A clock management method and system based on the internet of things

Abstract

The application relates to the technical field of clock management, and particularly discloses a clock management method and system based on the Internet of Things, which comprises the following steps: in the same time zone, extracting a median reference of clock instantaneous running time values reported by Internet of Things convergence nodes to generate a group reference time value; calculating a single-period outlying deviation amount of the instantaneous running time value based on the group reference time value, and completing clock equipment stability evaluation through zero-crossing rate statistics; fitting a fluctuation slope of a clock power supply voltage sampling value to solve equipment voltage degradation characteristic values; combining a fault discrimination threshold group and a time zone maintenance record, cross-mapping stability results and voltage characteristic values to generate a maintenance instruction and store the maintenance instruction in a device shadow; and grouping and sorting fault clocks in proximity according to stability, installation coordinates and maintenance urgency to output an operation and maintenance task execution sequence; the application can improve the remote operation and maintenance efficiency and fault early warning accuracy of clock equipment and reduce the cost of manual inspection.

Description

Technical Field

[0001] This invention relates to the field of watch management technology, and in particular to a watch management method and system based on the Internet of Things. Background Technology

[0002] Existing regional distributed clock timekeeping monitoring methods mostly rely on arithmetic averages or single-sample comparisons as benchmarks, without robustly extracting the median of instantaneous timekeeping values. This makes them susceptible to benchmark shifts caused by momentary equipment anomalies and communication jitter, resulting in poor anti-interference capabilities for group reference times and distorted consistency in timekeeping deviation calculations. Furthermore, equipment timekeeping stability is determined solely by fixed thresholds or statistical variances, lacking statistics on the zero-crossing rate of single-period outlier deviations. This makes it impossible to accurately capture fluctuations and random drifts in timekeeping errors, distinguish between short-term jitter and long-term instability, and provides poor early warning of faults.

[0003] Traditional clock power monitoring only compares discrete voltage samples with low-voltage thresholds, lacking methods for fitting voltage fluctuation slopes and extracting degradation features. This makes it impossible to quantify gradual voltage degradation trends and identify hidden faults such as power supply aging. Current maintenance practices separate timekeeping and voltage status analysis, lacking a cross-mapping mechanism between fault thresholds and maintenance knowledge, and failing to generate standardized maintenance instructions for storage in IoT device shadows. Maintenance tasks are not grouped and prioritized based on equipment stability, installation coordinates, and urgency, resulting in disorganized and inefficient maintenance operations. Therefore, relevant management solutions are urgently needed to address various technical pain points and improve the accuracy of remote maintenance and fault diagnosis for distributed clocks. Summary of the Invention

[0004] This invention provides a clock management method and system based on the Internet of Things to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a clock management method based on the Internet of Things, comprising:

[0006] A1: Within the same time zone, the median benchmark is extracted from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node to obtain the group reference time value of the clock devices.

[0007] A2: Based on the group reference time value, the deviation of the instantaneous time value is calculated to obtain the single-period outlier deviation of the clock device, and the zero-crossing rate of the single-period outlier deviation is statistically analyzed to evaluate the stability of the clock device.

[0008] A3: Fit the voltage fluctuation slope of the power supply voltage sample value of the watch device to obtain the voltage decay characteristic value of the watch device;

[0009] A4: Based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, the stability assessment result and the voltage decay characteristic value are cross-mapped with fault instructions, and the obtained maintenance instruction string is stored as an IoT management instruction in the IoT device shadow of the clock device;

[0010] A5: Based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the device in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain the task execution sequence for the IoT terminal devices of maintenance personnel.

[0011] In a preferred embodiment, the step of extracting the median benchmark from the instantaneous timekeeping values ​​reported by the clock devices via the IoT aggregation node within the same time zone to obtain the group reference time value of the clock devices includes:

[0012] Within the same time zone, the raw data frames received by the IoT aggregation node from the clock device are completely decapsulated to obtain the instantaneous timekeeping value of the clock device;

[0013] The instantaneous time values ​​are sorted in ascending order to obtain an ordered sequence of instantaneous time values;

[0014] The middle time value of the ordered time value sequence is used as the group reference time value. If the number of elements in the ordered time value sequence is even, the average of the two middle time values ​​is used as the group reference time value.

[0015] In a preferred embodiment, the step of calculating the deviation of the instantaneous timekeeping value based on the group reference time value to obtain the single-period outlier deviation of the clock device includes:

[0016] Based on the group reference time value, the difference between the instantaneous travel time value and the reference amplitude sequence of the instantaneous travel time value is obtained;

[0017] The extreme value elimination filter is applied to the deviation from the reference amplitude sequence to obtain the set of transient fluctuation amplitudes of the deviation from the reference amplitude sequence;

[0018] Perform sign zero-crossing identification on the transient fluctuation amplitude set to obtain the positive and negative transition identifier queue of the transient fluctuation amplitude set;

[0019] Periodic boundary cutting is performed on the positive and negative transition marker queue to obtain the deviation amplitude grouping queue of the positive and negative transition marker queue;

[0020] Outlier amplitude is extracted from the deviation amplitude grouping queue to obtain the single-cycle outlier deviation of the clock device.

[0021] In a preferred embodiment, the formula for calculating the single-period outlier deviation is:

[0022] ;

[0023] in, This refers to the single-period outlier deviation. The total number of groups in the deviation amplitude grouping queue. The first in the deviation magnitude grouping queue Instantaneous travel values ​​within each group, This is the reference time value for the group. The standard deviation of the instantaneous timekeeping value of the clock device under healthy conditions. This is the preset outlier sensitivity threshold.

[0024] In a preferred embodiment, the step of performing zero-crossing rate statistics on the single-period outlier deviation to evaluate the stability of the watchmaking device further includes:

[0025] The deviation time series of the clock device is constructed by arranging the single-period outlier deviations in chronological order.

[0026] The time series of the deviation is symbolized to obtain the deviation symbol sequence of the time series of the deviation;

[0027] The total number of symbol changes in the deviation symbol sequence is obtained by summing the number of adjacent symbol transitions.

[0028] The limit transition capacity of the deviation sign sequence is obtained by decreasing the sequence length value by a unit value.

[0029] The ratio of the total number of symbol changes to the limiting transition capacity is quantified to obtain the stability evaluation result of the clock device.

[0030] In a preferred embodiment, the step of fitting the voltage fluctuation slope of the power supply voltage sample value of the watch device to obtain the voltage decay characteristic value of the watch device includes:

[0031] The power supply voltage sample value of the clock device is divided into equal time windows to obtain the voltage segment subsequence and time window width of the power supply voltage sample value;

[0032] The voltage change amplitude value of the voltage segment subsequence is obtained by calculating the amplitude difference at the endpoints of the voltage segment subsequence;

[0033] Based on the time window width, the voltage change amplitude value is normalized to obtain the voltage fluctuation slope value of the voltage segmented subsequence;

[0034] By performing trend fitting on the voltage fluctuation slope value, the voltage decay characteristic value of the clock device is obtained.

