Flight simulator motor degradation monitoring and early warning method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of computers, and discloses a flight simulator motor degradation monitoring and early warning method and system, electronic equipment and a storage medium, and the method comprises the steps: building a health baseline map, and storing reference data of a motor in a health state; in the running process of the simulator, feature parameters are extracted and compared with the health baseline map, and the partition degradation index of each travel partition is calculated; collecting a vibration signal and an acoustic signal, and calculating and generating an abnormal sound quantization index; under the same motion instruction, synchronously collecting characteristic parameters, calculating a transverse difference rate, dynamically determining an adaptive threshold value, determining load intensity according to a partition degradation index historical record, calculating and updating a degradation integral based on the load intensity and duration, and predicting the remaining available life; and injecting an active excitation detection sequence at a preset triggering time, collecting data to generate a health detection report, integrating various indexes to execute hierarchical early warning logic, and outputting early warning information. The operation reliability of the flight simulator motion system can be improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, specifically to a method, system, electronic device, and storage medium for monitoring and early warning of motor degradation in a flight simulator. Background Technology

[0002] Flight simulator motion systems typically employ a six-DOF Stewart platform configuration, driven by six independent electric or hydraulic actuators. This is the core mechanical subsystem for achieving flight motion perception simulation. The drive motors of each actuator must respond precisely to real-time motion commands throughout their entire stroke range, demanding extremely high dynamic performance. Civil aviation regulations impose clear compliance requirements on the performance indicators of simulator motion systems; any perceptible abnormal motion response can lead to a decrease in training effectiveness or even training interruption, thereby affecting pilot training compliance.

[0003] Currently, the status monitoring of drive motors in simulator motion systems mainly relies on hardware protection alarms such as overcurrent, overvoltage, and overtemperature protection built into the drive controller, as well as regular manual inspections and preventative replacements, or subjective sensory abnormalities reported by trainees. All of these methods belong to passive response maintenance strategies and lack the ability to continuously and quantitatively monitor the motor performance degradation process.

[0004] The existing technology has the following drawbacks:

[0005] The alarm threshold of the drive controller protection mechanism is usually set at 150% to 200% or more of the rated parameters. It can only detect abnormalities at the level of hardware damage and cannot detect the performance degradation process of the motor within the rated parameter range, forming a complete technical monitoring blind spot. Motor performance degradation often has a strong correlation with stroke position. The full-stroke average monitoring method dilutes the degradation signal into the average calculation and cannot identify localized degradation features that only appear in specific stroke segments. The identification of actuator vibration and abnormal noise relies entirely on subjective feelings and lacks objective quantitative means, resulting in long fault confirmation cycles and low handling efficiency. Different training subjects have extremely different load requirements for the motion system. The early warning method based on static fixed thresholds cannot distinguish between normal current increase driven by load and abnormal current increase caused by degradation, resulting in a large number of false alarms in high-load subjects or failure to report degradation in low-load subjects. The six sets of actuator cylinder motors have inherent individual differences due to factors such as manufacturing tolerances, installation conditions, and historical running time. The uniform absolute reference threshold has poor applicability to different individuals. Passive monitoring based on runtime data relies on the natural occurrence of specific training subjects to trigger detection conditions. If degradation only appears under specific load-run combinations and the relevant training subjects have not been arranged for a long time, the degradation state may not be stimulated for a long time, resulting in missed detection. Even if the existing monitoring system can detect anomalies, it can only output a binary judgment of whether the current situation is abnormal. It cannot quantitatively assess the remaining usable life and cannot support risk-based predictive maintenance decisions.

[0006] Therefore, this application provides a method for monitoring and early warning of motor degradation in flight simulators to solve the above-mentioned technical problems. Summary of the Invention

[0007] The purpose of this invention is to provide a method, system, electronic device, and storage medium for monitoring and early warning of motor degradation in flight simulators, in order to solve the technical problems in the prior art, such as the inability to continuously and quantitatively monitor the performance degradation of motors in the motion system of flight simulators, the inability to identify position-related degradation, the lack of objective quantitative basis for abnormal noise and vibration, false alarms and missed alarms caused by static fixed thresholds, the failure of reference due to individual differences, the risk of missed detection due to passive monitoring, and the lack of collaborative capabilities for life prediction and maintenance planning.

[0008] To address the aforementioned technical problems, this invention provides a method for monitoring and early warning of motor degradation in flight simulators, comprising: A health baseline map is established with stroke position range and load level as indexes, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in a healthy state; During simulator operation, real-time operation data is collected and feature parameters are extracted. The real-time feature parameters are compared with the baseline data of the corresponding index unit in the health baseline map to calculate the partition degradation index of each travel partition. Real-time acquisition of vibration and acoustic signals, extraction of vibration and acoustic features, and fusion calculation to generate abnormal noise quantification index; Under the same motion command, key characteristic parameters of multiple actuator cylinder motors on the same motion platform are collected synchronously, the lateral difference rate is calculated and the adaptive threshold is dynamically determined. At the same time, the load intensity is determined according to the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral. An active stimulus detection sequence is injected at a preset trigger time. All data during the execution of the active stimulus detection sequence is collected and analyzed to generate a health detection report. The zonal degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report are combined to execute a graded early warning logic and output early warning information.

[0009] In some specific embodiments, a health baseline map is established indexed by stroke position range and load level. The health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in its healthy state, further including: When the actuator motor is in its initial healthy state, the actuator is controlled to move within its full stroke range and operate at different load levels. The drive current, stroke position, vibration and acoustic signals of the actuator cylinder motor are collected, and feature parameters of multiple dimensions are extracted. Among them, the feature parameters of multiple dimensions include at least the effective value of current, vibration intensity, kurtosis value and sound pressure level. Using the travel location range and load level as two-dimensional indexes, the mean and standard deviation of the extracted feature parameters from multiple dimensions are used as benchmark data and stored in a three-dimensional matrix structure to form a health baseline map. An exponentially weighted moving average method is used to continuously update the baseline data in the health baseline map based on newly added health status data.

[0010] In some specific embodiments, during simulator operation, real-time operation data is collected and feature parameters are extracted. These real-time feature parameters are compared with the baseline data of the corresponding index unit in the health baseline map to calculate the partition degradation index for each travel partition. This further includes: The entire stroke of the actuator is divided into multiple sub-sections, which are independent stroke zones; During simulator operation, feature parameters of multiple dimensions under the current travel partition are collected in real time. The feature parameters collected in real time from multiple dimensions are compared with the baseline data under the corresponding travel partition and load level in the health baseline map, and the normalized deviation of each feature parameter is calculated respectively. The normalized deviations of each feature parameter are weighted and summed to obtain the partition degradation index of the current travel partition.

[0011] In some specific embodiments, vibration and acoustic signals are acquired in real time, vibration and acoustic features are extracted and fused to generate an abnormal noise quantification index, further including: Real-time acquisition of vibration signals from the actuator cylinder and acoustic signals near the mounting point; The vibration intensity and kurtosis values ​​are extracted from the vibration signal, and the total sound pressure level is extracted from the acoustic signal; The extracted vibration intensity, kurtosis value, and total sound pressure level are compared with the corresponding vibration intensity reference value, kurtosis reference value, and sound pressure level reference value in the health baseline map, respectively, to obtain the normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation. The normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation are multiplied by preset fusion weighting coefficients and then summed to generate an abnormal noise quantification index.

[0012] In some specific embodiments, under the same motion command, key characteristic parameters of multiple actuator cylinder motors on the same motion platform are synchronously collected, the lateral difference rate is calculated, and an adaptive threshold is dynamically determined. Simultaneously, the load intensity is determined based on the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral. Further, this includes: After each motion command is executed, the key characteristic parameters of all actuator cylinder motors on the same motion platform are obtained, and the arithmetic mean of the key characteristic parameters of all actuator cylinder motors is calculated. The key characteristic parameters include at least the effective value of the current and the response time. The lateral difference rate of the corresponding actuator motor is calculated based on the ratio of the difference between the key characteristic parameters of a single actuator motor and the arithmetic mean to the arithmetic mean, and an adaptive threshold is dynamically determined based on the rolling standard deviation of the historical lateral difference rate sequence of the corresponding actuator motor. Within an update cycle, the degradation integral increment is calculated based on the product of the normalized load intensity and duration borne by the motor, combined with the material equivalent fatigue index. The degradation integral increment is then added to the degradation integral value of the previous moment to obtain the updated degradation integral. The degradation rate is fitted based on the degradation integral values ​​at multiple consecutive times, and the remaining usable lifetime is predicted based on the ratio of the difference between the preset degradation integral threshold and the current degradation integral value to the degradation rate.

