.

[0155] Clock calibration parameters are used to correct for internal clock deviations in the transmitter.

[0156] In some embodiments, clock calibration parameters include a time offset and a frequency correction factor. The time offset is a parameter used to compensate for clock deviations, and the frequency correction factor is a parameter used to compensate for the long-term drift rate of the clock.

[0157] In some embodiments, the industrial IoT management platform can use the negative of the synchronization index as the time offset. The occurrence times of multiple operating conditions corresponding to multiple operating conditions within a third preset time period are used as independent variables, and the time differences between these multiple operating conditions within the third preset time period are used as dependent variables. A linear fit is then performed, and the slope of the fitted line is used as the frequency correction coefficient.

[0158] The initial sampling time refers to the time interval from the current sampling interruption to the next sampling interruption.

[0159] In some embodiments, the industrial IoT management platform can determine the initial sampling time based on the time offset using a clock synchronization compensation algorithm, and set the determined initial sampling time as the initial sampling time of the transmitter.

[0160] The frequency division factor is used to reduce the counting frequency of the timer.

[0161] In some embodiments, the industrial IoT management platform can determine the frequency division coefficient based on the clock calibration parameters using an adaptive PID frequency compensation algorithm, and set the determined frequency division coefficient as the frequency division coefficient of the transmitter's sampling timer.

[0162] In some embodiments of the specification, the industrial IoT management platform can directly and accurately measure and determine synchronization indicators by comparing the occurrence time of the operating condition event of the monitored equipment with the acquisition time of the corresponding event by the transmitter, providing a basis for online clock calibration. By setting the initial sampling time and adjusting the frequency division coefficient of the sampling timer, the phase and frequency of the clock can be calibrated simultaneously, ensuring that the transmitter data acquisition remains synchronized with the system time base in a long-term and stable manner, laying the foundation for multi-transmitter data fusion and accurate event sequencing.

[0163] In some embodiments, the industrial IoT management platform determines the target monitoring transmitter based on the transmitter and the monitored device; acquires the monitoring data sequence of the transmitter and the monitoring data sequence of the target monitoring transmitter respectively, and determines the cross-correlation function; in response to the cross-correlation peak value being greater than the peak value threshold, determines the time shift corresponding to the cross-correlation peak value; determines the synchronization index of the transmitter based on the time shift value; in response to the synchronization index meeting preset conditions, generates clock calibration parameters based on the synchronization index; and controls the transmitter to set the initial sampling time based on the clock calibration parameters and adjusts the frequency division coefficient of the sampling timer.

[0164] When a transmitter monitors the operating events of the monitored equipment, the target monitoring transmitter refers to other transmitters that are associated with the operating events of the monitored equipment monitored by the transmitter.

[0165] In some embodiments, the industrial IoT management platform can use a transmitter that monitors the same monitored device as the transmitter as the target monitoring transmitter.

[0166] In some embodiments, the industrial IoT management platform may also use transmitters that are installed near the transmitter and associated with the monitored device as target monitoring transmitters. For example, when the monitored device is a valve on a pipeline, the target monitoring transmitter may be a transmitter installed on the downstream pipeline or on the valve monitored by the transmitter.

[0167] The transmitter's monitoring data sequence is a data sequence composed of monitoring data acquired by the transmitter within a third preset time period.

[0168] The monitoring data sequence of the target monitoring transmitter is a data sequence composed of monitoring data acquired by the target monitoring transmitter within a third preset time period.

[0169] For more information on monitoring data, please see [link / documentation]. Figure 2 And related content.

[0170] A cross-correlation function is a function that describes the similarity between two signals (such as a transmitter's monitoring data sequence and a target monitoring transmitter's monitoring data sequence). For example, a cross-correlation function describes the similarity between a transmitter's monitoring data sequence and a target monitoring transmitter's monitoring data sequence at different time shifts. The time shift refers to the relative time delay between the two signals.

[0171] In some embodiments, the industrial IoT management platform can perform cross-correlation calculations on the monitoring data sequences of the transmitter and the monitoring data sequences of the target monitoring transmitter to obtain a cross-correlation function.

[0172] The peak value of cross-correlation refers to the peak value of the cross-correlation function.

[0173] In some embodiments, the industrial IoT management platform may use the maximum value of the cross-correlation function as the cross-correlation peak value.

