Antenna life deduction method and system based on grey prediction

Abstract

The present application relates to the technical field of data processing, in particular to an antenna life deduction method and system based on grey prediction, comprising: obtaining operation data in each period before and after antenna maintenance; obtaining the load back-off coefficient of the antenna according to the operation data in each period, and then obtaining the maintenance recovery coefficient of the antenna, the anti-load repair coefficient of maintenance and the anti-degradation coefficient of maintenance; determining the maintenance state of the antenna according to the maintenance recovery coefficient of the antenna, the anti-load repair coefficient of maintenance and the anti-degradation coefficient of maintenance, in combination with the operation data of the verification task load period and the historical task load period; and predicting the antenna life according to the maintenance state of the antenna. The present application improves the accuracy of antenna life prediction by analyzing the influence of antenna operation load change on the health state, the actual health recovery degree after maintenance and the persistence under the task load.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to an antenna lifetime extrapolation method and system based on grey prediction. Background Technology

[0002] Antennas, as crucial devices in wireless communication, satellite communication, and radar detection systems, continuously perform different program tracking tasks through actuators such as azimuth and elevation axes. During long-term operation, the actuators are subjected to continuous motion loads, causing the mechanical transmission components, drive components, and related structural performance to gradually degrade. This leads to changes in the antenna's health status over time. When an abnormal antenna health status or performance degradation is detected, antenna repair is usually necessary to restore its operational performance. After repair, the antenna still needs to undertake program tracking tasks under different loads. Therefore, it is necessary to evaluate the repair effect based on the operational status before and after repair and further predict the remaining service life of the antenna after repair.

[0003] Existing antenna lifetime prediction methods typically establish health status degradation trends based on historical antenna operating data and predict the antenna's remaining lifetime based on these trends. However, changes in antenna health status after maintenance are not only affected by the maintenance effect but also by the differences in motion load from program tracking tasks before and after maintenance. When evaluating maintenance effectiveness, existing methods usually directly compare the health status before and after maintenance without fully considering the impact of changes in motion load such as azimuth and elevation axis operating speeds and cumulative rotation angles on the health status. This can easily lead to misjudging changes in health status caused by differences in motion load as maintenance recovery effects. Furthermore, there is a lack of effective differentiation between different scenarios such as short-term recovery, continuous recovery, and re-degradation after recovery of post-maintenance health status, making it difficult to accurately determine the post-maintenance health degradation trend and thus affecting the accuracy of antenna lifetime prediction based on maintenance status. Summary of the Invention

[0004] This invention provides an antenna lifetime prediction method and system based on grey prediction to solve the existing problem that existing antenna lifetime prediction methods are difficult to accurately predict antenna lifetime.

[0005] The antenna lifetime extrapolation method and system based on grey prediction of the present invention adopts the following technical solution: One embodiment of the present invention provides an antenna lifetime extrapolation method based on grey prediction, the method comprising the following steps: Obtain operational data for each time period before and after antenna maintenance, including: historical low load period, historical task load period, verification low load period, and verification task load period; The load reduction coefficient of the antenna is obtained based on the operating data of the verification task load period, historical low load period and historical task load period; Based on the operational data of the verification low load period, the verification task load period, and the historical task load period, as well as the antenna load drop coefficient, the antenna maintenance recovery coefficient is obtained. Combining the operational data of the verification task load period and the historical task load period, the maintenance load resistance repair coefficient is obtained. The maintenance resilience coefficient is obtained based on the operational data of the verification task load period and the historical task load period. The antenna's maintenance status is determined by combining the antenna's maintenance recovery coefficient, maintenance load repair coefficient, and maintenance decay resistance coefficient with the operational data from the verification task load period and historical task load periods. Predict antenna lifespan based on antenna maintenance status.

[0006] Preferably, the specific method for obtaining the operational data for each time period before and after antenna maintenance includes: The antenna's azimuth axis speed, pitch axis speed, and operating mode are acquired at each sampling time. The operating modes include: standby mode, manual low-speed operation mode, and program tracking mode. The preset analysis range is divided into time periods based on a complete health assessment cycle of the antenna health management system, and the health value of the antenna in each time period is obtained. For any given time period, the average azimuth axis operating speed, average pitch axis operating speed, cumulative azimuth axis rotation angle, and cumulative pitch axis rotation angle within that time period are taken as the antenna's operating data for that time period. If the operating mode is standby mode or manual low-speed operation mode at all sampling times within the time period, then the time period is recorded as a low-load time period; if the operating mode is program tracking mode at all sampling times within the time period, then the time period is recorded as a task-load time period. The low-load period and the high-load period that occur before the maintenance completion time are recorded as the historical low-load period and the historical high-load period, respectively. The preset observation time window after the maintenance completion time is used as the observation window, and the task load period within the observation window is recorded as the verification task load period. The low-load period that is after the maintenance completion time and before the first verification task load period is denoted as the verification low-load period.