[0013] In some specific embodiments, an active stimulus detection sequence is injected at a preset triggering time, and full data during the execution of the active stimulus detection sequence is collected and analyzed to generate a health detection report, further including: During the reset process after training or when a manual trigger command is received, an active excitation detection sequence is generated, which includes full-stroke low-speed scanning, step response test and constant load output test. The speed of the full-stroke low-speed scanning is a preset percentage of the rated speed. The active excitation detection sequence is injected into the motion control system to control the actuator to perform actions according to the sequence. During the sequence execution, the current, stroke position, vibration and acoustic signals of the actuator are collected synchronously, and the response time, overshoot, steady-state tracking error and full stroke zone degradation index distribution are calculated. The response time, overshoot, steady-state tracking error, and full-stroke partition degradation index distribution are compared with the initial baseline to generate a structured health test report containing execution timestamps, test feature values ​​for each segment, deviation, and comprehensive health index.

[0014] In some specific embodiments, the tiered early warning logic is executed and early warning information is output by combining the partition degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring report, and further includes: Real-time reception and integration of regional degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring reports from active stimulation detection; The received indicators are compared with the trigger conditions of multiple preset warning levels. The trigger conditions include at least the partition degradation index exceeding the baseline standard deviation multiple, the abnormal noise quantification index exceeding the threshold, and the remaining usable lifespan being lower than the preset alarm time limit. When any indicator meets the triggering conditions for any warning level, a warning event corresponding to the corresponding warning level is generated, wherein the warning event includes the warning level, device identifier and description of the triggering conditions; The warning event is pushed to the maintenance terminal, and the warning event and the associated feature parameter snapshot are written into the electronic health record. At the same time, based on the remaining usable life prediction results and training plan information, a dynamic maintenance window recommendation is calculated and output.

[0015] Based on the same concept, the present invention also provides a flight simulator motor degradation monitoring and early warning system, comprising: The health baseline map establishment module is configured to establish a health baseline map indexed by stroke position range and load level, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in a healthy state; The partition degradation index calculation module is configured to collect running data and extract feature parameters in real time during the operation of the simulator, compare the real-time feature parameters with the benchmark data of the corresponding index unit in the health baseline map, and calculate the partition degradation index of each travel partition. The abnormal noise quantification index generation module is configured to collect vibration and acoustic signals in real time, extract vibration and acoustic features, perform fusion calculations, and generate an abnormal noise quantification index. The adaptive threshold and predicted remaining usable life calculation module is configured to synchronously collect key characteristic parameters of multiple actuator cylinder motors on the same motion platform under the same motion command, calculate the lateral difference rate and dynamically determine the adaptive threshold. At the same time, it determines the load intensity based on the historical records of the partition degradation index, calculates and updates the degradation integral based on the load intensity and duration, and predicts the remaining usable life based on the degradation integral. The early warning information output module is configured to inject an active stimulus detection sequence at a preset trigger time, collect and analyze all data during the execution of the active stimulus detection sequence, generate a health detection report, and combine the partition degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report to execute hierarchical early warning logic and output early warning information.

[0016] Based on the same concept, the present invention also provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of a method for monitoring and warning of motor degradation in a flight simulator.

[0017] Based on the same concept, the present invention also provides a computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of a method for monitoring and warning of motor degradation in a flight simulator.

[0018] Compared with existing technologies, its advantages are as follows: This invention discloses a method, system, electronic device, and storage medium for monitoring and early warning of motor degradation in flight simulators, which can fill the blind spot in motor performance degradation monitoring: by establishing a three-dimensional health baseline map indexed by travel position intervals and load levels, and combining it with a partitioned degradation index, continuous quantitative monitoring is achieved within a wide degradation range below the hardware protection threshold of the motor drive controller, thus filling the complete monitoring blind spot of the degradation process from health to failure.

[0019] Achieve accurate identification of location-related degradation: By dividing the entire journey into several zones, establishing baselines for each zone, and independently calculating the degradation index, it is possible to accurately identify localized degradation that occurs only in specific travel segments, improving the monitoring accuracy from the average level of the entire journey to the fine level of each zone, and pinpointing the specific travel interval of degradation.

[0020] Transforming subjective abnormal noises and vibrations into objective quantitative indicators: By extracting features from the fusion of vibration and acoustic sensors, an abnormal noise quantitative index is constructed. This transforms the judgment of abnormal noises and vibrations, which relies entirely on subjective feelings, into objective, recordable, traceable, and comparable quantitative data, providing objective data support for fault confirmation and maintenance decisions.

[0021] Adaptive threshold elimination of false alarms and missed alarms: By using a horizontal comparison mechanism with multiple actuator cylinder motors on the same platform as mutual reference benchmarks, and by dynamically determining the adaptive threshold based on historical statistics, the impact of differences in training subject loads and individual equipment differences on monitoring accuracy is effectively eliminated, achieving low false alarm rate and low missed alarm rate under complex dynamic working conditions.

[0022] Upgrading from fault alarms to predictive maintenance: By modeling degradation trends based on equivalent fatigue integrals and predicting remaining usable life, the output of the monitoring system is upgraded from a binary judgment of whether the current situation is abnormal to a quantitative prediction of remaining usable time, providing a technical foundation for the transformation from planned preventive maintenance to condition-based predictive maintenance.

[0023] Eliminating the risk of missed detections in passive monitoring: By designing an active excitation detection sequence that includes full-stroke low-speed scanning, step response testing, and constant load output testing, and automatically executing it during training intervals, the system actively covers the entire stroke and all test conditions, thus eliminating the risk of missed detections that passive monitoring relies on the natural triggering of specific training subjects.

[0024] Achieve collaborative optimization of maintenance plans and training schedules: Through the dynamic maintenance window optimization module, the remaining usable life prediction results are deeply integrated with the training plan information stored in electronic data form, automatically recommending maintenance time nodes with the least impact on training continuity, and supporting the optimal allocation of maintenance resources while ensuring uninterrupted flight training.

[0025] Establish a full lifecycle health traceability record for equipment: Through the electronic health record module, a complete digital record of the health status of each actuator motor is established from installation to decommissioning. This supports retrospective analysis of degradation trends, evaluation of maintenance effectiveness, and cross-equipment horizontal comparison, providing data support for subsequent equipment selection, maintenance strategy optimization, and spare parts management. Attached Figure Description

[0026] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating some specific embodiments of the method for monitoring and early warning of motor degradation in a flight simulator according to the present invention; Figure 2 This is a schematic diagram of the structure of a flight simulator motor degradation monitoring and early warning system in some specific embodiments of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device according to some specific embodiments of the present invention; In the diagram, 710 is the processor; 720 is the memory; 730 is the input device; and 740 is the output device. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0029] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0030] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.

[0031] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0032] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0033] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0034] Reference Figure 1 A method for monitoring and early warning of motor degradation in flight simulators, comprising: S101, establish a health baseline map indexed by stroke position range and load level, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in a healthy state; S102, During the operation of the simulator, real-time operation data is collected and feature parameters are extracted. The real-time feature parameters are compared with the baseline data of the corresponding index unit in the health baseline map to calculate the partition degradation index of each travel partition. S103, real-time acquisition of vibration and acoustic signals, extraction of vibration and acoustic features and fusion calculation to generate abnormal noise quantification index; S104, under the same motion command, the key characteristic parameters of multiple actuator cylinder motors on the same motion platform are collected synchronously, the lateral difference rate is calculated and the adaptive threshold is dynamically determined. At the same time, the load intensity is determined according to the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral. S105, inject an active stimulus detection sequence at a preset trigger time, collect and analyze all data during the execution of the active stimulus detection sequence, generate a health detection report, and combine the partition degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report to execute graded early warning logic and output early warning information.