[0035] In a preferred embodiment, the stability assessment results and the voltage degradation characteristic value are cross-mapped based on a preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, and the resulting maintenance instruction string is stored as an IoT management instruction in the IoT device shadow of the watch device, including:

[0036] Based on a preset fault discrimination threshold group, the stability evaluation results are compared with the stability level threshold points to obtain the stability level identifier of the watch device.

[0037] The voltage degradation characteristic value is mapped to the fault discrimination threshold group to obtain the degradation level identifier of the watch device;

[0038] Based on the combined key value composed of the stability level identifier and the degradation level identifier, the fault code mapping entries of the maintenance knowledge record in the same time zone are indexed and matched to obtain the maintenance instruction string of the clock device.

[0039] Based on the maintenance instruction string, the maintenance instruction field in the IoT device shadow of the clock device is overwritten with the maintenance instruction string to obtain the IoT management instruction of the IoT device shadow.

[0040] In a preferred embodiment, the step of indexing and matching fault code mapping entries in the maintenance knowledge records of the same time zone based on the combined key value composed of the stability level identifier and the degradation level identifier to obtain the maintenance instruction string of the watch device includes:

[0041] The stability level identifier and the degradation level identifier are encoded by binary bit concatenation to obtain the combined key value byte stream of the watch device;

[0042] Based on the combined key-value byte stream, hash slot positioning is performed on the fault code mapping entries of the maintenance knowledge records in the same time zone to obtain the matching address offset of the clock device.

[0043] Based on the matching address offset, extract the original fault data record from the maintenance knowledge record;

[0044] The original fault data record is parsed to obtain the maintenance instruction string for the clock device.

[0045] In a preferred embodiment, based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the devices in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain a task execution sequence for the IoT terminal devices of maintenance personnel, including:

[0046] Based on the urgency of the maintenance of the clock devices in the maintenance instruction string, the field values ​​of the clock devices are filtered to obtain a set of clock devices to be processed;

[0047] Based on the installation coordinates of the clock device, the devices to be processed in the set of devices to be processed are spatially clustered to obtain the spatial cluster groups of the devices to be processed.

[0048] The spatial clusters are sorted in descending order according to the stability assessment results to obtain the maintenance priority queue of the equipment.

[0049] The maintenance priority queue is joined end-to-end to obtain a task execution sequence for IoT terminal devices for maintenance personnel.

[0050] To address the aforementioned problems, the present invention also provides a watch management system based on the Internet of Things, the system comprising:

[0051] The median reference extraction module is used to extract the median reference from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node within the same time zone, so as to obtain the group reference time value of the clock devices.

[0052] The deviation zero-crossing statistics module is used to calculate the deviation of the instantaneous timekeeping value based on the group reference time value, obtain the single-cycle outlier deviation of the clock device, and perform zero-crossing rate statistics on the single-cycle outlier deviation to evaluate the stability of the clock device.

[0053] The voltage decay fitting module performs voltage fluctuation slope fitting on the power supply voltage sampling value of the watch device to obtain the voltage decay characteristic value of the watch device.

[0054] The fault instruction mapping module is used to perform fault instruction cross-mapping on the stability assessment results and the voltage decay characteristic value based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, and store the obtained maintenance instruction string as an IoT management instruction in the IoT device shadow of the clock device.

[0055] The grouping and sorting maintenance module is used to sort the devices to be processed in the clock device according to the stability assessment results and the installation coordinates of the clock device, combined with the maintenance urgency of the devices in the maintenance instruction string, to obtain a task execution sequence for the IoT terminal devices of maintenance personnel.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. Compared with existing technologies, this invention extracts the median benchmark of the instantaneous timekeeping value of the clock, avoiding benchmark offset caused by arithmetic averaging and single-sample comparison, effectively improving the robustness and anti-interference ability of the group reference time value. By calculating the outlier deviation in a single period and performing zero-crossing rate statistics, it can accurately capture the fluctuation and random drift characteristics of timekeeping error, refine the evaluation of equipment stability, distinguish between short-term jitter and long-term instability, and improve the accuracy of early warning of timekeeping faults. At the same time, by fitting the slope of the voltage sampling value and extracting the decay feature value, it can quantify the gradual voltage decay trend, accurately identify hidden faults such as power supply aging and power supply instability, and enhance the ability to predict and diagnose the health status of the power supply.

[0058] 2. This invention combines fault identification threshold groups with time zone maintenance records, cross-mapping stability assessment results with voltage decay characteristic values ​​to generate standardized maintenance instructions stored in the IoT device's shadow. This overcomes the drawbacks of separating timekeeping and voltage status analysis, enabling multi-dimensional joint fault assessment and remote collaborative management. By grouping and prioritizing devices based on stability, installation coordinates, and maintenance urgency, a maintenance task sequence is generated. This avoids disorderly cross-regional travel for maintenance personnel, reduces resource and path losses, and significantly improves the scheduling efficiency and task execution order of large-scale distributed clock devices. Attached Figure Description

[0059] Figure 1 A flowchart illustrating a clock management method based on the Internet of Things (IoT) according to an embodiment of the present invention is shown below.

[0060] Figure 2 A functional block diagram of a watch management system based on the Internet of Things provided in an embodiment of the present invention;

[0061] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0062] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0063] This application provides a clock management method based on the Internet of Things (IoT). The executing entity of this IoT-based clock management method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the IoT-based clock management method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0064] Reference Figure 1 The diagram shown is a flowchart illustrating a watch management method based on the Internet of Things (IoT) according to an embodiment of the present invention. In this embodiment, the watch management method based on the IoT includes:

[0065] A1: Within the same time zone, the median benchmark is extracted from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node to obtain the group reference time value of the clock devices.

[0066] In this embodiment of the invention, the step of extracting the median benchmark from the instantaneous timekeeping values ​​reported by the clock devices via the IoT aggregation node within the same time zone to obtain the group reference time value of the clock devices includes:

[0067] Within the same time zone, the raw data frames received by the IoT aggregation node from the clock device are completely decapsulated to obtain the instantaneous timekeeping value of the clock device;

[0068] The instantaneous time values ​​are sorted in ascending order to obtain an ordered sequence of instantaneous time values;

[0069] The middle time value of the ordered time value sequence is used as the group reference time value. If the number of elements in the ordered time value sequence is even, the average of the two middle time values ​​is used as the group reference time value.