[0007] Preferably, the method for obtaining the antenna load reduction coefficient based on the operational data of the verification task load period, historical low load period, and historical task load period includes: For any historical task load period, if the previous period of the historical task load period is not a historical task load period, the historical task load period is recorded as a candidate matching period. Based on the running data of the first verification task load period and all candidate matching periods, the nearest neighbor matching algorithm is used to match the first verification task load period with all candidate matching periods to obtain the matching period of the first verification task load period. For any matching period, the historical low-load period that is closest to and precedes the matching period is recorded as the source period of the matching period; the difference between the health value of the source period and the health value of the matching period is used as the load reduction factor of the matching period. The average load reduction factor of all matching periods during the first verification task load period is used as the antenna load reduction coefficient.

[0008] Preferably, the method for obtaining the antenna maintenance recovery coefficient based on the operating data of the verification low-load period, the verification task load period, and the historical task load period, as well as the antenna load reduction coefficient, includes the following specific methods: For any verification task load period, based on the running data of the verification task load period and all historical task load periods, the nearest neighbor matching algorithm is used to match the verification task load period with all historical task load periods, and the historical task load period with the closest matching distance is used as the benchmark matching period for the verification task load period. The difference between the mean of all health values ​​during low-load verification periods and the antenna's load reduction coefficient is used as the antenna's estimated task-state health value. The difference between the antenna's estimated task-state health value and the health value during the baseline matching period of the first verification task load period is used as the antenna's maintenance recovery coefficient.

[0009] Preferably, the specific method for obtaining the repair load-bearing coefficient is as follows: The difference between the health value of the first verification task load period and the health value of its baseline matching period is recorded as the health repair amount of the first verification task load period. The ratio of the health repair amount during the first verification task load period to the antenna's maintenance recovery coefficient is used as the maintenance load resistance repair coefficient.

[0010] Preferably, the specific method for obtaining the maintenance resilience coefficient based on the operational data of the verification task load period and the historical task load period includes: The minimum health repair amount among all verification task load periods is divided by the health repair amount of the first verification task load period, and the ratio is used as the maintenance degradation resistance coefficient.

[0011] Preferably, the method for determining the antenna's maintenance status based on the antenna's maintenance recovery coefficient, maintenance load repair coefficient, and maintenance anti-fading coefficient, combined with the operating data of the verification task load period and historical task load periods, includes the following specific method: based on the operating data of the first verification task load period and all candidate matching periods, the first verification task load period is matched with all candidate matching periods using the nearest neighbor matching algorithm to obtain the matching period of the first verification task load period; If the antenna's repair recovery coefficient is greater than the preset fluctuation threshold, the repair load resistance repair coefficient is greater than the preset load resistance repair threshold, the repair anti-fading coefficient is greater than the preset anti-fading threshold, and the health repair amount during each verification task load period is greater than the preset fluctuation threshold, the antenna's repair status is recorded as a stable repair state. If the antenna's repair recovery coefficient is less than or equal to the preset fluctuation threshold, or if there is a verification task load period where the amount of health repair is less than or equal to the preset fluctuation threshold, the antenna's repair status will be recorded as a short-term recovery state. If the antenna's maintenance status is not in a stable repair state or a short-term recovery state, the antenna's maintenance status will be recorded as an observation state.

[0012] Preferably, the method for predicting antenna lifespan based on antenna maintenance status includes: If the antenna is in a stable repair state, the sequence of health values ​​for all verification task load periods is used as a new gray prediction sequence. Gray prediction is then performed based on the new gray prediction sequence to predict the antenna lifetime. If the antenna's maintenance status is in a short-term recovery state, the health values ​​of all verification task load periods are spliced ​​into the original gray prediction sequence before maintenance, the gray prediction sequence is updated, and gray prediction is performed based on the updated gray prediction sequence to predict the antenna's lifespan. If the antenna is under observation during maintenance, the antenna lifespan is predicted based on the original gray prediction sequence before maintenance.

[0013] Preferably, if the antenna is under maintenance and in an observation state, the antenna lifespan is predicted based on the original gray prediction sequence before maintenance. The specific methods include: Grey prediction is performed based on the original grey prediction sequence before maintenance to obtain the remaining mission cycles of the antenna. The remaining task duration of the antenna is the product of the remaining task cycle number of the antenna and the corresponding duration of the task load period. The average daily load duration is recorded as the average duration of the antenna being in program tracking mode within the preset statistical range. The antenna's lifespan is the ratio of its remaining operational time to its average daily load time.

[0014] Another embodiment of the present invention provides an antenna lifetime extrapolation system based on grey prediction, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any of the above-described antenna lifetime extrapolation methods based on grey prediction.