[0035] Specifically, in this embodiment of the invention, a health baseline map is established. During the initial operation phase after a new installation or major overhaul of the actuator motor, the actuator is controlled to move within its full stroke range and operate at different load levels. Drive current, stroke position, vibration signals, and acoustic signals are collected, and multiple characteristic parameters are extracted from them, including at least the effective value of the current, vibration intensity, kurtosis value, and sound pressure level. Using the stroke position range and load level as two-dimensional indices, the mean and standard deviation of the above characteristic parameters under healthy conditions are used as benchmark data and stored in a three-dimensional matrix structure to form a health baseline map. During normal operation of the simulator, the operating data of the actuator cylinder motor is continuously collected, and the same multidimensional feature parameters as those in the health baseline map are extracted. The real-time collected stroke position and load level are mapped to the corresponding index unit of the health baseline map. The real-time feature parameters are compared with the benchmark data in the corresponding index unit. For each stroke zone, the normalized deviation of each feature parameter is calculated, that is, the real-time feature parameter minus the corresponding benchmark mean and then divided by the benchmark standard deviation. The normalized deviation of each feature parameter is weighted and summed according to the preset weight to obtain the zone degradation index of each stroke zone. Simultaneously, vibration signals from the actuator cylinder and acoustic signals near the installation point are acquired in real time. Vibration intensity and kurtosis values ​​are extracted from the vibration signals, and total sound pressure level is extracted from the acoustic signals. The extracted vibration intensity, kurtosis, and total sound pressure level are normalized and compared with the corresponding vibration intensity, kurtosis, and sound pressure level benchmark values ​​in the health baseline map to obtain normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation. These three types of deviations are then multiplied by preset fusion weight coefficients and summed to generate an abnormal noise quantification index. After each motion command is executed, key characteristic parameters of all actuator motors on the same motion platform are synchronously acquired. The arithmetic mean of the key characteristic parameters of all actuator motors is calculated. The lateral difference rate of an actuator motor is calculated based on the ratio of the difference between the key characteristic parameter of a single actuator motor and the arithmetic mean to the arithmetic mean. An adaptive threshold is dynamically determined based on the rolling standard deviation of the historical lateral difference rate sequence of the actuator motor. The load intensity borne by the motor is determined based on the historical records of the degradation index of the zone. Within one update cycle, the degradation integral increment is calculated based on the product of the load intensity and its duration, combined with the material equivalent fatigue index. This increment is added to the degradation integral value of the previous moment to obtain the updated degradation integral. The degradation rate is then fitted based on the degradation integral values ​​of multiple consecutive moments. The remaining usable life is predicted based on the ratio of the difference between the preset degradation integral threshold and the current degradation integral value to the degradation rate.During the reset process after training or upon receiving a manual trigger command, an active excitation detection sequence is generated. This sequence includes a full-stroke low-speed scan, a step response test, and a constant load output test. The sequence is injected into the motion control system to control the actuator. During the sequence execution, the actuator's current, stroke position, vibration, and acoustic signals are simultaneously acquired. The response time, overshoot, steady-state tracking error, and full-stroke zone degradation index distribution are calculated. These indicators are compared with the initial baseline to generate a structured health inspection report containing the execution timestamp, test characteristic values ​​for each segment, deviation, and a comprehensive health index. Finally, the zone degradation index, abnormal noise quantification index, lateral difference rate, remaining usable life, and health inspection report are comprehensively received and integrated. Each indicator is compared with the trigger conditions of multiple preset warning levels. When any indicator meets the trigger conditions of any warning level, a warning event corresponding to that warning level is generated. The warning event is pushed to the maintenance terminal, and a snapshot of the warning event and associated characteristic parameters is written into the electronic health record. Simultaneously, based on the remaining usable life prediction results and the training plan information stored in electronic data form, a dynamic maintenance window recommendation is calculated and output.

[0036] In some applications, a health baseline map is established, indexed by stroke position range and load level. This health baseline map stores benchmark data of multiple dimensions of characteristic parameters of the actuator motor in its healthy state. This includes controlling the actuator to move within its full stroke range and operating at different load levels when the actuator motor is in its initial healthy state; collecting the actuator motor's drive current, stroke position, vibration, and acoustic signals, and extracting multiple dimensions of characteristic parameters, including at least the effective value of current, vibration intensity, kurtosis value, and sound pressure level; using the stroke position range and load level as two-dimensional indexes, the mean and standard deviation of the extracted multiple dimensions of characteristic parameters are used as benchmark data and stored in a three-dimensional matrix structure to form the health baseline map; and using an exponentially weighted moving average method, the benchmark data in the health baseline map is continuously updated based on newly added health state data.

[0037] Understandably, when the actuator motor is in its initial healthy state—that is, after the motor has been newly installed or overhauled and commissioned to ensure no abnormal operation—the actuator is controlled to move within its full stroke range, traversing all position intervals from minimum to maximum stroke, and operating at different load levels, covering various working conditions from no-load to rated load. During this process, sensors collect the actuator motor's drive current, stroke position, vibration signals, and acoustic signals. Multiple characteristic parameters are extracted from the collected signals. These parameters include at least the effective current value representing the current load level, the vibration intensity representing the total vibration energy, the kurtosis value representing the vibration impact characteristics, and the total sound pressure level representing the overall acoustic intensity. Using the stroke position interval and load level as a two-dimensional index, the mean and standard deviation of the above-mentioned multiple characteristic parameters in the healthy state are used as baseline data and stored in a three-dimensional matrix structure according to the stroke partition, load level, and characteristic dimensions, thus forming a healthy baseline map. During subsequent operation, when new health status data is collected, the baseline data in the health baseline map is updated in a rolling manner using an exponentially weighted moving average method. That is, the measured mean of the newly added health status data and the current baseline mean are weighted by a preset smoothing coefficient to obtain the updated baseline mean. This allows the baseline to slowly adjust smoothly with the normal aging drift of the equipment, while avoiding contamination by degraded data, thereby maintaining the timeliness and accuracy of the baseline map.

[0038] In some applications, during simulator operation, real-time operating data is collected and feature parameters are extracted. These real-time feature parameters are compared with the baseline data of the corresponding index unit in the health baseline map to calculate the partition degradation index of each stroke partition. This includes dividing the entire stroke of the actuator into multiple sub-intervals as independent stroke partitions. During simulator operation, feature parameters of multiple dimensions under the current stroke partition are collected in real time. The real-time collected feature parameters of multiple dimensions are compared with the baseline data of the corresponding stroke partition and load level in the health baseline map to calculate the normalized deviation of each feature parameter. The normalized deviations of each feature parameter are weighted and summed to obtain the partition degradation index of the current stroke partition.

[0039] Understandably, the entire stroke of the actuator is divided into multiple sub-intervals, each serving as an independent stroke zone, to achieve refined monitoring of position-related degradation. During normal simulator training, the actuator motor's operating data is collected in real time. Based on the current stroke position, the stroke zone is determined, and multiple characteristic parameters of the same dimensions as those in the health baseline map are simultaneously collected within that stroke zone, including at least the RMS current value, vibration intensity, kurtosis value, and sound pressure level. These real-time collected characteristic parameters are compared item by item with the baseline data for the corresponding stroke zone and current load level in the health baseline map. For each characteristic parameter, the difference between its real-time value and the baseline mean is calculated, and then divided by the baseline standard deviation of that characteristic parameter to obtain the normalized deviation. This deviation value characterizes the degree of deviation of the characteristic parameter from its healthy state. After calculating the normalized deviation of all characteristic parameters, based on the weighting coefficients preset for each characteristic parameter during system initial calibration, the normalized deviation of each characteristic parameter is multiplied by its corresponding weighting coefficient and summed to obtain the zone degradation index for the current stroke zone. As a comprehensive quantitative indicator, the degradation index of this zone can reflect the overall degradation degree of the actuator motor within a specific stroke zone, providing basic data for subsequent early warning judgment and life prediction.