[0070] Within the same time zone, the instantaneous time values ​​reported by clock devices via the IoT aggregation node form the basis for extracting the group reference time value. The same time zone refers to a 24 / 24 longitude area defined by Earth's longitude, where all clock devices share the same standard time. The clock devices are smart clocks with built-in IoT communication modules, capable of proactively initiating reporting actions according to pre-set time cycles. Reporting refers to the complete process by which the clock device transmits the current displayed time value stored in its internal register to the IoT aggregation node via a wireless communication link. The IoT aggregation node is a centralized data collection device deployed in the IoT network, specifically responsible for receiving reported data from multiple clock devices. The instantaneous time value refers to the time value indicated by the dial hand or digital display at the moment the clock device triggers the reporting action; this value is accurate to the second or even smaller units. Median benchmark extraction is a method of finding the middle value from a set of values ​​as a benchmark value, without relying on any statistical assumptions or pre-set thresholds. The group reference time value refers to a unified time value that is obtained through the median benchmark extraction operation and can represent the overall timekeeping level of all participating clock devices in the same time zone.

[0071] The system receives raw data frames from watch devices and performs a complete frame decapsulation operation on these frames to obtain the instantaneous time value of each watch device. A raw data frame is a complete data packet encapsulated by the watch device according to the IoT communication protocol before sending reported data. This data packet includes a frame start identifier, a device address field, a time data payload, a checksum field, and a frame end identifier. Complete frame decapsulation involves, after receiving the raw data frame, first scanning the data stream and identifying the frame start and end identifiers. Then, starting from the start identifier, it reads the entire data frame byte by byte. Next, it locates the byte position of the time data payload according to the protocol specification. Finally, it extracts the binary data representing the time value from this position and converts it into a readable decimal time value. After completing this series of actions, the frame start identifier, device address field, checksum field, and frame end identifier are discarded, retaining only the extracted time value, which is the instantaneous time value of the watch device.

[0072] All acquired instantaneous time-lapse values ​​are sorted in ascending order to obtain an ordered sequence of instantaneous time-lapse values. Ascending sorting involves placing all collected instantaneous time-lapse values ​​in a sortable storage area, then comparing the size of any two values ​​one by one, moving the smaller value to the front of the sequence and the larger value to the back, repeating this comparison and swapping process until every value at the beginning of the sequence is no greater than any value after it. During the sorting process, each comparison is performed on two specific instantaneous time-lapse values; swapping positions only changes the order of the values ​​in the storage area, not the values ​​themselves. The ordered sequence of time-lapse values ​​is a list of values ​​obtained after the ascending sorting process, where the instantaneous time-lapse values ​​are arranged in ascending order, with the first value being the smallest and the last value being the largest.

[0073] After obtaining the ordered time value sequence, the parity of the number of elements in the sequence is determined, and different methods are selected to extract the group reference time value based on the result. The number of elements refers to the total number of all instantaneous time values ​​in the ordered time value sequence, which is exactly equal to the number of clock devices participating in this reporting action within the same time zone. If the number of elements in the ordered time value sequence is odd, the exact center position of the sequence is found, where the number of elements before and after this position is equal. The time value corresponding to this center position is then directly used as the group reference time value. The time value at the center position is simply this central time value and does not require any mathematical operations.

[0074] If the number of elements in the ordered time value sequence is even, then the sequence does not have a single central position, but rather two central positions. Find the two time values ​​located at the exact center; these two values ​​occupy the left and right central positions of the sequence, respectively, with the number of elements to their left equal to the number of elements to their right. Extract these two time values ​​and perform an averaging operation: add the two values ​​together and divide by two. The result is used as the group reference time value. The averaging operation requires no external parameters; it simply adds the two time values ​​and then divides them, resulting in a new time value. At this point, the group reference time value has been successfully extracted. This value will be used for subsequent time management operations of all clocks within the same time zone.

[0075] The beneficial effect is that by extracting the median benchmark from the instantaneous time values ​​reported by the clock devices through the IoT aggregation node within the same time zone, a group reference time value can be obtained. This can effectively eliminate the interference of abnormal time values ​​caused by individual clock devices due to malfunctions or communication interference on the overall time benchmark, significantly improve the accuracy and stability of group time synchronization, and provide a reliable and unified time reference for the IoT clock management system.

[0076] A2: Based on the group reference time value, the deviation of the instantaneous time value is calculated to obtain the single-period outlier deviation of the clock device, and the zero-crossing rate of the single-period outlier deviation is statistically analyzed to evaluate the stability of the clock device.

[0077] In this embodiment of the invention, the step of calculating the deviation of the instantaneous timekeeping value based on the group reference time value to obtain the single-period outlier deviation of the clock device includes:

[0078] Based on the group reference time value, the difference between the instantaneous travel time value and the reference amplitude sequence of the instantaneous travel time value is obtained;

[0079] The extreme value elimination filter is applied to the deviation from the reference amplitude sequence to obtain the set of transient fluctuation amplitudes of the deviation from the reference amplitude sequence;

[0080] Perform sign zero-crossing identification on the transient fluctuation amplitude set to obtain the positive and negative transition identifier queue of the transient fluctuation amplitude set;

[0081] Periodic boundary cutting is performed on the positive and negative transition marker queue to obtain the deviation amplitude grouping queue of the positive and negative transition marker queue;

[0082] Outlier amplitude is extracted from the deviation amplitude grouping queue to obtain the single-cycle outlier deviation of the clock device.

[0083] The formula for calculating the single-period outlier bias is as follows:

[0084] ;

[0085] in, This refers to the single-period outlier deviation. The total number of groups in the deviation amplitude grouping queue. The first in the deviation magnitude grouping queue Instantaneous travel values ​​within each group, This is the reference time value for the group. The standard deviation of the instantaneous timekeeping value of the clock device under healthy conditions. This is the preset outlier sensitivity threshold.

[0086] The method of performing zero-crossing rate statistics on the single-period outlier deviation to evaluate the stability of the watchmaking device further includes:

[0087] The deviation time series of the clock device is constructed by arranging the single-period outlier deviations in chronological order.

[0088] The time series of the deviation is symbolized to obtain the deviation symbol sequence of the time series of the deviation;

[0089] The total number of symbol changes in the deviation symbol sequence is obtained by summing the number of adjacent symbol transitions.

[0090] The limit transition capacity of the deviation sign sequence is obtained by decreasing the sequence length value by a unit value.

[0091] The ratio of the total number of symbol changes to the limiting transition capacity is quantified to obtain the stability evaluation result of the clock device.

[0092] The group reference time value is a time value extracted from the previous action, representing the overall timekeeping level of all clock devices within the same time zone. The instantaneous time value is the specific time reading obtained by the IoT aggregation node from the decapsulated raw data frames reported by each clock device. Difference calculation involves subtracting the group reference time value from each instantaneous time value to obtain the time difference between each instantaneous time value and the group reference time value. All differences are arranged in the reporting order to form a deviation from the reference amplitude sequence, where positive numbers indicate that the instantaneous time value is faster than the group reference time value, and negative numbers indicate that it is slower.