[0015] The beneficial effects of the technical solution of this invention are as follows: This invention acquires operational data from historical low-load periods, historical task load periods, verification low-load periods, and verification task load periods before and after antenna maintenance, and obtains a load reduction coefficient based on the operational data from the verification task load period, historical low-load periods, and historical task load periods. This characterizes the change in the health status of the antenna caused by the motion load during the transition from a low-load state to a task load state, providing a load impact benchmark for subsequent maintenance recovery effect evaluation. Simultaneously, based on the operational data from the verification low-load period, verification task load period, and historical task load periods, and combined with the load reduction coefficient, a maintenance recovery coefficient is obtained. Furthermore, based on the same load operational data before and after maintenance, load resistance repair coefficient and anti-fading coefficient are obtained. This provides a basis for evaluating the maintenance recovery degree, the degree of recovery fulfillment under task load, and the maintenance recovery effect. The effectiveness of antenna repair is evaluated based on factors such as the degree of maintenance, thereby reducing the interference of different motion loads on the judgment of repair effectiveness and improving the accuracy of antenna repair status determination. Furthermore, the repair status of the antenna is determined based on the repair recovery coefficient, load resistance repair coefficient, anti-fading coefficient, and changes in health status before and after repair. Based on different repair statuses, a prediction method for antenna health degradation trend is determined, so that the life prediction can be adapted to the health status change process under different repair effects, thereby improving the accuracy of antenna life prediction after repair. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the steps of the antenna lifetime extrapolation method based on grey prediction of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the antenna lifetime estimation method and system based on gray prediction proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The specific scheme of the antenna lifetime extrapolation method and system based on gray prediction provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Please see Figure 1 The diagram illustrates a flowchart of an antenna lifetime extrapolation method based on gray prediction provided by an embodiment of the present invention. The method includes the following steps: Step S001: Obtain operational data for each time period before and after antenna maintenance.

[0022] It should be noted that during long-term operation, the antenna's actuators are subjected to continuous motion loads, causing the mechanical transmission components, drive components, and related structural performance to gradually degrade. This results in changes in the antenna's health status over time. After experiencing low-load operating states such as standby and manual adjustment, the antenna re-enters normal program tracking tasks. If the changes in health status after maintenance are directly considered as maintenance effects, the temporary health improvement caused by the switching of operating modes may be mistaken for a true recovery of equipment performance, thus reducing the accuracy of antenna lifespan prediction. Therefore, it is necessary to correlate and classify the health status data before and after maintenance with their corresponding operating modes and loads. Thus, this step divides time periods based on the antenna health management system's evaluation cycle. Each time period is marked as a low-load period or a task load period according to its operating mode. The maintenance completion time serves as the boundary, and each time period is assigned to either the historical stage or the verification stage. This provides a structured data foundation for subsequently distinguishing between the "apparent advantages brought by low load" and the "real improvements brought by maintenance."

[0023] Specifically, the azimuth axis running speed, pitch axis running speed, and operating mode of the antenna are obtained at each sampling time. The operating modes include: standby mode, manual low-speed operation mode, and program tracking mode.

[0024] The antenna health management system uses a complete health assessment cycle as the basic unit to divide the preset analysis range into time periods and obtain the antenna's health value in each time period. The specific values ​​of the analysis range can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, the analysis range is described as 12 months before maintenance to 1 month after maintenance.

[0025] It should be noted that a complete health assessment cycle refers to the time period corresponding to one complete health assessment of the antenna's health status by the antenna health management system. Specifically, the antenna health management system continuously monitors and assesses the antenna's operating status according to a preset health assessment cycle. Upon completion of each complete health assessment cycle, it outputs the corresponding health value for that cycle. One complete health assessment cycle corresponds to a time period; the operating data within this time period is the operating data for that complete health assessment cycle, and the health value for that time period is the health value corresponding to the complete health assessment cycle. The specific duration of the health assessment cycle can be preset by the antenna health management system according to actual monitoring and assessment requirements. This embodiment does not impose a rigid requirement; in this embodiment, a complete health assessment cycle of 24 hours is used as an example.

[0026] Furthermore, for any given time period, the average azimuth axis operating speed, average pitch axis operating speed, cumulative azimuth axis rotation angle, and cumulative pitch axis rotation angle within that time period are taken as the antenna's operating data for that time period.

[0027] If the operating mode at all sampling times within the time period is standby mode or manual low-speed operation mode, then the time period is recorded as a low-load time period; if the operating mode at all sampling times within the time period is program tracking mode, then the time period is recorded as a task-load time period.

[0028] The low-load period and the high-load period that occur before the maintenance is completed are respectively recorded as the historical low-load period and the historical high-load period.

[0029] The preset observation time window after the maintenance completion time is used as the observation window. The task load period within the observation window is recorded as the verification task load period. The specific value of the observation time window can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, the observation time window is 15 days as an example.

[0030] The low-load period that is after the maintenance completion time and before the first verification task load period is denoted as the verification low-load period.

[0031] It should be noted that the antenna health management system is used to continuously monitor the operating status of the antenna during operation and to conduct health assessments of the antenna's operating status according to the health assessment cycle. The health value is used to characterize the comprehensive health status of the antenna within the corresponding health assessment cycle. It can be obtained by the antenna health management system based on the status parameters, performance parameters, or abnormal status parameters collected during the antenna operation. The specific method of obtaining the health value can be determined according to the actual health assessment mechanism, and this embodiment does not impose specific limitations.