[0040] In some applications, vibration and acoustic signals are acquired in real time, vibration and acoustic features are extracted and fused to generate an abnormal noise quantification index. This includes real-time acquisition of vibration signals from the actuator cylinder and acoustic signals near the installation point; extraction of vibration intensity and kurtosis values ​​from the vibration signals and extraction of total sound pressure level from the acoustic signals; normalization comparison of the extracted vibration intensity, kurtosis, and total sound pressure level with the corresponding vibration intensity, kurtosis, and sound pressure level reference values ​​in the health baseline spectrum to obtain normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation; and summing of the normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation by preset fusion weighting coefficients to generate the abnormal noise quantification index.

[0041] Understandably, a triaxial accelerometer is installed on the flange face at the end of the actuator cylinder to collect vibration signals from the cylinder in three directions in real time. Simultaneously, a MEMS microphone is fixed at a predetermined distance from the actuator cylinder surface to collect acoustic signals generated during the actuator's operation. After continuously acquiring raw data from these two types of sensors, the system performs time-domain and frequency-domain analysis on the vibration signals, extracting vibration intensity and kurtosis values. Vibration intensity is obtained by calculating the effective value of the vibration signal, reflecting the overall level of vibration energy. Kurtosis is obtained by calculating the fourth moment of the vibration signal, reflecting the strength of the impact transient component in the vibration waveform and being sensitive to early mechanical damage. Spectral analysis is performed on the acoustic signals to extract the total sound pressure level as an acoustic feature, reflecting the overall intensity of the abnormal noise. The extracted vibration intensity, kurtosis value, and total sound pressure level are compared with the corresponding baseline values ​​for vibration intensity, kurtosis, and sound pressure level stored in the health baseline map for the corresponding stroke zones and load levels. The normalized deviation of each characteristic parameter is calculated by subtracting the baseline value from the real-time value and then dividing by the baseline value, yielding the normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation. After calculating these three types of deviations, based on the fusion weight coefficients determined by expert knowledge during the initial system calibration phase, the normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation are multiplied by their respective fusion weight coefficients and summed to generate an abnormal noise quantification index. This index, as a dimensionless comprehensive quantification indicator, has its value range normalized to a preset interval and is used to objectively characterize the degree of vibration and abnormal noise during the operation of the actuator cylinder.

[0042] In some applications, under the same motion command, key characteristic parameters of multiple actuator motors on the same motion platform are synchronously acquired, the lateral difference rate is calculated, and an adaptive threshold is dynamically determined. Simultaneously, the load intensity is determined based on the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral. This includes acquiring the key characteristic parameters of all actuator motors on the same motion platform after each motion command execution, calculating the arithmetic mean of all key characteristic parameters of the actuator motors, wherein the key characteristic parameters include at least the effective value of the current and the response time; and based on the key characteristics of a single actuator motor... The lateral difference rate of the corresponding actuator motor is calculated by the ratio of the difference between the parameter and the arithmetic mean to the arithmetic mean, and an adaptive threshold is dynamically determined based on the rolling standard deviation of the historical lateral difference rate sequence of the corresponding actuator motor. Within an update cycle, the degradation integral increment is calculated based on the product of the normalized load intensity and duration borne by the motor, combined with the material equivalent fatigue index. The degradation integral increment is accumulated to the degradation integral value of the previous moment to obtain the updated degradation integral. The degradation rate is fitted based on the degradation integral values ​​of multiple consecutive moments, and the remaining usable life is predicted based on the ratio of the difference between the preset degradation integral threshold and the current degradation integral value to the degradation rate.

[0043] Understandably, after each motion command is executed, the system synchronously acquires the key characteristic parameters of all actuator motors on the same motion platform. These key characteristic parameters include at least the effective value of the current characterizing the motor load level and the response time characterizing the dynamic response performance. The arithmetic mean of the key characteristic parameters of all actuator motors is calculated as the intra-group reference benchmark under that motion command. For a single actuator motor, the difference between its key characteristic parameters and the intra-group arithmetic mean is divided by the arithmetic mean to obtain the lateral deviation rate of that motor. This ratio reflects the degree of performance deviation of that motor relative to other motors on the same platform. Based on the historical lateral deviation rate sequence of the motor, its rolling standard deviation is calculated, and the rolling standard deviation is multiplied by a preset multiple according to a preset confidence coefficient to dynamically determine an adaptive threshold. When the absolute value of the lateral deviation rate exceeds the adaptive threshold, a lateral anomaly is determined to exist. Meanwhile, regarding degradation integral updates, the normalized load intensity borne by the motor during operation is determined based on the historical records of the degradation index for each zone. Within an update cycle, the load intensity corresponding to each time period within that cycle is indexed, i.e., the material equivalent fatigue index of the load intensity is raised to the power of the index, multiplied by the duration of that time period, and the product of all time periods is integrated and summed to obtain the degradation integral increment for that update cycle. This increment is then added to the degradation integral value of the previous moment to obtain the updated degradation integral. This degradation integral increases monotonically with operating time, comprehensively reflecting the cumulative fatigue loss throughout the motor's entire life cycle. Regarding remaining usable life prediction, a linear fitting method is used to fit the degradation rate based on the degradation integral values ​​at multiple consecutive moments. This degradation rate characterizes the increase in degradation integral per unit time. The difference between the preset degradation integral threshold and the current degradation integral value is divided by this degradation rate to obtain the predicted remaining usable life. This predicted value is output in training flight hours or calendar days to support predictive maintenance decisions.

[0044] In some applications, an active excitation detection sequence is injected at a preset trigger time. All data during the execution of the active excitation detection sequence is collected and analyzed to generate a health monitoring report. This includes generating the active excitation detection sequence during the reset process after training or upon receiving a manual trigger command. The active excitation detection sequence includes full-stroke low-speed scanning, step response testing, and constant load output testing. The speed of the full-stroke low-speed scanning is a preset percentage of the rated speed. The active excitation detection sequence is injected into the motion control system to control the actuator to perform actions according to the sequence. During sequence execution, the current, stroke position, vibration, and acoustic signals of the actuator are simultaneously collected. The response time, overshoot, steady-state tracking error, and full-stroke partition degradation index distribution are calculated. The response time, overshoot, steady-state tracking error, and full-stroke partition degradation index distribution are compared with the initial baseline to generate a structured health monitoring report containing execution timestamps, test characteristic values ​​for each segment, deviation, and a comprehensive health index.

[0045] Understandably, during the reset process after training or upon receiving a manual trigger command from maintenance personnel, the system automatically generates an active excitation detection sequence. This sequence consists of three standardized tests: the first is a full-stroke low-speed scan, where the actuator is driven at a preset percentage of the rated speed from the minimum stroke to the maximum stroke and then back to the middle position, used to acquire continuous state data across the entire stroke range; the second is a step response test, where step position commands are applied at multiple preset stroke positions to evaluate the motor's dynamic response performance; the third is a constant load output test, where a rated load is applied at a specified stroke position and held for a preset duration to evaluate the motor's steady-state output capability under rated operating conditions. The system injects the generated active excitation detection sequence into the motion control system, controlling the actuator to execute the action commands set in the sequence sequentially. During sequence execution, the system synchronously acquires the actuator's current signal, stroke position signal, vibration signal, and acoustic signal, and calculates multiple health indicators from the acquired data, including response time, overshoot, and settling time in the step response test, steady-state tracking error in the constant load test, and the full-stroke zone degradation index distribution during the full-stroke low-speed scan. The calculated health indicators are compared with the corresponding benchmark values ​​stored in the initial baseline map to calculate the deviation of each indicator. The comprehensive health index is generated by combining the overall performance of each test segment. Finally, a structured health test report is generated, which includes the execution timestamp, characteristic values ​​of each test segment, deviation, and comprehensive health index. This report is written into the electronic health record for historical traceability and trend analysis.