[0093] The fluctuation amplitude of the deviation from the reference amplitude sequence is extracted to obtain a transient fluctuation amplitude set. Fluctuation amplitude extraction refers to iterating through each deviation from the reference amplitude, extracting its absolute magnitude, and ignoring the positive or negative sign. All absolute magnitudes are arranged in their original order to form a transient fluctuation amplitude set, where each value represents the absolute degree of deviation from the group reference time value within the reporting period. Positive and negative transition indicators refer to re-referencing the original deviation from the reference amplitude sequence and observing the positive or negative sign of each deviation from the reference amplitude in sequence. For the first value, its sign is directly recorded as positive or negative. For each subsequent value, its sign is compared with the previous one: if they are the same, an unchanged indicator is recorded; otherwise, a transition indicator is recorded. All indicators form a positive and negative transition indicator queue in sequence, and each indicator describes whether the deviation direction has changed relative to the previous one.

[0094] The positive and negative transition marker queue is periodically segmented to obtain a deviation amplitude grouping queue. Periodic boundary segmentation refers to dividing the queue into several consecutive groups based on the transition marker positions as the cutting boundaries. Scanning starts from the beginning, and when a transition marker is encountered, a segment is cut before that marker, grouping adjacent unchanged markers into the same group. Each sub-segment corresponds to a time period, during which the deviation direction remains unchanged. These sub-segments are arranged in chronological order to form the deviation amplitude grouping queue, with each sub-segment containing several markers with the same symbol.

[0095] The first group in the deviation amplitude grouping queue is retrieved. Based on its identified time position, all fluctuation amplitude values ​​within the corresponding time range are extracted from the transient fluctuation amplitude set, and these values ​​are added together to obtain the cumulative fluctuation amplitude of that group. This process of taking values ​​and accumulating is repeated for each group to obtain the cumulative fluctuation amplitude for each group. The maximum value among all the cumulative fluctuation amplitudes of the groups is calculated and taken as the group deviation. The group deviation represents the maximum cumulative degree of deviation of the instantaneous time value from the group reference time value over the longest period in which the deviation direction remains unchanged. This group deviation is the single-period outlier deviation of the clock device.

[0096] The total number of groups in the deviation amplitude grouping queue is derived from the number of groups obtained after periodically dividing the positive and negative transition marker queues. The instantaneous time value within each group is derived from the original time data of that group. The group reference time value is derived from the result of extracting the median benchmark of the instantaneous time values ​​of all clocks within the time zone. The standard deviation of the instantaneous time values ​​of a clock in a healthy state is derived from the statistical value of the dispersion of instantaneous time values ​​recorded at the time of manufacture or during historical normal operation. The preset outlier sensitivity threshold is a judgment boundary value pre-set by maintenance personnel according to the accuracy requirements of the clocks, used to control the sensitivity to outlier deviations.

[0097] The significance of this formula lies in calculating the outlier deviation of a clock device within a single cycle. First, for each instantaneous timekeeping value within each group, the formula calculates the absolute value of the difference between that instantaneous timekeeping value and the group reference time value. Then, this absolute value is divided by the standard deviation under healthy conditions to obtain a normalized deviation factor. Next, a preset outlier sensitivity threshold is subtracted from this normalized factor; if the result is positive, it is retained; otherwise, it is zero. Then, all the processed values ​​within the same group are summed to obtain the cumulative deviation for that group. Finally, the maximum cumulative deviation value is found among all groups, and this maximum value is multiplied by the standard deviation under healthy conditions. The entire formula outputs the product of the maximum cumulative deviation of the device over multiple fluctuation cycles and the healthy standard deviation, reflecting the total total timekeeping deviation of the device during extreme fluctuation cycles.

[0098] The formula exhibits the following trend: When the deviations between the instantaneous travel values ​​and the group reference time values ​​within all groups are small, and each normalized deviation factor does not exceed the preset outlier sensitivity threshold, the cumulative result within each group is zero, and the single-period outlier deviation is zero. When some instantaneous travel values ​​within a group have larger deviations, causing the normalized deviation factor to exceed the sensitivity threshold, the excess portion is accumulated, increasing the cumulative deviation of that group and consequently increasing the single-period outlier deviation. As the standard deviation of instantaneous travel values ​​increases under healthy conditions, the single-period outlier deviation amplifies proportionally. As the preset outlier sensitivity threshold increases, only larger deviations are included, resulting in a decrease in the overall single-period outlier deviation; conversely, as the sensitivity threshold decreases, more deviations are included, leading to an increase in the overall single-period outlier deviation.

[0099] For each single-cycle outlier deviation obtained by the same clock device in multiple consecutive reporting periods, they are arranged sequentially according to the reporting period, with each deviation occupying a position, forming a deviation time series. This deviation time series is then symbolized to obtain a deviation symbol sequence. Symbolization involves iterating through each single-cycle outlier deviation: values ​​greater than zero are converted to a positive sign, values ​​equal to zero are converted to a zero sign, and values ​​less than zero are converted to a negative sign. All symbols are arranged sequentially to form the deviation symbol sequence, with each symbol describing the positive or negative nature of the cumulative deviation within the corresponding period.

[0100] The accumulation of adjacent symbol transition counts refers to comparing the current symbol with the previous symbol sequentially, starting from the second symbol in the deviation symbol sequence: if they are different, the transition count is incremented by one; otherwise, it is not. After comparing all adjacent positions, the transition count is the total number of symbol changes, which describes the total frequency of positive and negative changes in the cumulative deviation over a continuous period. The sequence length of the deviation symbol sequence is subtracted by a unit value to obtain the limiting transition capacity. The sequence length is the total number of symbols in the deviation symbol sequence, equal to the total number of reporting periods involved in the evaluation. Subtracting one unit value from the sequence length yields the limiting transition capacity, representing the maximum possible number of adjacent symbol changes. The ratio is calculated by dividing the total number of symbol changes by the limiting transition capacity, resulting in a proportional value between zero and one. Level quantization divides stability into multiple discrete levels based on this proportional value: a larger proportional value indicates more frequent symbol changes and a lower stability level; a smaller proportional value indicates fewer symbol changes and a higher stability level. The final determined stability level is output as the stability evaluation result for the clock device.

[0101] The beneficial effects are that by calculating the outlier deviation in a single period and performing zero-crossing rate statistics, the cumulative deviation of each clock device relative to the group reference time can be accurately quantified. At the same time, the stability can be evaluated by the sign change frequency, thereby effectively identifying unstable clock devices. This provides a reliable basis for judging the device status of the IoT clock management system, significantly improving the system's ability to identify abnormal devices and the overall operational reliability.