[0032] It should be further explained that the azimuth axis running speed and the pitch axis running speed reflect the speed of the antenna's azimuth axis and pitch axis movement at the corresponding moment, respectively, and can characterize the intensity of the motion load borne by the antenna actuator at the corresponding moment; the cumulative rotation angle of the azimuth axis and the cumulative rotation angle of the pitch axis reflect the actual amount of motion completed by the corresponding rotation axis within a health assessment cycle. The running speed reflects the instantaneous motion load intensity, and the cumulative rotation angle reflects the total amount of motion continuously borne during the period. The two together describe the motion load characteristics of the period. Because the antenna movement is low in standby mode and manual low-speed operation mode, while the antenna needs to continuously perform azimuth and elevation movements according to the tracking task in program tracking mode, the corresponding actuator motion loads are significantly different. Therefore, the health assessment period is divided into low-load and task load periods by working mode, so that different health values ​​can be correlated with the motion load conditions at the time of their formation. Furthermore, the historical stage and the verification stage are divided with the maintenance completion time as the boundary, and the normal program tracking period within the post-maintenance observation window is used as the verification task load period, and the low-load period before the first occurrence of task load is used as the verification low-load period. This provides a basis for establishing a comparative benchmark for the subsequent use of the health status under the normal task load before maintenance, and for distinguishing between health changes caused by maintenance and health changes caused by changes in operating load.

[0033] Step S002: Obtain the antenna load reduction coefficient based on the operating data of the verification task load period, historical low load period, and historical task load period.

[0034] It should be noted that when an antenna transitions from a low-load state to normal tracking mode, its health status typically exhibits a natural decline due to the increased operating load. This decline is not caused by maintenance actions but is an inherent characteristic of equipment operating status changes. Directly using the change in health status from low load to task load after maintenance as the evaluation criterion for maintenance effectiveness could confuse the natural decline with the actual change in maintenance results. Therefore, this step utilizes historical operating data from before maintenance, using the motion characteristics of the first task load switch after this maintenance as a matching template to find historical low-load to task load switch processes with similar operating conditions. The typical decline in health status during such switch processes is calculated as a load recovery coefficient. This coefficient is used to subtract the natural impact of load recovery from the observed health changes after maintenance, providing a benchmark for accurately identifying the true health improvement brought about by maintenance.

[0035] Specifically, for any historical task load period, if the previous period of the historical task load period is not a historical task load period, the historical task load period is recorded as a candidate matching period.

[0036] Based on the running data of the first verification task load period and all candidate matching periods, the first verification task load period is matched with all candidate matching periods using the nearest neighbor matching algorithm to obtain the matching period of the first verification task load period. Since the nearest neighbor matching algorithm is a well-known prior art, it will not be described in detail in this embodiment.

[0037] For any matching period, the historical low-load period that is closest to and precedes the matching period is recorded as the source period of the matching period; the difference between the health value of the source period and the health value of the matching period is used as the load reduction factor of the matching period.

[0038] The average load reduction factor of all matching periods during the first verification task load period is used as the antenna load reduction coefficient.

[0039] It should be noted that the purpose of this step is to obtain the degree of natural decline in health status caused by the increase in motion load when the antenna transitions from a low-load state to a task load state without maintenance, based on historical operational data prior to maintenance. Using historical task load periods that were not previously part of the historical task load period as candidate matching periods ensures that the selected historical task load period corresponds to a complete load switching process from low load to task load, avoiding the use of intermediate periods in continuous task load processes as the start of load switching, which would cause health status changes to be simultaneously affected by the previous task load state. Further, nearest neighbor matching is performed using operational data such as azimuth axis average operating speed, pitch axis average operating speed, azimuth axis cumulative rotation angle, and pitch axis cumulative rotation angle, ensuring that the candidate matching period has motion load conditions as close as possible to the first verification task load period. Then, the historical low-load period preceding and closest to the matching period is used as the source period, and the health values ​​of the source period and the matching period are compared to obtain the degree of health status decline corresponding to a load transition from low to high. Since a single historical load switching process may be affected by factors such as fluctuations in the equipment's own condition, the average load reduction factor of multiple matching periods is taken to reduce the impact of single operation fluctuations on the estimation of the load reduction degree, so as to obtain a load reduction coefficient that can represent the degree of health state reduction caused by normal antenna load loading.

[0040] Step S003: Based on the operating data of the verification low load period, the verification task load period, the historical task load period, and the antenna load reduction coefficient, obtain the antenna maintenance recovery coefficient. Combine the operating data of the verification task load period and the historical task load period to obtain the maintenance load resistance repair coefficient.

[0041] It should be noted that when the antenna is under low load after maintenance, its health value is usually higher than that under normal workload. This difference is partly due to the apparent advantage brought about by low load operation, rather than the actual performance improvement produced by the maintenance. To identify the true effect of the maintenance, it is necessary to first use the load reduction coefficient to subtract the natural influence caused by the difference in load conditions from the low load health value after maintenance, to obtain the theoretically achievable health level under normal workload. Then, this should be compared with the health value under the same load condition before maintenance to obtain the maintenance recovery coefficient, which is used to measure the actual health improvement formed by maintenance after deducting the load influence. Furthermore, the maintenance recovery coefficient is a theoretical calculation result based on historical patterns. Whether this improvement can be truly realized when the antenna actually undertakes its first program tracking task still needs to be verified through actual operating data. Therefore, the health improvement measured in the first actual task is compared with the maintenance recovery coefficient to obtain the load resistance repair coefficient, which is used to measure the degree to which the theoretically calculated maintenance effect is realized under actual workload.