[0046] In some applications, the tiered early warning logic is executed and early warning information is output by integrating the zonal degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring reports. This includes real-time reception and integration of the zonal degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring reports from active incentive detection; comparison of the received indicators with the triggering conditions of multiple preset early warning levels, wherein the triggering conditions include at least the zonal degradation index exceeding the baseline standard deviation multiple, the abnormal noise quantification index exceeding a threshold, and the remaining usable life being lower than a preset alarm time limit; when any indicator meets the triggering condition of any early warning level, an early warning event corresponding to the corresponding early warning level is generated, wherein the early warning event includes the early warning level, device identifier, and description of the triggering conditions; the early warning event is pushed to the maintenance terminal, and the early warning event and associated feature parameter snapshots are written into the electronic health record; simultaneously, based on the remaining usable life prediction results and training plan information, a dynamic maintenance window recommendation is calculated and output.

[0047] Understandably, the system receives and integrates outputs from various analysis modules in real time, including the partition degradation index for each travel partition, the abnormal noise quantification index representing the degree of jitter and abnormal noise, the lateral difference rate reflecting the lateral deviation within the same platform, the predicted remaining usable life representing cumulative fatigue loss, and the structured health detection report generated after active stimulation detection. These multi-dimensional health indicators are compared item by item with preset trigger conditions for multiple warning levels. The preset trigger conditions include at least: the partition degradation index exceeding the baseline standard deviation by a preset multiple; the abnormal noise quantification index exceeding a preset threshold; the remaining usable life falling below a preset alarm timeout; the lateral difference rate continuously exceeding the adaptive threshold by a preset multiple; and the active detection health index falling below a preset score. When any indicator meets the trigger condition for any warning level, a warning event corresponding to that warning level is generated based on the highest level of the met condition. This warning event includes at least a warning level identifier, the device number or actuator number where the abnormality occurred, a description of the specific conditions triggering the warning, and a snapshot of the characteristic parameters at the trigger time. The system pushes generated early warning events to the maintenance terminal in real time for maintenance personnel to view. Simultaneously, it writes the early warning events and their associated feature parameter snapshots, trigger timestamps, and other information into the device's structured electronic health record, forming a complete historical traceability record. Based on this, and using the remaining usable life prediction results and training plan information stored in electronic data form, the system calculates a comprehensive score for each candidate maintenance time node. The comprehensive score considers at least two dimensions: maintenance safety margin and the degree of impact on training continuity. The system selects the highest-scoring candidate windows as dynamic maintenance window recommendations for maintenance supervisors' decision-making reference.

[0048] The following describes another embodiment of the method for monitoring and early warning of motor degradation in a flight simulator according to the present invention: In this embodiment, the definitions of abbreviations and key terms include: Health Baseline Map (HBM): During the factory commissioning or initial online phase of the equipment, a three-dimensional parameter baseline data set indexed by the stroke position is established by systematically collecting the operating parameters such as current, speed, and power of the motor throughout its full stroke and full load range. This data is used for real-time comparison during subsequent operation to identify degradation deviations.

[0049] Abnormal Response Index (ARI): A dimensionless quantitative index obtained by combining vibration acceleration sensor and acoustic sensor signals through feature extraction and fusion calculation; used to convert the vibration and abnormal noise in the operation of the actuator from subjective perception into objective and recordable values, supporting graded early warning and historical traceability.

[0050] Degradation Integral (DI): A dimensionless comprehensive degradation quantification value calculated based on the historical operating load intensity and duration of the motor according to the equivalent fatigue accumulation model; the DI value increases monotonically with the operating time, and a corresponding level of warning is triggered when the DI reaches a preset threshold.

[0051] Remaining Useful Life (RUL): Based on the current degradation integral value and degradation rate, predict the remaining usable time (expressed in training flight hours or calendar days) before the motor reaches the point of unacceptable performance degradation while maintaining the current usage intensity.

[0052] Zonal Degradation Index (ZDI): The local degradation quantification value is calculated independently in each sub-interval by dividing the entire stroke of the actuator into several sub-intervals; it is used to identify position-related degradation that occurs only in a specific stroke segment, which is different from the full-stroke mean monitoring method.

[0053] Lateral Deviation Ratio (LDR): Under the same motion command, the normalized deviation of key parameters (such as current and response time) of each actuator motor on the same motion platform relative to the mean of the group; used to achieve adaptive threshold judgment by using other normal motors on the same platform as mutual reference benchmarks.

[0054] Active Excitation Detection Sequence (AEDS): A set of standardized motion commands automatically injected into the motion control system by the monitoring system during training breaks or equipment reset; including low-speed full-stroke scanning, step response testing and constant torque loading testing, used to periodically and actively acquire motor health status characteristics without waiting for specific training subjects to trigger naturally.

[0055] Electronic Health Record (EHR): A structured and persistent digital archive of historical monitoring data, active detection results, degradation trend curves, early warning records and maintenance events for each actuator motor; supports multi-dimensional retrieval and auditing by device number, time period, early warning level and other dimensions.

[0056] Dynamic Maintenance Window Optimization (DMWO): This function automatically recommends maintenance time nodes that have the least impact on training continuity by combining RUL prediction results with training center scheduling plans.

[0057] The monitoring and early warning system in this embodiment adopts a four-layer architecture of "perception layer - analysis layer - decision layer - output layer". Each layer has independent functions and standard interfaces, and can be deployed and implemented in a bypass manner without changing the existing motion control system hardware architecture.

[0058] Sensing layer: responsible for acquiring raw signals, including motor drive current, actuator stroke position encoder signal, cylinder vibration acceleration signal, and acoustic signals near the mounting point.

[0059] Analysis layer: Based on the signal from the perception layer, it performs feature extraction and degradation quantization calculation, including three-dimensional feature map comparison, abnormal noise quantization index calculation, lateral difference rate calculation and degradation integral update.

[0060] Decision-making level: Based on the output of the analysis layer, it performs hierarchical early warning judgment, remaining life prediction, and maintenance window optimization recommendation.

[0061] Output layer: Provides maintenance personnel with real-time early warning push, electronic health records of equipment, degradation trend reports and maintenance suggestion sheets, and provides standardized data interface support for integration with maintenance management system (MMS) or training management system (TMS).

[0062] It includes the following functional modules: Sensor access module: responsible for signal acquisition, A / D conversion and timestamp alignment of motor drive current sensor (Hall type or shunt type), stroke encoder, triaxial accelerometer (installed at the end of actuator cylinder body) and MEMS microphone (installed in the adjacent structure of actuator cylinder); sampling rate requirements: current and position signals not less than 1kHz, vibration signals not less than 5kHz, acoustic signals not less than 20kHz.

[0063] Feature extraction module: preprocesses and calculates features of the raw sensor signal, and outputs feature vectors for subsequent analysis; specifically including: RMS current, peak current, current waveform distortion rate, Kurtosis value of vibration signal, vibration intensity (RMS), energy of specified frequency band, sound pressure level (SPL), and amplitude of acoustic anomaly frequency components.

[0064] Health Baseline Map Module: Stores and maintains the three-dimensional health baseline map (HBM) for each motor, indexed by stroke position range (recommended resolution not less than 5% stroke / grid) and load level, and stores the initial baseline mean and confidence interval of each feature; supports baseline initialization, periodic updates and version management.

[0065] Partition Degradation Analysis Module: Compares the real-time feature vector with the baseline under the corresponding travel-load index to calculate the partition degradation index (ZDI) of each travel partition; when the ZDI of any partition continuously exceeds the threshold, the partition is marked as a degradation interval and recorded.

[0066] Vibration-Acoustic Fusion Module: Performs fusion calculation of vibration feature vector and acoustic feature vector, and outputs the Abnormal Noise Quantization Index (ARI).

[0067] Lateral Comparison Analysis Module: During the execution of the same motion command, key feature values ​​of all six sets of actuators on the same platform are collected synchronously, and the lateral difference rate (LDR) of each motor is calculated; when the LDR exceeds the adaptive threshold, a lateral anomaly marker is triggered.

[0068] Degradation Integral and Lifetime Prediction Module: Calculates degradation integral on a rolling basis to predict remaining usable life (RUL); triggers an alert when RUL falls below a preset alarm time limit.

[0069] Active stimulus detection module: Monitors the operating status of the equipment and injects an active stimulus detection sequence (AEDS) into the motion control system during the reset process after training or when manually triggered by maintenance personnel; collects and analyzes all sensor data during the execution of AEDS, generates periodic health monitoring reports and writes them into the electronic health record.