[0102] A3: Fit the voltage fluctuation slope of the power supply voltage sample value of the watch device to obtain the voltage decay characteristic value of the watch device;

[0103] In this embodiment of the invention, the step of fitting the voltage fluctuation slope of the power supply voltage sample value of the watch device to obtain the voltage decay characteristic value of the watch device includes:

[0104] The power supply voltage sample value of the clock device is divided into equal time windows to obtain the voltage segment subsequence and time window width of the power supply voltage sample value;

[0105] The voltage change amplitude value of the voltage segment subsequence is obtained by calculating the amplitude difference at the endpoints of the voltage segment subsequence;

[0106] Based on the time window width, the voltage change amplitude value is normalized to obtain the voltage fluctuation slope value of the voltage segmented subsequence;

[0107] By performing trend fitting on the voltage fluctuation slope value, the voltage decay characteristic value of the clock device is obtained.

[0108] The voltage sensor built into the watch continuously collects power supply voltage values ​​at fixed time intervals, obtaining one voltage sample value each time. These voltage sample values ​​are received and arranged into a complete voltage sample value sequence according to the chronological order of collection. Then, starting from the beginning of the sequence, a fixed-length time window is selected, containing several consecutive voltage sample values. All voltage sample values ​​within this time window are extracted to form a voltage segmented subsequence. The time span covered by this time window is recorded; this span is the time window width. After the first extraction, the time window is slid forward by one step, extracting the next voltage segmented subsequence and its corresponding time window width. This sliding and extraction process is repeated until the time window covers the end of the sequence. All extracted voltage segmented subsequences are arranged in chronological order, forming a time window segmentation sequence of voltage sample values, with each voltage segmented subsequence accompanied by its corresponding time window width.

[0109] For each voltage segment subsequence, first find the first voltage sample value in the subsequence, i.e., the voltage amplitude at the starting endpoint. Then find the last voltage sample value in the subsequence, i.e., the voltage amplitude at the ending endpoint. Subtract the voltage amplitude at the starting endpoint from the voltage amplitude at the ending endpoint to obtain the difference between the two endpoints. This difference may be positive or negative. Take the absolute magnitude of this difference, ignoring the sign; the result is the voltage change amplitude. The voltage change amplitude describes the absolute total change in the power supply voltage from the starting time to the ending time within that time window, regardless of whether the voltage is rising or falling.

[0110] Retrieve the previously recorded time window width corresponding to the voltage segment subsequence. The time window width represents the length of time during which the voltage change occurs. Divide the voltage change amplitude by the time window width to obtain the average rate of voltage change per unit time. This average rate is the voltage fluctuation slope value. The sign of the voltage fluctuation slope value is determined by the actual direction of the voltage change amplitude: if the voltage at the end point is lower than the voltage at the beginning point, the voltage change amplitude value is negative, and the voltage fluctuation slope value is negative, indicating that the voltage is decreasing; if the voltage at the end point is higher than the voltage at the beginning point, the voltage fluctuation slope value is positive, indicating that the voltage is increasing. Repeat the above normalization operation for each voltage segment subsequence to obtain the voltage fluctuation slope value corresponding to each time window.

[0111] Arrange all voltage fluctuation slope values ​​in chronological order according to the time window, forming a voltage fluctuation slope value sequence. Each value in this sequence reflects the rate and direction of voltage change within the corresponding time window. Observe the overall trend of these slope values ​​as the time window number increases. If the slope values ​​gradually change from positive to negative, or from smaller negative values ​​to larger negative values, it indicates that the voltage drop is accelerating. Take the last slope value in the voltage fluctuation slope value sequence, and simultaneously take the cumulative result of all slope values ​​in the sequence. Combine the last slope value with the cumulative result, and determine the voltage decay characteristic value by judging the overall direction of slope value change. When the last slope value is negative and its absolute value is greater than the absolute values ​​of most of the preceding slope values, the voltage decay characteristic value is judged as accelerated decay. When the last slope value is close to zero and the preceding slope values ​​are also close to zero, the voltage decay characteristic value is judged as stable voltage. When the last slope value is positive, the voltage decay characteristic value is judged as voltage recovery. Output this voltage decay characteristic value as the basis for judging the power health status of the clock device.

[0112] The beneficial effects are that by fitting the voltage fluctuation slope of the power supply voltage sampling value of the clock device, the voltage decay characteristic value can be obtained, which can accurately capture the slow downward trend of voltage over time, detect early signs of power performance degradation in advance, provide a reliable basis for power health status assessment of the clock device, thereby effectively avoiding timekeeping errors or shutdown failures caused by insufficient voltage, and improving the device maintenance foresight and operational stability of the IoT clock management system.

[0113] A4: Based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, the stability assessment result and the voltage decay characteristic value are cross-mapped with fault instructions, and the obtained maintenance instruction string is stored as an IoT management instruction in the IoT device shadow of the clock device;

[0114] In this embodiment of the invention, the method of cross-mapping the stability assessment results and the voltage degradation characteristic value based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, and storing the resulting maintenance instruction string as an IoT management instruction in the IoT device shadow of the watch device, includes:

[0115] Based on a preset fault discrimination threshold group, the stability evaluation results are compared with the stability level threshold points to obtain the stability level identifier of the watch device.

[0116] The voltage degradation characteristic value is mapped to the fault discrimination threshold group to obtain the degradation level identifier of the watch device;

[0117] Based on the combined key value composed of the stability level identifier and the degradation level identifier, the fault code mapping entries of the maintenance knowledge record in the same time zone are indexed and matched to obtain the maintenance instruction string of the clock device.

[0118] Based on the maintenance instruction string, the maintenance instruction field in the IoT device shadow of the clock device is overwritten with the maintenance instruction string to obtain the IoT management instruction of the IoT device shadow.

[0119] The combined key value composed of the stability level identifier and the degradation level identifier is used to perform index matching retrieval of fault code mapping entries in the maintenance knowledge records of the same time zone to obtain the maintenance instruction string of the watch equipment, including:

[0120] The stability level identifier and the degradation level identifier are encoded by binary bit concatenation to obtain the combined key value byte stream of the watch device;

[0121] Based on the combined key-value byte stream, hash slot positioning is performed on the fault code mapping entries of the maintenance knowledge records in the same time zone to obtain the matching address offset of the clock device.

[0122] Based on the matching address offset, extract the original fault data record from the maintenance knowledge record;

[0123] The original fault data record is parsed to obtain the maintenance instruction string for the clock device.