[0174] The peak threshold is a threshold used to determine whether the peak value of a cross-correlation function has sufficient confidence. For example, when the peak value of the cross-correlation function is greater than the peak threshold, it can be determined that the peak value is not a meaningless peak value caused by noise.

[0175] In some embodiments, the peak threshold can be preset based on human experience.

[0176] In some embodiments, there are two known cases: the cross-correlation peak value of the cross-correlation function is greater than the peak threshold and the cross-correlation peak value is not greater than the peak threshold. When the cross-correlation peak value of the cross-correlation function is greater than the peak threshold, the industrial IoT management platform can use the time shift corresponding to the cross-correlation peak value as the synchronization index of the transmitter.

[0177] In some embodiments, the industrial IoT management platform can use the time shift corresponding to the cross-correlation peak as a synchronization indicator of the transmitter.

[0178] In some embodiments, there are known cases where the synchronization index meets preset conditions and cases where the synchronization index does not meet preset conditions. In response to the synchronization index meeting the preset conditions, clock calibration parameters are generated based on the synchronization index; the transmitter is controlled to set the initial sampling time based on the clock calibration parameters and adjust the division factor of the sampling timer. For more information on preset conditions, clock calibration parameters, setting the initial sampling time, and adjusting the division factor of the sampling timer, please refer to the relevant content above.

[0179] In some embodiments of the specification, the industrial IoT management platform uses the target monitoring transmitter as a reference benchmark and determines the timing relationship between the two transmitters by calculating the cross-correlation function, thereby further determining the synchronization index of the transmitters. In scenarios where the timestamps of control system events are unreliable or difficult to obtain (such as unreliable timestamps due to old equipment), it provides an effective means of synchronization verification. By setting a peak threshold, it effectively filters out mismatches caused by noise interference or unrelated operating condition fluctuations, thereby improving the robustness and accuracy of synchronization judgment.

[0180] In some embodiments, the industrial IoT management platform determines the transmission delay time based on the physical distance between the transmitter and the target monitoring transmitter; and determines the synchronization index of the transmitter based on the time shift and the transmission delay time.

[0181] Physical distance refers to the straight-line distance, or the distance along the medium propagation path, between the location of the monitored device by the transmitter and the location of the monitored device by the target monitoring transmitter. When the transmitter and the target monitoring transmitter are monitoring the same monitored device, the physical distance is 0.

[0182] In some embodiments, the industrial IoT management platform can obtain the location of the monitored device monitored by the transmitter and the location of the monitored device monitored by the target monitoring transmitter from the installation records of the transmitter and the target monitoring transmitter, and then further determine the physical distance based on the location of the monitored device monitored by the transmitter and the location of the monitored device monitored by the target monitoring transmitter.

[0183] In some embodiments, the industrial IoT management platform can obtain physical distance from manually measured results.

[0184] Transmission delay time refers to the time required for a physical disturbance (such as a pressure wave or temperature field change) to propagate from the location of the monitored device monitored by the transmitter to the location of the monitored device monitored by the target monitoring transmitter. When the transmitter and the target monitoring transmitter are monitoring the same monitored device, the transmission delay time is 0.

[0185] In some embodiments, the industrial IoT management platform can query a third preset table based on operating conditions to determine physical disturbances (such as pressure waves or temperature field changes), and then query a fourth preset table based on the physical disturbances to determine the propagation rate corresponding to the physical disturbances. The ratio of physical distance to propagation rate is used as the transmission delay time. The third preset table records operating conditions and their corresponding physical disturbances, while the fourth preset table records physical disturbances and their corresponding propagation rates. Both the third and fourth preset tables can be preset based on human experience.

[0186] In some embodiments, the industrial IoT management platform can use the difference between the time shift and the transmission delay as a synchronization indicator.

[0187] In some embodiments of the specification, when the transmitter and the target monitoring transmitter are not monitoring the same monitored device, by introducing a transmission delay time, the calculated synchronization index can purely reflect the clock deviation, which significantly improves the accuracy of clock calibration and avoids miscalibration.