[0070] Early warning management module: integrates the output of zonal degradation analysis, vibration acoustic fusion, lateral comparison analysis, degradation integral and active detection, executes hierarchical early warning logic, generates early warning events (including level, equipment number, travel position, characteristic deviation value, timestamp) and pushes them to the maintenance terminal.

[0071] Electronic Health Record Module: Establishes an independent structured electronic health record for each actuator cylinder motor, persistently storing historical monitoring data, proactive detection reports, early warning events, and maintenance records; supports multi-dimensional retrieval, trend visualization, and data export; the record data includes timestamps and integrity verification, supporting audit traceability.

[0072] Dynamic Maintenance Window Optimization Module: Reads RUL prediction results and external training scheduling plans (obtained through a standard interface), calculates the impact score of each possible maintenance time point on training continuity, and outputs an optimal maintenance window recommendation list and maintenance suggestion sheet.

[0073] In summary, the process of this embodiment includes: Baseline initialization: During the initial operation phase after a new motor is installed or overhauled (it is recommended that the cumulative operating time be no less than 100 hours), the system collects motor operation data under full stroke and multi-load conditions to establish an initial health baseline map (HBM).

[0074] Real-time online monitoring: During normal training operation of the simulator, the sensor access module continuously collects raw signals, the feature extraction module calculates feature vectors in real time, and outputs them to the subsequent analysis module at a frequency of no less than 1 time / second.

[0075] Step-by-step partition degradation calculation: The partition degradation analysis module maps real-time feature values ​​to corresponding HBM index cells and calculates the partition degradation index: ; in, For the first Partition degradation index for each travel partition; This is the route partition number, with a value ranging from 1 to the total number of partitions in the entire route; The feature dimension is numbered, with values ​​ranging from 1 to... ; The total number of dimensions of the features involved in the degradation calculation; This is to perform a weighted summation operation across all feature dimensions; For the first The weight coefficients corresponding to the dimensional features are determined by the initial system calibration, and the sum of all weight coefficients is 1. This is a maximum value function used to filter out negative biases where the feature value is below the baseline, retaining only the positive biases corresponding to performance degradation; For the first The first travel zone, the first Real-time acquisition and calculation of 3D feature values; In the health baseline map, the first The first travel zone, the first The initial baseline mean corresponding to the dimensional feature quantity; In the health baseline map, the first The first travel zone, the first The initial baseline standard deviation corresponding to the dimensional feature.

[0076] Abnormal Noise Quantification Index Calculation: The vibration-acoustic fusion module calculates the abnormal noise quantization index: ; in, This is the abnormal noise quantification index, with values ​​normalized to [0, 10]. These are the fusion weighting coefficients corresponding to the vibration and impact characteristics; These are the fusion weight coefficients corresponding to the acoustic features; These are the fusion weighting coefficients corresponding to the vibration intensity characteristics; satisfy The weighting coefficients can be determined according to the equipment type during the initial calibration stage, and iterative updates based on fault samples are supported. This represents the real-time kurtosis value of the vibration signal; The kurtosis reference value of the vibration signal stored in the health baseline map; This represents the real-time total sound pressure level of the acoustic signal; The sound pressure level reference value stored in the health baseline map; This represents the real-time vibration intensity (RMS) of the vibration signal. The vibration intensity benchmark values ​​are stored in the health baseline map.

[0077] Calculation of cross-sectional difference rate: After each motion command is executed, the lateral comparison analysis module calculates the lateral difference rate of each motor on the same platform: ; in, For the first Lateral difference rate of the actuator cylinder motor; Number the actuator cylinder motor on the same motion platform; Under the same motion command, the first Real-time calculated values ​​of key characteristic quantities of motor No. 1 (such as RMS current, response time, etc.). This is the real-time arithmetic mean of the key characteristic quantities corresponding to all actuator cylinder motors on the same motion platform under the same motion command.

[0078] Matching threshold judgment symbols: Adaptive threshold : Determined dynamically by the rolling standard deviation of the historical LDR time series of the motor; This is the threshold confidence coefficient, with a default value of 2. Judgment logic: when When this occurs, the lateral abnormality flag for the motor is triggered.

[0079] Degradation Integral Update and RUL Prediction: The degradation integral and lifetime prediction module updates the degradation integral after the training task ends: ; in, for The constantly updated integral value of motor degradation, which increases monotonically with running time; For the last update The initial value of the motor degradation integral; The update time step for the degenerate integral is set to be executed by default after the end of a single training task; To perform a summation operation on all load time periods within this update cycle; This is the load time period number within this update cycle, with a value range of 1~ ; This represents the total number of load time periods divided within this update cycle; For the first The normalized load intensity borne by the motor within a time period, with a value range of [0,1]; The material equivalent fatigue index is used to describe the nonlinear relationship between load strength and fatigue degradation, and is calibrated by equipment manufacturer data or historical failure statistics. For the first The duration of operation corresponding to each load intensity.

[0080] The remaining usable lifetime is predicted using the following formula: ; in, The remaining usable life of the motor, expressed in training flight hours or calendar days; A preset degradation integral threshold for when the motor performance reaches an unacceptable degradation state; for Real-time degradation integral calculation value of the motor at any given moment; The current degradation rate of the motor is calculated from recent continuous data. DI The time series is obtained through linear fitting and represents the growth rate of the degradation integral per unit time.

[0081] Active stimulus detection: During the training completion and reset process, the active excitation detection module is injected into AEDS and executes the following sequentially: full-stroke low-speed scan (speed is 20% of the rated speed, covering 0% to 100% of the stroke); step response test (step position commands at multiple stroke positions); constant torque loading test (applying rated load at a specified stroke point and holding it). All sensor data from the above processes are collected, and health indicators such as response time, overshoot, steady-state tracking error, and full-stroke ZDI distribution are calculated to generate periodic health monitoring reports.

[0082] Tiered early warning and output: The early warning management module triggers early warnings and pushes them to the maintenance terminal according to the following hierarchical logic. Simultaneously, all early warning events, detection data, and maintenance records are written to the electronic health record.

[0083] The Abnormal Resonance Index (ARI) is calculated by fusing the following three types of features: Vibration characteristics: Vibration intensity (RMS) reflects the total amount of vibration energy; The Kurtosis value reflects the characteristics of impact transient vibration and is sensitive to degradation modes such as early pitting and raceway defects. The energy ratio of a specified frequency band is used to extract the proportion of abnormal frequency band energy related to the inherent frequency of the motor's mechanical structure.

[0084] Acoustic characteristics: Overall sound pressure level (SPL) reflects the overall intensity of abnormal noise; Abnormal frequency component amplitudes are identified by FFT, which identifies the amplitudes of abnormal frequency components that deviate from the normal spectral template. Acoustic envelope modulation index, used to detect the amplitude modulation characteristics of acoustic signals (related to mechanical loosening, gear wear, etc.).

[0085] Fusion method: Weighted linear fusion is used, with weight coefficients... , , During the initial calibration phase of the system, the values ​​are jointly determined by expert knowledge and historical samples, and iterative updates based on accumulated fault samples are supported. The ARI value range is normalized to [0,10], with ARI≥1.5 indicating mild anomaly, ARI≥3.0 indicating moderate anomaly, and ARI≥5.0 indicating severe anomaly.

[0086] Among them, the early warning level is defined as follows:

[0087] The Healthy Baseline Map (HBM) is stored using a three-dimensional matrix structure: ; in, The three-dimensional health baseline map of a single motor is stored using a three-dimensional matrix structure; The trip is divided into 20 sub-intervals (each sub-interval accounts for 5% of the total trip), with values ​​ranging from 1 to 20. The load level is numbered, with each level divided into 20% of the rated load, for a total of 5 levels, with values ​​ranging from 1 to 5. Number the feature dimensions, covering no less than 5 feature dimensions such as current RMS, current peak value, vibration intensity, kurtosis value, and sound pressure level; For the corresponding Itinerary zoning, Load level, Under the feature dimension, the initial baseline mean of the feature quantity; For the corresponding Itinerary zoning, Load level, Under the feature dimension, the initial baseline standard deviation of the feature quantity; This represents the number of valid data samples used to calculate the baseline within the corresponding index unit. This is the most recent update timestamp of the baseline data for this index unit.