[0124] The preset fault discrimination threshold set is a multi-level numerical boundary set pre-stored in the IoT aggregation node, with each boundary level corresponding to a stability level. The stability assessment result is the stability level value of the clock device obtained through zero-crossing rate statistics in the previous steps. The stability level threshold comparison refers to the IoT aggregation node retrieving the stability assessment result and comparing it sequentially with each boundary level in the fault discrimination threshold set to find the level interval into which the value falls. When the stability assessment result is less than the first-level boundary, the threshold is the lowest stability level. When the stability assessment result is between the first and second-level boundaries, the threshold is the second lowest stability level. This process continues until a corresponding level interval is found. The level name corresponding to the threshold is output as the stability level identifier. The stability level identifier is a text or symbolic label that directly indicates the stability level of the clock device.

[0125] The voltage degradation characteristic value is a numerical value describing the downward trend of the power supply voltage, obtained by fitting the voltage fluctuation slope in the previous step. The fault discrimination threshold group also contains multi-level boundaries for the voltage degradation characteristic value. Mapping the voltage degradation characteristic value to the fault discrimination threshold group means that the IoT aggregation node extracts the voltage degradation characteristic value and matches its value with the pre-divided intervals in the threshold group. If the voltage degradation characteristic value is positive and less than a certain positive boundary, it is mapped to a normal voltage indicator. If the voltage degradation characteristic value is negative and the absolute value is small, it is mapped to a slight degradation indicator. If the voltage degradation characteristic value is negative and the absolute value is large, it is mapped to a severe degradation indicator. A degradation level indicator is output based on the matching result. The degradation level indicator describes the degree of degradation in the power health of the clock device.

[0126] The combined key value, consisting of the stability level identifier and the degradation level identifier, refers to merging these two identifiers into a unique search keyword. The stability level identifier and the degradation level identifier are concatenated in a fixed order, for example, placing the stability level identifier first and the degradation level identifier last, forming a new composite identifier. This composite identifier is the combined key value. The maintenance knowledge record for the same time zone is a pre-built data table that stores fault code mapping entries summarized from historical maintenance cases of all clock equipment within that time zone. Each fault code mapping entry contains an input key value and a corresponding output maintenance instruction string. Index matching retrieval refers to using the combined key value as the search condition to compare each fault code mapping entry in the maintenance knowledge record, finding an entry that completely matches the combined key value. Once found, the maintenance instruction string stored in that entry is retrieved. The maintenance instruction string is a text describing the specific maintenance operations that should be performed for the current stability and voltage degradation conditions, such as replacing the battery, calibrating the timekeeping, and cleaning the contacts.

[0127] After the maintenance instruction string is retrieved, it needs to be stored in the IoT device shadow of the watch device. The IoT device shadow is a virtual data storage area maintained by the IoT platform for each watch device. This storage area saves the expected state and actual state of the device in key-value pairs. A dedicated field named the maintenance instruction field stores the most recently issued maintenance instruction to the device. Overwriting the maintenance instruction string involves retrieving the newly retrieved maintenance instruction string, locating it in the maintenance instruction field within the IoT device shadow, deleting all existing content from that field, and then writing the new maintenance instruction string into it. After writing, the content of the maintenance instruction field is updated to the latest maintenance instruction string. At this point, this maintenance instruction field in the IoT device shadow becomes the IoT management instruction. The IoT management instruction is the platform's command carrier for managing the device. It can subsequently be sent to the corresponding watch device via the IoT communication link to guide on-site maintenance personnel or the device itself to perform corresponding operations.

[0128] To accurately retrieve maintenance instruction strings from the maintenance knowledge record, further processing of the composite key value is required. The stability level identifier and degradation level identifier are encoded using binary bit concatenation. First, the stability level identifier is converted into a fixed-length binary value; for example, the lowest stability level corresponds to 0-0, the next lowest to 0-1, and so on. Then, the degradation level identifier is also converted into a fixed-length binary value. Finally, these two binary values ​​are concatenated to form a new binary sequence. This binary sequence is the composite key value byte stream. The composite key value byte stream is a continuous block of binary data, the length of which is equal to the sum of the binary lengths of the two identifiers.

[0129] Hash slot location is performed on fault code mapping entries in maintenance knowledge records within the same time zone using a combined key-value byte stream. Hash slot location involves inputting the combined key-value byte stream into a hash function, which maps a binary sequence of arbitrary length to an integer hash value within a fixed range. This hash value corresponds to a slot number in the maintenance knowledge record's storage space. Based on this slot number, the system directly jumps to the corresponding location in the storage space; this location is the matching address offset. The matching address offset is a numerical value representing the byte distance from the starting position in the storage space to the target entry.

[0130] Based on the matching address offset, the original fault data record is extracted from the storage space of the maintenance knowledge record. The original fault data record refers to the unparsed binary data block stored at the matching address offset, which contains information such as fault codes, maintenance descriptions, and instruction codes. The extraction process involves reading a fixed-length byte data segment starting from the offset and copying it entirely. Then, this original fault data record is parsed using a format string. Format string parsing involves disassembling each field in the binary data block according to a pre-defined data format template. After parsing, a plain text maintenance instruction string is extracted. This maintenance instruction string can then be used for subsequent IoT device shadow overwrite operations.

[0131] The beneficial effect is that by cross-mapping the stability assessment results and voltage degradation characteristic values ​​based on the preset fault discrimination threshold group and maintenance knowledge records, the system can automatically generate maintenance instruction strings for each watch device and store them in the IoT device shadow. This achieves accurate matching and instant issuance of fault diagnosis and maintenance instructions, significantly improving the intelligent operation and maintenance efficiency of the IoT watch management system and the timeliness of equipment fault response.

[0132] A5: Based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the device in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain the task execution sequence for the IoT terminal devices of maintenance personnel.

[0133] In this embodiment of the invention, based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the devices in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain a task execution sequence for the IoT terminal devices of maintenance personnel, including:

[0134] Based on the urgency of the maintenance of the clock devices in the maintenance instruction string, the field values ​​of the clock devices are filtered to obtain a set of clock devices to be processed;

[0135] Based on the installation coordinates of the clock device, the devices to be processed in the set of devices to be processed are divided into spatial proximity clusters to obtain spatial cluster groups of the devices to be processed.

[0136] The spatial clusters are sorted in descending order according to the stability assessment results to obtain the maintenance priority queue of the equipment.

[0137] The maintenance priority queue is joined end-to-end to obtain a task execution sequence for IoT terminal devices for maintenance personnel.

[0138] The maintenance urgency level of the clock devices in the maintenance instruction string is a predefined field within the string. This field uses fixed-level vocabulary to indicate the urgency at which the device needs to be addressed, such as urgent, moderate, or deferred. Field value filtering involves the IoT aggregation node traversing all clock devices that have generated maintenance instruction strings and reading the maintenance urgency level field from each string. All clock devices with an urgent maintenance urgency level are selected and placed into a temporary set. The IoT aggregation node continues filtering for devices with moderate or deferred maintenance urgency levels, placing them into their corresponding sets. Finally, all clock devices with a non-empty maintenance urgency level that require processing are merged together to form a set of devices awaiting processing. Each device in this set requires at least one maintenance operation and is initially categorized according to its urgency level.