[0042] Specifically, for any verification task load period, based on the running data of the verification task load period and all historical task load periods, the nearest neighbor matching algorithm is used to match the verification task load period with all historical task load periods, and the historical task load period with the closest matching distance is used as the benchmark matching period for the verification task load period.

[0043] The difference between the mean of all health values ​​during low-load verification periods and the antenna's load reduction coefficient is used as the antenna's estimated task-state health value. The difference between the antenna's estimated task-state health value and the health value during the baseline matching period of the first verification task load period is used as the antenna's maintenance recovery coefficient.

[0044] It should be noted that finding the benchmark matching period with the closest operating load in the historical task load period is to ensure that the comparison of health status before and after maintenance is based on similar operating load conditions, reducing comparison bias caused by differences in task load before and after maintenance. Since the antenna usually experiences low-load operation before entering the program tracking task after maintenance, verifying the health value during the low-load period can reflect the health status before the normal task load is reloaded after maintenance. Subtracting the load reduction coefficient from the average health value is equivalent to estimating the health level that the antenna should reach when it resumes task load after maintenance based on historical normal load loading patterns, thus obtaining the estimated task-state health value. Further comparing the estimated task-state health value with the benchmark health value under similar task load conditions before maintenance can distinguish the health status difference caused by low-load operation from the health improvement caused by maintenance, thereby obtaining the maintenance recovery coefficient that can characterize the actual recovery degree of this maintenance. That is, after deducting the apparent health status advantage caused by low-load operation after maintenance, it estimates the actual improvement degree of antenna health status caused by maintenance.

[0045] Specifically, the difference between the health value of the first verification task load period and the health value of its baseline matching period is recorded as the health repair amount for the first verification task load period.

[0046] The ratio of the health repair amount during the first verification task load period to the antenna's maintenance recovery coefficient is used as the maintenance load resistance repair coefficient.

[0047] It should be noted that the repair recovery coefficient reflects the degree of repair recovery based on low load conditions and historical load variation patterns. However, this degree of recovery has not been verified by the actual task load borne by the antenna after repair. Therefore, it is necessary to further determine whether the health improvement formed by the repair can be reflected under the actual task load. Therefore, this embodiment compares the health value of the first verified task load period with the health value of its reference matching period. This directly yields the degree of health improvement of the antenna under actual task load conditions relative to the same task load conditions before repair. This degree of improvement is used as the health repair amount. Furthermore, comparing the ratio of the health repair amount under the actual task load to the repair recovery coefficient reflects the proportion of the health improvement realized under the actual task load relative to the estimated repair recovery degree. When this ratio is high, it indicates that the health improvement formed by the repair can withstand the normal task load well. When this ratio is low, it indicates that the health improvement after repair has not been fully realized under the actual task load. Therefore, this ratio is used as the load resistance repair coefficient to characterize the ability of the repair recovery effect to withstand the task load.

[0048] Step S004: Obtain the maintenance resistance coefficient based on the operating data of the verification task load period and the historical task load period.

[0049] It should be noted that while some repair or maintenance operations may perform well initially after resumption of operation, existing abnormal conditions may gradually reappear as normal tasks are continuously performed, leading to a decline in the effectiveness of the repairs. Therefore, the repair effectiveness verified during the initial program tracking task cannot be directly considered as having reached a stable new health stage. Thus, it is necessary to continuously acquire data during subsequent task load periods within the pre-set maintenance acceptance observation period to observe the changing trend of the repair effectiveness. This step takes the minimum health improvement across all task load periods within the observation period and compares it with the health improvement during the initial task load period to obtain the anti-fading coefficient. This coefficient measures the extent to which the initially realized repair effectiveness can be retained under continuous normal operating conditions, thus providing a basis for judging the sustainability of the repair effectiveness.

[0050] Specifically, the minimum health repair amount among all verification task load periods is obtained, and the ratio is obtained by dividing it by the health repair amount of the first verification task load period. This ratio is used as the maintenance anti-degradation coefficient.

[0051] Step S005: Based on the antenna's repair recovery coefficient, repair load resistance coefficient, and repair degradation resistance coefficient, and combined with the operating data of the verification task load period and the historical task load period, determine the antenna's repair status.

[0052] It should be noted that the assessment of antenna repair effectiveness requires a comprehensive examination from three dimensions: First, after deducting the apparent impact of differences in low-load conditions, does the repair still result in a health improvement exceeding the equipment's natural fluctuation range? Second, can this improvement be effectively realized under the initial real-world workload? Third, can the realized improvement be maintained under continuous workloads? Therefore, this step compares the repair recovery coefficient, load-resistance repair coefficient, and anti-fading coefficient with their corresponding preset thresholds, while requiring that the amount of health improvement in each verification workload period exceeds the equipment's natural fluctuation range. Based on this, the repair status is determined as a stable repair state, a short-term recovery state, or a state under observation. A stable repair state indicates that the repair has resulted in a real health improvement that can be maintained sustainably; a short-term recovery state indicates that although the health status has improved to some extent after the repair, this improvement fails to meet the recovery, realization, or maintenance requirements of a stable repair state; a state under observation indicates that the current data is insufficient to determine the repair effectiveness, and its status category is temporarily uncertain.