[0188] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0189] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An industrial Internet of Things (IoT) system for online maintenance of transmitters, characterized in that, Includes an industrial IoT management platform, which is configured as follows: Acquire the monitoring data of the transmitter and the operating status data of the monitored equipment within the first preset time period; Based on the monitoring data and the operating condition data, a cross-correlation coefficient is determined, which is configured to characterize the correlation between the fluctuation trends of the monitoring data and the operating condition data. In response to the cross-correlation coefficient being less than a coefficient threshold, the abnormal fluctuation pattern corresponding to the monitoring data is identified; The reliability index of the transmitter is determined based on the data characteristics of the monitoring data; the data characteristics include at least one of packet loss rate, signal-to-noise ratio, and extreme value data. The common-mode drift component is determined based on the median of the changes in the monitoring data from multiple transmitters. Based on the data characteristics of the monitoring data, the operating status data of the monitored equipment, and the common mode drift component, the reliability index is determined through a prediction model, wherein the prediction model is a machine learning model. In response to the reliability index being less than a first reliability threshold, filtering parameters are generated based on the signal-to-noise ratio and spectral distribution characteristics of the monitoring data, and the transmitter is controlled to adjust the filtering window and / or cutoff frequency based on the filtering parameters. In response to the reliability index being less than or equal to the second reliability threshold, the monitored equipment is controlled to shut down, an alarm is triggered, and a manual maintenance work order is generated, wherein the first reliability threshold is greater than the second reliability threshold.

2. The system according to claim 1, characterized in that, The input to the prediction model also includes the onboard temperature data and power supply voltage data of the multiple transmitters.

3. The system according to claim 1, characterized in that, The industrial IoT management platform is further configured as follows: In response to the reliability index being greater than the second preset threshold, the delay index of the transmitter is determined based on the time difference sequence between the platform receiving time and the actual acquisition time of the monitoring data within the second preset time period, wherein the second preset time period includes multiple first preset time periods.

4. The system according to claim 3, characterized in that, The industrial IoT management platform is further configured as follows: In response to the latency index being greater than the index threshold, a frequency reduction parameter is generated based on the latency index; The transmitter is controlled to adjust the trigger frequency of the transmission timer based on the down-frequency parameter, and the retransmission waiting time is extended.

5. A method for online maintenance of a transmitter, characterized in that, The method is executed by an industrial IoT management platform, and the method includes: Acquire the monitoring data of the transmitter and the operating status data of the monitored equipment within the first preset time period; Based on the monitoring data and the operating condition data, a cross-correlation coefficient is determined, which is configured to characterize the correlation between the fluctuation trends of the monitoring data and the operating condition data. In response to the cross-correlation coefficient being less than a coefficient threshold, the abnormal fluctuation pattern corresponding to the monitoring data is identified; The reliability index of the transmitter is determined based on the data characteristics of the monitoring data; the data characteristics include at least one of packet loss rate, signal-to-noise ratio, and extreme value data. The common-mode drift component is determined based on the median of the changes in the monitoring data from multiple transmitters. Based on the data characteristics of the monitoring data, the operating status data of the monitored equipment, and the common mode drift component, the reliability index is determined through a prediction model, wherein the prediction model is a machine learning model. In response to the reliability index being less than a first reliability threshold, filtering parameters are generated based on the signal-to-noise ratio and spectral distribution characteristics of the monitoring data, and the transmitter is controlled to adjust the filtering window and / or cutoff frequency based on the filtering parameters. In response to the reliability index being less than or equal to the second reliability threshold, the monitored equipment is controlled to shut down, an alarm is triggered, and a manual maintenance work order is generated, wherein the first reliability threshold is greater than the second reliability threshold.

6. The method according to claim 5, characterized in that, The input to the prediction model also includes the onboard temperature data and power supply voltage data of the multiple transmitters.

7. The method according to claim 5, characterized in that, The method further includes: In response to the reliability index being greater than the second preset threshold, the delay index of the transmitter is determined based on the time difference sequence between the platform receiving time and the actual acquisition time of the monitoring data within the second preset time period, wherein the second preset time period includes multiple first preset time periods.

8. The method according to claim 7, characterized in that, The method further includes: In response to the latency index being greater than the index threshold, a frequency reduction parameter is generated based on the latency index; The transmitter is controlled to adjust the trigger frequency of the transmission timer based on the down-frequency parameter, and the retransmission waiting time is extended.

Images