[0139] The installation coordinates of clock devices refer to the geographical coordinates recorded during deployment. These coordinates are typically composed of longitude and latitude values ​​and are stored in the device archive of the IoT platform. Spatial proximity clustering involves the IoT aggregation node taking all devices in the set of devices to be processed and reading the installation coordinates of each device one by one. Then, it calculates the geographical distance between the installation coordinates of every two devices. The IoT aggregation node sets a fixed distance threshold; when the distance between two devices is less than this threshold, the two devices are grouped into the same group. Starting with the first device to be processed, the IoT aggregation node uses it as the seed for the first group and then iterates through the other devices, adding all devices whose distance to the seed device is less than the threshold to that group. After the first group is built, the next ungrouped device is taken as the seed for a new group, and the distance comparison and grouping process is repeated until all devices to be processed are assigned to a group. These groups are the spatial clusters of the devices to be processed. Clock devices within each spatial cluster are physically close to each other, facilitating maintenance personnel to reach the area at once to perform multiple maintenance tasks.

[0140] The stability assessment result is the stability level value of the clock device obtained through zero-crossing rate statistics in the previous steps. The higher the value, the more unstable the device. Sort the spatial clusters in descending order of stability assessment results. This means that the IoT aggregation node calculates the maximum stability assessment result of all devices to be processed within each spatial cluster, and uses this maximum value as the representative stability value for the entire group. Then, all spatial clusters are sorted in descending order of this representative value. After sorting, the group with the worst stability is placed at the front of the queue, and the groups with better stability are placed at the back. This sorted list of groups is the device maintenance priority queue. The maintenance priority queue determines which groups maintenance personnel should process first, with groups with poor stability receiving higher processing priority.

[0141] The head-to-tail merging of the maintenance priority queue refers to the IoT aggregation node internally sorting the devices to be processed within each spatial cluster of the maintenance priority queue according to their stability assessment results from highest to lowest. Then, starting with the first device in the first group, information for each device is retrieved sequentially, followed by the second device in the first group, and so on, until all devices in the first group have been retrieved. This process continues until all devices in all groups have been retrieved, and then the first and second devices in the second group, and so on. Once all devices in all groups have been retrieved, the device information is concatenated into a long sequence according to the retrieval order. If a group has a small number of devices, the process moves directly to the next group after retrieving all devices from that group. This head-to-tail merging method ensures that devices from different groups appear alternately, prioritizing devices with high stability assessment results while also considering the spatial clustering characteristics between groups. The resulting sequence is the task execution sequence for the IoT terminal devices used by maintenance personnel. Each entry in the task execution sequence corresponds to a device to be processed and includes the device's identifier, installation coordinates, maintenance instruction string, and execution sequence number. After receiving this task execution sequence, the maintenance personnel's IoT terminal device navigates to the corresponding installation coordinates in the order of the devices in the sequence and executes the corresponding maintenance instructions.

[0142] The beneficial effects are that by combining stability assessment results, installation coordinates, and maintenance urgency, the devices to be processed are grouped and sorted by proximity to generate a task execution sequence. This enables maintenance personnel to plan the optimal work path that takes into account both the severity of equipment failure and geographical proximity, significantly reducing unnecessary travel and response delays, and improving the on-site maintenance efficiency and resource scheduling accuracy of the IoT watch management system.

[0143] like Figure 2 The diagram shown is a functional block diagram of a watch management system based on the Internet of Things provided in an embodiment of the present invention.

[0144] The IoT-based watch management system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the IoT-based watch management system 100 may include a median reference extraction module 101, a deviation zero-crossing statistics module 102, a voltage decay fitting module 103, a fault command mapping module 104, and a grouping and sorting maintenance module 105. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0145] In this embodiment, the functions of each module / unit are as follows:

[0146] The median reference extraction module 101 is used to extract the median reference from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node within the same time zone, so as to obtain the group reference time value of the clock devices.

[0147] The deviation zero-crossing statistics module 102 is used to calculate the deviation of the instantaneous timekeeping value based on the group reference time value, obtain the single-cycle outlier deviation of the clock device, and perform zero-crossing rate statistics on the single-cycle outlier deviation to evaluate the stability of the clock device.

[0148] The voltage decay fitting module 103 performs voltage fluctuation slope fitting on the power supply voltage sampling value of the watch device to obtain the voltage decay characteristic value of the watch device.

[0149] The fault instruction mapping module 104 is used to perform fault instruction cross-mapping on the stability assessment result and the voltage decay characteristic value based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, and store the obtained maintenance instruction string as an Internet of Things management instruction in the Internet of Things device shadow of the clock device.

[0150] The grouping and sorting maintenance module 105 is used to sort the devices to be processed in the clock device by proximity based on the stability assessment results and the installation coordinates of the clock device, combined with the maintenance urgency of the devices in the maintenance instruction string, so as to obtain a task execution sequence for the Internet of Things terminal devices for maintenance personnel.

[0151] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0152] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0153] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0154] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0155] The embodiments of this application can acquire and process relevant data based on an artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A clock management method based on the Internet of Things, characterized in that, The method includes: A1: Within the same time zone, the median benchmark is extracted from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node to obtain the group reference time value of the clock devices. A2: Based on the group reference time value, the deviation of the instantaneous time value is calculated to obtain the single-period outlier deviation of the clock device, and the zero-crossing rate of the single-period outlier deviation is statistically analyzed to evaluate the stability of the clock device. A3: Fit the voltage fluctuation slope of the power supply voltage sample value of the watch device to obtain the voltage decay characteristic value of the watch device; A4: Based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, the stability assessment result and the voltage decay characteristic value are cross-mapped with fault instructions, and the obtained maintenance instruction string is stored as an IoT management instruction in the IoT device shadow of the clock device; A5: Based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the device in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain the task execution sequence for the IoT terminal devices of maintenance personnel.

2. The watch management method based on the Internet of Things as described in claim 1, characterized in that, Within the same time zone, the median benchmark is extracted from the instantaneous timekeeping values ​​reported by the clock devices via the IoT aggregation node to obtain the group reference time value of the clock devices, including: Within the same time zone, the raw data frames received by the IoT aggregation node from the clock device are completely decapsulated to obtain the instantaneous timekeeping value of the clock device; The instantaneous time values ​​are sorted in ascending order to obtain an ordered sequence of instantaneous time values; The middle time value of the ordered time value sequence is used as the group reference time value. If the number of elements in the ordered time value sequence is even, the average of the two middle time values ​​is used as the group reference time value.