[0053] Specifically, if the antenna's repair recovery coefficient is greater than the preset fluctuation threshold, the repair load resistance repair coefficient is greater than the preset load resistance repair threshold, the repair anti-fading coefficient is greater than the preset anti-fading threshold, and the health repair amount during each verification task load period is greater than the preset fluctuation threshold, the antenna's repair status is recorded as a stable repair state.

[0054] If the antenna's maintenance recovery coefficient is less than or equal to the preset fluctuation threshold, or if there is a verification task load period where the amount of health repair is less than or equal to the preset fluctuation threshold, the antenna's maintenance status will be recorded as a short-term recovery state.

[0055] If the antenna's maintenance status is not in a stable repair state or a short-term recovery state, the antenna's maintenance status will be recorded as an observation state.

[0056] It should be noted that the specific values ​​of the fluctuation threshold, the load repair threshold, and the anti-decay threshold can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, the process of obtaining the fluctuation threshold is as follows: calculate the standard deviation of the health values ​​of all historical task load periods, and use the standard deviation as the fluctuation threshold to characterize the normal fluctuation range of the health status of the equipment itself when there is no maintenance behavior.

[0057] The process of obtaining the load resistance repair threshold is as follows: retrieve the load resistance repair coefficients of each maintenance of the same type of antenna from the database, and take the minimum value among all the load resistance repair coefficients of the same type of antenna as the load resistance repair threshold.

[0058] The process of obtaining the anti-fading threshold is as follows: retrieve the anti-fading coefficient of each maintenance of the same type of antenna from the database, and take the minimum value among all maintenance anti-fading coefficients of the same type of antenna as the anti-fading threshold.

[0059] It should be further clarified that this embodiment does not judge the repair effect solely based on whether the health value increases after repair. Instead, it comprehensively judges the degree of health improvement after repair, the degree of recovery realization under actual task load, and the degree of recovery maintenance under subsequent task load. A repair recovery coefficient greater than a preset fluctuation threshold indicates that the health status after repair still has an improvement beyond the normal health fluctuation range after deducting the impact of load changes; a load resistance repair coefficient greater than a preset load resistance repair threshold indicates that the repair improvement can be fully realized under actual task load; a decay resistance coefficient greater than a preset decay resistance threshold indicates that the realized repair improvement can be maintained under subsequent task load; at the same time, it is required that the health repair amount in each verification task load period is greater than the preset fluctuation threshold to avoid determining that the repair has formed a stable recovery based solely on an abnormal increase in the health status in a certain period. When all the above conditions are met, it indicates that the health improvement formed by the repair has been verified by actual task load and has sustainability, and is therefore judged as a stable recovery state. If the degree of repair recovery does not exceed the normal health fluctuation range, or if the amount of health repair during the verification task load period does not exceed the normal health fluctuation range, it indicates that the health improvement after repair is insufficient to support the judgment of stable recovery, and it is judged as a short-term recovery state. If the situation neither reaches a stable repair state nor meets the judgment conditions of a short-term recovery state, it is retained as an observation state to avoid making premature judgments on the antenna repair effect before the repair effect has been fully verified.

[0060] Step S006: Predict the antenna lifespan based on the antenna's maintenance status.

[0061] It's important to note that different maintenance status assessments correspond to different lifetime prediction sequence construction methods. If maintenance is determined to be in a stable repair state, it indicates that the equipment has reached a new, sustainable health state under normal workload. Pre-maintenance health degradation data no longer accurately reflects the post-maintenance degradation trend. In this case, a new grey prediction sequence should be constructed using data from the verification workload period to re-establish the health degradation trend. If maintenance is determined to be in a short-term recovery state, it indicates that the observed health recovery after maintenance has not formed a sustainable new state under normal workload. In this case, the original degradation process should not be interrupted due to the short-term recovery; instead, data from the verification workload period should be integrated into the original pre-maintenance grey prediction sequence to continue the original degradation trend. If maintenance is in an observation state, it indicates that the current data is insufficient to determine the maintenance effect. In this case, the original pre-maintenance grey prediction sequence should continue to be used for lifetime estimation, and unverified post-maintenance data should not be used to update or reconstruct the sequence. Through these methods, it is ensured that lifetime prediction is always based on the actual health degradation trend verified by workload.

[0062] Specifically, if the antenna is in a stable repair state, the sequence of health values ​​for all verification task load periods is used as a new gray prediction sequence. Gray prediction is then performed based on the new gray prediction sequence to predict the antenna lifetime. Since gray prediction is a well-known existing technology, it will not be described in detail in this embodiment.

[0063] If the antenna's maintenance status is in a short-term recovery state, the health values ​​of all verification task load periods are spliced ​​into the original gray prediction sequence before maintenance, the gray prediction sequence is updated, and gray prediction is performed based on the updated gray prediction sequence to predict the antenna's lifespan.