[0088] The baseline is updated using an exponentially weighted moving average (EWMA) on a rolling basis, with the following update formula: ; in, This is the updated baseline mean; The measured mean of the characteristic quantities under the newly added healthy state; The baseline mean before the update; This is a smoothing coefficient, ranging from 0.05 to 0.15, used to control the smoothness of baseline updates and prevent degraded data from polluting the baseline.

[0089] The Active Activated Detection Sequence (AEDS) consists of three standardized tests, with a total execution time not exceeding 5 minutes. It is executed during the reset process after training and does not consume additional training time. Section 1: Low-speed scan throughout the entire travel path Drive the actuator at a constant speed of 20% of the rated speed from the minimum stroke to the maximum stroke, and then return to the middle position. Collect the current and vibration data throughout the stroke, calculate the ZDI distribution map of the entire stroke, and identify the degradation range.

[0090] Section 2: Multi-point Step Response Test A step position command at rated speed is applied at three positions: 25%, 50%, and 75% of the travel. The response time, overshoot, and settling time at each point are recorded and compared with the initial baseline to identify dynamic response performance degradation.

[0091] Section 3: Constant Load Output Test Apply the rated load at 50% of the stroke and hold for 10 seconds. Record the average current and fluctuation during the holding period to evaluate the steady-state output capability under rated conditions. It is most sensitive to degradation due to insufficient torque.

[0092] The AEDS execution results generate a structured health assessment report, including: execution timestamp, characteristic values ​​for each test segment, deviation from the initial baseline, segmented health scores, and a comprehensive health index. The report is written into an electronic health record, supporting historical trend viewing.

[0093] Among them, the degradation parameter table is as follows:

[0094] Sensor installation requirements include:

[0095] The electronic health record data structure is as follows: Electronic health records are created with independent files based on device serial number (actuator serial number), and each file must contain at least the following data fields:

[0096] All archive data is accompanied by a record timestamp, and key fields (early warning events, maintenance records) are accompanied by the operator's digital signature, supporting integrity verification and anti-tampering auditing.

[0097] The dynamic maintenance window optimization module performs the following calculation process: enter: Current RUL Predictions and Lower Confidence Bounds ; External training scheduling plan (including the subject types, expected load intensity and time arrangement of each training task in the next N days); Expected downtime for maintenance work .

[0098] deal with: Calculate the RUL margin for each candidate maintenance window: ; in, Candidate maintenance time points The corresponding maintenance safety margin; the larger the margin, the higher the safety redundancy at the maintenance time point. Candidate repair time points; It serves as a confidence lower bound for the predicted Remaining Usable Life (RUL) value, used to mitigate safety risks arising from prediction errors. The current time for performing maintenance window optimization calculations; This represents the current degradation rate of the motor. The preset degradation integral threshold is used to define the unacceptable degradation state of motor performance.

[0099] Calculate the impact score of each candidate maintenance window on the training plan: ; in, Candidate maintenance time points The impact on the training plan is scored, with higher scores indicating a greater impact on training continuity. Candidate repair time points; To perform a summation operation on all discrete-time units within the candidate maintenance cycle; To cover all candidate repair time ranges that cover the lower bound of the RUL confidence; It is a discrete time unit on the time axis; for The importance weight of each training task is assigned at any time and can be configured according to the level and priority of the training task. This is a time overlap calculation function that outputs candidate repair windows and... The temporal overlap of the training tasks at each moment, with a value range of [0,1]. This indicates the estimated downtime required for this maintenance operation.

[0100] Overall score ranking: ; in, Candidate maintenance time points The higher the overall recommendation score, the higher the overall priority of the repair window. Candidate repair time points; These are the weighting coefficients for the safety margin dimension; To train the weight coefficients that influence the dimensions; satisfy It can be configured according to the training center's operation and maintenance strategy; Candidate maintenance time points Corresponding maintenance safety margin; Candidate maintenance time points Impact on training plan rating.

[0101] The top 3 candidate windows with the highest scores are selected as recommended maintenance time nodes, along with the margin value and impact score for each node for maintenance supervisors to use as a reference for decision-making.

[0102] Output: Recommended maintenance window list (including recommendation ranking, planned maintenance time, remaining safety margin, and impact on training plan score) and maintenance suggestion sheet (including warning level, description of degradation characteristics, suggested inspection / replacement parts, and reference maintenance man-hours).

[0103] For the purpose of simplicity, the method steps disclosed in the above embodiments are described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0104] like Figure 2 As shown, the present invention also provides a flight simulator motor degradation monitoring and early warning system, comprising: The health baseline map establishment module 201 is configured to establish a health baseline map indexed by stroke position range and load level, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in the health state; The partition degradation index calculation module 202 is configured to collect running data and extract feature parameters in real time during the operation of the simulator, compare the real-time feature parameters with the benchmark data of the corresponding index unit in the health baseline map, and calculate the partition degradation index of each travel partition. The abnormal noise quantification index generation module 203 is configured to collect vibration signals and acoustic signals in real time, extract vibration features and acoustic features, perform fusion calculation, and generate an abnormal noise quantification index. The adaptive threshold and predicted remaining usable life calculation module 204 is configured to synchronously collect key characteristic parameters of multiple actuator cylinder motors on the same motion platform under the same motion command, calculate the lateral difference rate and dynamically determine the adaptive threshold, and at the same time, determine the load intensity based on the historical records of the partition degradation index, calculate and update the degradation integral based on the load intensity and duration, and predict the remaining usable life based on the degradation integral. The early warning information output module 205 is configured to inject an active stimulus detection sequence at a preset trigger time, collect and analyze all data during the execution of the active stimulus detection sequence, generate a health detection report, and combine the partition degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report to execute hierarchical early warning logic and output early warning information.

[0105] It is worth noting that although only some basic functional modules are disclosed in the embodiments of this invention, it does not mean that the composition of this system is limited to the above-mentioned basic functional modules. On the contrary, what this embodiment intends to express is that, based on the above-mentioned basic functional modules, those skilled in the art can arbitrarily add one or more functional modules in combination with existing technology to form an infinite number of embodiments or technical solutions. That is to say, this system is open rather than closed. The fact that this embodiment only discloses a few basic functional modules should not be considered as the scope of protection of the claims of this invention being limited to the disclosed basic functional modules. At the same time, for the convenience of description, the above device is described separately according to its functions as various units and modules. Of course, in implementing this invention, the functions of each unit and module can be implemented in one or more software and / or hardware.

[0106] like Figure 3 As shown, the present invention also provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of a method for monitoring and warning of motor degradation in a flight simulator.

[0107] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. For example... Figure 3 The structure shown in this embodiment of the invention includes an electronic device comprising one or more processors 710 and a memory 720; the processors 710 in this electronic device may be one or more. Figure 3Taking a processor 710 as an example; a memory 720 is used to store one or more programs; the one or more programs are executed by the one or more processors 710, so that the one or more processors 710 implement a method for monitoring and early warning of motor degradation in a flight simulator as described in any one of the embodiments of the present invention.

[0108] The electronic device may also include an input device 730 and an output device 740.

[0109] The processor 710, memory 720, input device 730, and output device 740 in this electronic device can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0110] The memory 720 in this electronic device serves as a computer-readable storage medium, capable of storing one or more programs. These programs can be software programs, computer-executable programs, or modules, such as the program instructions / modules corresponding to the flight simulator motor degradation monitoring and early warning method provided in this embodiment of the invention. The processor 710 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 720, thereby implementing the flight simulator motor degradation monitoring and early warning method described in the above embodiment.

[0111] The memory 720 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 720 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 720 may further include memory remotely located relative to the processor 710, which can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0112] Input device 730 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the electronic device. Output device 740 may include display devices such as a display screen.

[0113] The present invention also provides a computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of a method for monitoring and warning of motor degradation in a flight simulator.