3. The watch management method based on the Internet of Things as described in claim 1, characterized in that, The step of calculating the deviation of the instantaneous timekeeping value based on the group reference time value to obtain the single-period outlier deviation of the clock device includes: Based on the group reference time value, the difference between the instantaneous travel time value and the reference amplitude sequence of the instantaneous travel time value is obtained; The extreme value elimination filter is applied to the deviation from the reference amplitude sequence to obtain the set of transient fluctuation amplitudes of the deviation from the reference amplitude sequence; Perform sign zero-crossing identification on the transient fluctuation amplitude set to obtain the positive and negative transition identifier queue of the transient fluctuation amplitude set; Periodic boundary cutting is performed on the positive and negative transition marker queue to obtain the deviation amplitude grouping queue of the positive and negative transition marker queue; Outlier amplitude is extracted from the deviation amplitude grouping queue to obtain the single-cycle outlier deviation of the clock device.

4. The watch management method based on the Internet of Things as described in claim 3, characterized in that, The formula for calculating the single-period outlier bias is as follows: ； in, This refers to the single-period outlier deviation. The total number of groups in the deviation amplitude grouping queue. The first in the deviation magnitude grouping queue Instantaneous travel values ​​within each group, This is the reference time value for the group. The standard deviation of the instantaneous timekeeping value of the clock device under healthy conditions. This is the preset outlier sensitivity threshold.

5. A clock management method based on the Internet of Things as described in claim 1, characterized in that, The method of performing zero-crossing rate statistics on the single-period outlier deviation to evaluate the stability of the watchmaking device further includes: The deviation time series of the clock device is constructed by arranging the single-period outlier deviations in chronological order. The time series of the deviation is symbolized to obtain the deviation symbol sequence of the time series of the deviation; The total number of symbol changes in the deviation symbol sequence is obtained by summing the number of adjacent symbol transitions. The limit transition capacity of the deviation sign sequence is obtained by decreasing the sequence length value by a unit value. The ratio of the total number of symbol changes to the limiting transition capacity is quantified to obtain the stability evaluation result of the clock device.

6. The watch management method based on the Internet of Things as described in claim 1, characterized in that, The step of fitting the voltage fluctuation slope of the power supply voltage sample values ​​of the watch device to obtain the voltage decay characteristic values ​​of the watch device includes: The power supply voltage sample value of the clock device is divided into equal time windows to obtain the voltage segment subsequence and time window width of the power supply voltage sample value; The voltage change amplitude value of the voltage segment subsequence is obtained by calculating the amplitude difference at the endpoints of the voltage segment subsequence; Based on the time window width, the voltage change amplitude value is normalized to obtain the voltage fluctuation slope value of the voltage segmented subsequence; By performing trend fitting on the voltage fluctuation slope value, the voltage decay characteristic value of the clock device is obtained.

7. The watch management method based on the Internet of Things as described in claim 5, characterized in that, The system, based on a preset fault discrimination threshold group and maintenance knowledge records for the same time zone, performs fault command cross-mapping on the stability assessment results and the voltage degradation characteristic values, and stores the resulting maintenance command string as an IoT management command in the IoT device shadow of the watch device, including: Based on a preset fault discrimination threshold group, the stability evaluation results are compared with the stability level threshold points to obtain the stability level identifier of the watch device. The voltage degradation characteristic value is mapped to the fault discrimination threshold group to obtain the degradation level identifier of the watch device; Based on the combined key value composed of the stability level identifier and the degradation level identifier, the fault code mapping entries of the maintenance knowledge record in the same time zone are indexed and matched to obtain the maintenance instruction string of the clock device. Based on the maintenance instruction string, the maintenance instruction field in the IoT device shadow of the clock device is overwritten with the maintenance instruction string to obtain the IoT management instruction of the IoT device shadow.

8. A clock management method based on the Internet of Things as described in claim 7, characterized in that, The combined key value composed of the stability level identifier and the degradation level identifier is used to perform index matching retrieval of fault code mapping entries in the maintenance knowledge records of the same time zone to obtain the maintenance instruction string of the watch equipment, including: The stability level identifier and the degradation level identifier are encoded by binary bit concatenation to obtain the combined key value byte stream of the watch device; Based on the combined key-value byte stream, hash slot positioning is performed on the fault code mapping entries of the maintenance knowledge records in the same time zone to obtain the matching address offset of the clock device. Based on the matching address offset, extract the original fault data record from the maintenance knowledge record; The original fault data record is parsed to obtain the maintenance instruction string for the clock device.

9. A clock management method based on the Internet of Things as described in claim 1, characterized in that, Based on the stability assessment results and the installation coordinates of the clock device, and combined with the maintenance urgency of the devices in the maintenance instruction string, the devices to be processed in the clock device are grouped and sorted by proximity to obtain a task execution sequence for the IoT terminal devices of maintenance personnel, including: Based on the urgency of the maintenance of the clock devices in the maintenance instruction string, the field values ​​of the clock devices are filtered to obtain a set of clock devices to be processed; Based on the installation coordinates of the clock device, the devices to be processed in the set of devices to be processed are spatially clustered to obtain the spatial cluster groups of the devices to be processed. The spatial clusters are sorted in descending order according to the stability assessment results to obtain the maintenance priority queue of the equipment. The maintenance priority queue is joined end-to-end to obtain a task execution sequence for IoT terminal devices for maintenance personnel.

10. A clock management system based on the Internet of Things, characterized in that, The system is used to implement the IoT-based watch management method according to claim 1, the system comprising: The median reference extraction module is used to extract the median reference from the instantaneous timekeeping values ​​reported by the clock devices via the Internet of Things aggregation node within the same time zone, so as to obtain the group reference time value of the clock devices. The deviation zero-crossing statistics module is used to calculate the deviation of the instantaneous timekeeping value based on the group reference time value, obtain the single-cycle outlier deviation of the clock device, and perform zero-crossing rate statistics on the single-cycle outlier deviation to evaluate the stability of the clock device. The voltage decay fitting module performs voltage fluctuation slope fitting on the power supply voltage sampling value of the watch device to obtain the voltage decay characteristic value of the watch device. The fault instruction mapping module is used to perform fault instruction cross-mapping on the stability assessment results and the voltage decay characteristic value based on the preset fault discrimination threshold group and the maintenance knowledge record of the same time zone, and store the obtained maintenance instruction string as an IoT management instruction in the IoT device shadow of the clock device. The grouping and sorting maintenance module is used to sort the devices to be processed in the clock device according to the stability assessment results and the installation coordinates of the clock device, combined with the maintenance urgency of the devices in the maintenance instruction string, to obtain a task execution sequence for the IoT terminal devices of maintenance personnel.

Images