[0064] If the antenna is under maintenance and in observation status, perform gray prediction based on the original gray prediction sequence before maintenance to obtain the remaining number of mission cycles for the antenna.

[0065] The remaining task duration of the antenna is the product of the remaining task cycle number of the antenna and the corresponding duration of the task load period.

[0066] The average daily duration of the antenna in program tracking mode within the preset statistical range is recorded as the average daily load duration. The specific value of the preset statistical range can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, the 6 months prior to the maintenance are used as the statistical range.

[0067] The antenna's lifespan is the ratio of its remaining operational time to its average daily load time.

[0068] It should be noted that different maintenance states indicate different degrees of correlation between the post-maintenance health degradation process and the original pre-maintenance degradation process. Therefore, identical health data sequences cannot be used for lifetime prediction. When the maintenance state is a stable recovery state, it indicates that the post-maintenance health improvement has been verified under normal workloads and can be maintained. The pre-maintenance health degradation trend can no longer directly represent the current post-maintenance health state. Therefore, a new grey prediction sequence is formed using the health values ​​of each verified workload period after maintenance, so that lifetime prediction starts from the new health state verified by the workload. When the maintenance state is a short-term recovery state, it indicates that although the health state has recovered after maintenance, the recovery is not sufficient to form a stable new health degradation stage. Therefore, the original degradation trend is not severed by the post-maintenance health recovery. Instead, the health values ​​of the verified workload period are integrated into the original pre-maintenance grey prediction sequence, so that the actual operating data after maintenance can continue to participate in the updating of the original degradation trend. When the maintenance status is under observation, it indicates that the current maintenance effect has not been fully verified. If the post-maintenance data is used to rebuild the prediction sequence, the unconfirmed health recovery may have too much impact on the life prediction. Therefore, life prediction is continued based on the original gray prediction sequence before maintenance, so that the health degradation trend on which the life prediction is based matches the certainty of the current maintenance effect.

[0069] It needs further clarification that the remaining task cycles obtained from the grey prediction represent the number of cycles the antenna can continue to undertake normal program tracking tasks under the current health degradation trend. However, the actual task duration corresponding to a single health assessment cycle may not be directly used as the antenna's remaining usage time. Therefore, it is necessary to further convert the remaining task cycles into a remaining task duration with actual time meaning. By multiplying the remaining task cycles by the duration corresponding to the task load period, the theoretical remaining task duration of the antenna under continuous task load conditions can be obtained. Considering that the time the antenna is in program tracking mode each day during actual operation is not fixed, directly using the duration of a single task load period as the lifespan time scale may not reflect the actual usage rhythm of the antenna. Therefore, the average duration of the antenna in program tracking mode each day within a preset range is statistically calculated as the daily average load duration. The remaining available time under continuous task load conditions is converted into a lifespan calculated according to the actual daily task load level by using the ratio of the remaining task duration to the daily average load duration. This allows the prediction result to be expressed as the lifespan of the antenna under actual usage conditions.

[0070] Another embodiment of the present invention provides an antenna lifetime extrapolation system based on gray prediction, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the antenna lifetime extrapolation method based on gray prediction in steps S001 to S006.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for extrapolating antenna lifetime based on grey prediction, characterized in that, The method includes the following steps: Obtain operational data for each time period before and after antenna maintenance, including: historical low load period, historical task load period, verification low load period, and verification task load period; The load reduction coefficient of the antenna is obtained based on the operating data of the verification task load period, historical low load period and historical task load period; Based on the operational data of the verification low load period, the verification task load period, and the historical task load period, as well as the antenna load drop coefficient, the antenna maintenance recovery coefficient is obtained. Combining the operational data of the verification task load period and the historical task load period, the maintenance load resistance repair coefficient is obtained. The maintenance resilience coefficient is obtained based on the operational data of the verification task load period and the historical task load period. The antenna's maintenance status is determined by combining the antenna's maintenance recovery coefficient, maintenance load repair coefficient, and maintenance decay resistance coefficient with the operational data from the verification task load period and historical task load periods. Predict antenna lifespan based on antenna maintenance status.

2. The antenna lifetime extrapolation method based on grey prediction according to claim 1, characterized in that, The specific methods for obtaining operational data for each time period before and after antenna maintenance include: The antenna's azimuth axis speed, pitch axis speed, and operating mode are acquired at each sampling time. The operating modes include: standby mode, manual low-speed operation mode, and program tracking mode. The preset analysis range is divided into time periods based on a complete health assessment cycle of the antenna health management system, and the health value of the antenna in each time period is obtained. For any given time period, the average azimuth axis operating speed, average pitch axis operating speed, cumulative azimuth axis rotation angle, and cumulative pitch axis rotation angle within that time period are taken as the antenna's operating data for that time period. If the operating mode is standby mode or manual low-speed operation mode at all sampling times within the time period, then the time period is recorded as a low-load time period; if the operating mode is program tracking mode at all sampling times within the time period, then the time period is recorded as a task-load time period. The low-load period and the high-load period that occur before the maintenance completion time are recorded as the historical low-load period and the historical high-load period, respectively. The preset observation time window after the maintenance completion time is used as the observation window, and the task load period within the observation window is recorded as the verification task load period. The low-load period that is after the maintenance completion time and before the first verification task load period is denoted as the verification low-load period.