[0114] Specifically, the computer storage medium in this embodiment of the invention can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be—but is not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring and early warning of motor degradation in a flight simulator, characterized in that, include: A health baseline map is established with stroke position range and load level as indexes, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in a healthy state; During simulator operation, real-time operation data is collected and feature parameters are extracted. The real-time feature parameters are compared with the baseline data of the corresponding index unit in the health baseline map to calculate the partition degradation index of each travel partition. Real-time acquisition of vibration and acoustic signals, extraction of vibration and acoustic features, and fusion calculation to generate abnormal noise quantification index; Under the same motion command, key characteristic parameters of multiple actuator cylinder motors on the same motion platform are collected synchronously, the lateral difference rate is calculated and the adaptive threshold is dynamically determined. At the same time, the load intensity is determined according to the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral. An active stimulus detection sequence is injected at a preset trigger time. All data during the execution of the active stimulus detection sequence is collected and analyzed to generate a health detection report. The zonal degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report are combined to execute a graded early warning logic and output early warning information.

2. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 1, characterized in that, A health baseline map is established, indexed by stroke position range and load level. This health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor under healthy conditions, and further includes: When the actuator motor is in its initial healthy state, the actuator is controlled to move within its full stroke range and operate at different load levels. The drive current, stroke position, vibration and acoustic signals of the actuator cylinder motor are collected, and feature parameters of multiple dimensions are extracted. Among them, the feature parameters of multiple dimensions include at least the effective value of current, vibration intensity, kurtosis value and sound pressure level. Using the travel location range and load level as two-dimensional indexes, the mean and standard deviation of the extracted feature parameters from multiple dimensions are used as benchmark data and stored in a three-dimensional matrix structure to form a health baseline map. An exponentially weighted moving average method is used to continuously update the baseline data in the health baseline map based on newly added health status data.

3. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 2, characterized in that, During simulator operation, real-time operation data is collected and feature parameters are extracted. These real-time feature parameters are compared with the baseline data of the corresponding index units in the health baseline map to calculate the partition degradation index for each travel partition. This further includes: The entire stroke of the actuator is divided into multiple sub-sections, which are independent stroke zones; During simulator operation, feature parameters of multiple dimensions under the current travel partition are collected in real time. The feature parameters collected in real time from multiple dimensions are compared with the baseline data under the corresponding travel partition and load level in the health baseline map, and the normalized deviation of each feature parameter is calculated respectively. The normalized deviations of each feature parameter are weighted and summed to obtain the partition degradation index of the current travel partition.

4. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 1, characterized in that, Real-time acquisition of vibration and acoustic signals, extraction of vibration and acoustic features, and fusion calculation to generate an abnormal noise quantification index, further including: Real-time acquisition of vibration signals from the actuator cylinder and acoustic signals near the mounting point; The vibration intensity and kurtosis values ​​are extracted from the vibration signal, and the total sound pressure level is extracted from the acoustic signal; The extracted vibration intensity, kurtosis value, and total sound pressure level are compared with the corresponding vibration intensity reference value, kurtosis reference value, and sound pressure level reference value in the health baseline map, respectively, to obtain the normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation. The normalized vibration intensity deviation, normalized kurtosis deviation, and normalized sound pressure level deviation are multiplied by preset fusion weighting coefficients and then summed to generate an abnormal noise quantification index.

5. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 1, characterized in that, Under the same motion command, key characteristic parameters of multiple actuator cylinder motors on the same motion platform are synchronously collected, the lateral difference rate is calculated, and an adaptive threshold is dynamically determined. Simultaneously, the load intensity is determined based on the historical records of the partition degradation index, and the degradation integral is calculated and updated based on the load intensity and duration. The remaining usable life is predicted based on the degradation integral, further including: After each motion command is executed, the key characteristic parameters of all actuator cylinder motors on the same motion platform are obtained, and the arithmetic mean of the key characteristic parameters of all actuator cylinder motors is calculated. The key characteristic parameters include at least the effective value of the current and the response time. The lateral difference rate of the corresponding actuator motor is calculated based on the ratio of the difference between the key characteristic parameters of a single actuator motor and the arithmetic mean to the arithmetic mean, and an adaptive threshold is dynamically determined based on the rolling standard deviation of the historical lateral difference rate sequence of the corresponding actuator motor. Within an update cycle, the degradation integral increment is calculated based on the product of the normalized load intensity and duration borne by the motor, combined with the material equivalent fatigue index. The degradation integral increment is then added to the degradation integral value of the previous moment to obtain the updated degradation integral. The degradation rate is fitted based on the degradation integral values ​​at multiple consecutive times, and the remaining usable lifetime is predicted based on the ratio of the difference between the preset degradation integral threshold and the current degradation integral value to the degradation rate.

6. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 1, characterized in that, An active stimulus detection sequence is injected at a preset trigger time, and all data during the execution of the active stimulus detection sequence is collected and analyzed to generate a health detection report, further including: During the reset process after training or when a manual trigger command is received, an active excitation detection sequence is generated, which includes full-stroke low-speed scanning, step response test and constant load output test. The speed of the full-stroke low-speed scanning is a preset percentage of the rated speed. The active excitation detection sequence is injected into the motion control system to control the actuator to perform actions according to the sequence. During the sequence execution, the current, stroke position, vibration and acoustic signals of the actuator are collected synchronously, and the response time, overshoot, steady-state tracking error and full stroke zone degradation index distribution are calculated. The response time, overshoot, steady-state tracking error, and full-stroke partition degradation index distribution are compared with the initial baseline to generate a structured health test report containing execution timestamps, test feature values ​​for each segment, deviation, and comprehensive health index.

7. The method for monitoring and early warning of motor degradation in a flight simulator according to claim 1, characterized in that, Based on the aforementioned zonal degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring report, a tiered early warning logic is executed and early warning information is output, further including: Real-time reception and integration of regional degradation index, abnormal noise quantification index, horizontal difference rate, remaining usable life, and health monitoring reports from active stimulation detection; The received indicators are compared with the trigger conditions of multiple preset warning levels. The trigger conditions include at least the partition degradation index exceeding the baseline standard deviation multiple, the abnormal noise quantification index exceeding the threshold, and the remaining usable lifespan being lower than the preset alarm time limit. When any indicator meets the triggering conditions for any warning level, a warning event corresponding to the corresponding warning level is generated, wherein the warning event includes the warning level, device identifier and description of the triggering conditions; The warning event is pushed to the maintenance terminal, and the warning event and the associated feature parameter snapshot are written into the electronic health record. At the same time, based on the remaining usable life prediction results and training plan information, a dynamic maintenance window recommendation is calculated and output.

8. A monitoring and early warning system for motor degradation in a flight simulator, characterized in that, include: The health baseline map establishment module is configured to establish a health baseline map indexed by stroke position range and load level, wherein the health baseline map stores baseline data of multiple dimensions of characteristic parameters of the actuator motor in a healthy state; The partition degradation index calculation module is configured to collect running data and extract feature parameters in real time during the operation of the simulator, compare the real-time feature parameters with the benchmark data of the corresponding index unit in the health baseline map, and calculate the partition degradation index of each travel partition. The abnormal noise quantification index generation module is configured to collect vibration and acoustic signals in real time, extract vibration and acoustic features, perform fusion calculations, and generate an abnormal noise quantification index. The adaptive threshold and predicted remaining usable life calculation module is configured to synchronously collect key characteristic parameters of multiple actuator cylinder motors on the same motion platform under the same motion command, calculate the lateral difference rate and dynamically determine the adaptive threshold. At the same time, it determines the load intensity based on the historical records of the partition degradation index, calculates and updates the degradation integral based on the load intensity and duration, and predicts the remaining usable life based on the degradation integral. The early warning information output module is configured to inject an active stimulus detection sequence at a preset trigger time, collect and analyze all data during the execution of the active stimulus detection sequence, generate a health detection report, and combine the partition degradation index, abnormal sound quantification index, horizontal difference rate, remaining usable life and health detection report to execute hierarchical early warning logic and output early warning information.

9. An electronic device, characterized in that, include: The system includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method according to any one of claims 1 to 7.

Images