3. The antenna lifetime extrapolation method based on grey prediction according to claim 1, characterized in that, The method for obtaining the antenna load reduction coefficient based on operational data from verification task load periods, historical low load periods, and historical task load periods includes the following specific methods: For any historical task load period, if the previous period of the historical task load period is not a historical task load period, the historical task load period is recorded as a candidate matching period. Based on the running data of the first verification task load period and all candidate matching periods, the nearest neighbor matching algorithm is used to match the first verification task load period with all candidate matching periods to obtain the matching period of the first verification task load period. For any matching time period, the historical low-load time period that is closest to the matching time period and precedes the matching time period is recorded as the source time period of the matching time period; The difference between the health value of the source period and the health value of the matching period is used as the load reduction factor for the matching period. The average load reduction factor of all matching periods during the first verification task load period is used as the antenna load reduction coefficient.

4. The antenna lifetime extrapolation method based on grey prediction according to claim 1, characterized in that, The method for obtaining the antenna's maintenance recovery coefficient based on operational data from verification low-load periods, verification task load periods, historical task load periods, and the antenna's load drop coefficient includes the following specific methods: For any verification task load period, based on the running data of the verification task load period and all historical task load periods, the nearest neighbor matching algorithm is used to match the verification task load period with all historical task load periods, and the historical task load period with the closest matching distance is used as the benchmark matching period for the verification task load period. The difference between the mean of all health values ​​during low-load verification periods and the antenna's load reduction coefficient is used as the antenna's estimated task-state health value. The difference between the antenna's estimated task-state health value and the health value during the baseline matching period of the first verification task load period is used as the antenna's maintenance recovery coefficient.

5. The antenna lifetime extrapolation method based on grey prediction according to claim 4, characterized in that, The specific method for obtaining the repair load-bearing coefficient is as follows: The difference between the health value of the first verification task load period and the health value of its baseline matching period is recorded as the health repair amount of the first verification task load period. The ratio of the health repair amount during the first verification task load period to the antenna's maintenance recovery coefficient is used as the maintenance load resistance repair coefficient.

6. The antenna lifetime extrapolation method based on grey prediction according to claim 5, characterized in that, The method for obtaining the maintenance resilience coefficient based on operational data from the verification task load period and historical task load periods includes: The minimum health repair amount among all verification task load periods is divided by the health repair amount of the first verification task load period, and the ratio is used as the maintenance degradation resistance coefficient.

7. The antenna lifetime extrapolation method based on grey prediction according to claim 4, characterized in that, The method for determining the antenna's maintenance status based on its maintenance recovery coefficient, maintenance load resistance coefficient, and maintenance degradation resistance coefficient, combined with operational data from the verification task load period and historical task load periods, includes the following specific methods: If the antenna's repair recovery coefficient is greater than the preset fluctuation threshold, the repair load resistance repair coefficient is greater than the preset load resistance repair threshold, the repair anti-fading coefficient is greater than the preset anti-fading threshold, and the health repair amount during each verification task load period is greater than the preset fluctuation threshold, the antenna's repair status is recorded as a stable repair state. If the antenna's repair recovery coefficient is less than or equal to the preset fluctuation threshold, or if there is a verification task load period where the amount of health repair is less than or equal to the preset fluctuation threshold, the antenna's repair status will be recorded as a short-term recovery state. If the antenna's maintenance status is not in a stable repair state or a short-term recovery state, the antenna's maintenance status will be recorded as an observation state.

8. The antenna lifetime extrapolation method based on grey prediction according to claim 7, characterized in that, The specific methods for predicting antenna lifespan based on its maintenance status include: If the antenna is in a stable repair state, the sequence of health values ​​for all verification task load periods is used as a new gray prediction sequence. Gray prediction is then performed based on the new gray prediction sequence to predict the antenna lifetime. If the antenna's maintenance status is in a short-term recovery state, the health values ​​of all verification task load periods are spliced ​​into the original gray prediction sequence before maintenance, the gray prediction sequence is updated, and gray prediction is performed based on the updated gray prediction sequence to predict the antenna's lifespan. If the antenna is under observation during maintenance, the antenna lifespan is predicted based on the original gray prediction sequence before maintenance.

9. The antenna lifetime extrapolation method based on grey prediction according to claim 8, characterized in that, If the antenna is under maintenance and in an observation state, the antenna lifespan is predicted based on the original gray prediction sequence before maintenance. The specific methods include: Grey prediction is performed based on the original grey prediction sequence before maintenance to obtain the remaining mission cycles of the antenna. The remaining task duration of the antenna is the product of the remaining task cycle number of the antenna and the corresponding duration of the task load period. The average daily load duration is recorded as the average duration of the antenna being in program tracking mode within the preset statistical range. The antenna's lifespan is the ratio of its remaining operational time to its average daily load time.

10. An antenna lifetime estimation system based on grey prediction, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the antenna lifetime extrapolation method based on gray prediction as described in any one of claims 1-9.

Images