Wave impact intelligent locking protection method and system based on artificial intelligence

Abstract

The invention discloses a wave impact intelligent locking protection method and system based on artificial intelligence, and belongs to the technical field of wave energy. Collecting wave oscillation waveform data, identifying wave crest time nodes, dividing time periods, monitoring a reciprocating vibration state of a vibration body and fitting a waveform; extracting a negative vibration characteristic waveform, and quantifying the observation sampling frequency of the pressure monitoring sensor; capturing a load pressure value and a time node at a negative half waveform trough, and constructing a dual-scale wavelet window model to quantify a high-frequency energy degree; load sampling frequency is calculated based on the high-frequency energy degree, and the sampling frequency is calibrated through the intelligent locking protection model and fed back to the sensor. According to the method, the time domain and energy domain characteristics of the vibration frequency are combined, the problems that in the prior art, a sampling strategy is rigid, characteristic analysis is one-sided, and a calibration mechanism is lacked are effectively solved, and the adaptability and efficiency of intelligent locking protection in the wave impact environment are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wave energy, in particular to a wave impact intelligent locking protection method and system based on artificial intelligence. BACKGROUND

[0002] In the field of ocean engineering, port facilities and ship equipment, the dynamic load of wave impact on the vibration body and its base is an important factor leading to structural fatigue and damage. In the prior art, the monitoring and protection system for wave impact usually monitors the vibration body load pressure with a fixed sampling frequency, which is difficult to adapt to the periodicity, nonlinearity and uneven energy distribution of wave impact in real time.

[0003] Specifically, the traditional method has the following technical problems: Single sampling strategy: based on fixed time interval sampling method, the sampling frequency cannot be dynamically adjusted according to the energy characteristics of wave impact (such as high frequency impact area), which may cause data loss in high energy impact scene or resource waste in low energy scene; Incomplete feature extraction: only relying on time domain features of vibration body vibration (such as vibration frequency), ignoring the influence of high frequency energy distribution in negative vibration waveform on load pressure, it is difficult to accurately capture the instantaneous high intensity load of wave impact; Insufficient intelligent calibration capability: lacking adaptive calibration mechanism for sampling frequency of pressure monitoring sensor, unable to dynamically optimize monitoring strategy combined with real-time load pressure and reference value, resulting in limited response accuracy and efficiency of locking protection system. SUMMARY

[0004] The purpose of the present application is to provide a wave impact intelligent locking protection method and system based on artificial intelligence to solve the problems raised in the background.

[0005] In order to solve the above technical problems, the present application provides the following technical scheme: A wave impact intelligent locking protection system based on artificial intelligence, the system comprises: data acquisition module, feature extraction module, frequency calculation module and intelligent calibration control module; The data acquisition module is used for collecting wave oscillation waveform data and vibration body reciprocating vibration state, generating reciprocating waveform of vibration body vibration, collecting load pressure value of vibration body acting on base through pressure monitoring sensor; The feature extraction module captures the load pressure value and time node at the negative half wave trough based on the negative vibration characteristic waveform, constructs a double scale wavelet window model, and quantifies the high frequency energy degree of each negative half wave; The frequency calculation module calculates the observation sampling frequency based on the reciprocating waveform, and calculates the load sampling frequency based on the high frequency energy degree; The intelligent calibration control module is used to construct an intelligent locking protection model to calculate the calibration sampling frequency and feed it back to the pressure monitoring sensor to adjust the sampling strategy.

[0006] Furthermore, the data acquisition module includes a wave monitoring unit, a vibration monitoring unit, and a pressure monitoring unit; The wave monitoring unit is used to collect wave oscillation waveform data, identify periodic wave peak time nodes, and divide the sampling period. The vibration monitoring unit is used to monitor the reciprocating vibration state of the vibrating body during the sampling period, and to fit the vibration amplitude to form a reciprocating waveform with the non-impact static position as the reference plane. The pressure monitoring unit is used to collect the load pressure value of the vibrating body acting on the base.

[0007] Furthermore, the feature extraction module includes a negative half-wave processing unit, a dual-scale window construction unit, and an energy calculation unit; The negative half-wave processing unit is used to extract the negative envelope of the reciprocating waveform, obtain the negative half-wave, and count the number of negative half-waves within the sampling period. The dual-scale window construction unit is used to set the time node scale and the load pressure scale, construct a rectangular window and map it to the negative half waveform, and determine the time and pressure value range of the load pressure value within the window. The energy measurement unit is used to calculate the ratio of the overlapping area of ​​the rectangular window and the negative half-wave to the total area of ​​the window, so as to obtain the high-frequency energy measurement to characterize the energy distribution characteristics of the negative half-wave.

[0008] Furthermore, the frequency calculation module includes an observation frequency calculation unit and a load frequency calculation unit; The observation frequency calculation unit is used to calculate the observation sampling frequency based on the sampling period duration and the number of negative half-waves. The load frequency calculation unit calculates the load sampling frequency by weighting the time interval between adjacent troughs based on high-frequency energy.

[0009] Furthermore, the intelligent calibration control module includes a weight adaptive unit, a fusion calibration unit, and a feedback adjustment unit; The weighted adaptive unit dynamically allocates the fusion weights of the observation sampling frequency and the load sampling frequency based on the real-time load pressure value and the benchmark load pressure value. The fusion calibration unit is used to dynamically adjust the calibration sampling frequency through an intelligent locking protection model; The feedback adjustment unit is used to feed back the calibration sampling frequency to the pressure monitoring sensor, driving the pressure monitoring sensor to collect the load pressure value according to the calibration sampling frequency.

[0010] An artificial intelligence-based wave impact intelligent locking protection method, the method comprises the following steps: S1: Collecting the oscillation waveform data of the wave, and identifying the time nodes representing the periodic wave peak characteristics in the oscillation waveform data; dividing the time period between adjacent wave peak time nodes, continuously monitoring the reciprocating vibration state of the vibrating body due to wave action in each time period, and fitting the reciprocating waveform of the vibrating body vibration; S2: Based on the time nodes of adjacent wave peaks in the wave oscillation waveform data, the reciprocating waveform of the vibrating body vibration is analyzed, the waveform representing the negative vibration characteristics of the vibrating body is extracted, and the observation sampling frequency of the pressure monitoring sensor in the cycle is quantified, the pressure monitoring sensor is used to collect the load pressure value generated when the vibrating body acts on the vibrating body base under the action of wave impact; S3: Based on the waveform of the negative vibration characteristics, the load pressure value collected by the pressure monitoring sensor at the negative half-wave trough and the time node triggered by the load pressure value are captured, and a double-scale wavelet window model is constructed to quantify the high-frequency energy degree of each negative half-wave; S4: Based on the high-frequency energy degree, the load sampling frequency of the pressure monitoring sensor in the cycle is quantified; an intelligent locking protection model is constructed, and based on the observation sampling frequency and the load sampling frequency, the calibration sampling frequency is calculated and fed back to the pressure monitoring sensor.

[0011] Further, the fitting method of the reciprocating waveform of the vibrating body vibration is as follows: Collect the oscillation waveform of the wave, and mark the time node of each wave peak in the oscillation waveform; Record the reciprocating behavior of the vibrating body vibration between the adjacent two wave peak time nodes, and represent the reciprocating behavior of the vibrating body vibration in the vibration two-dimensional coordinate system: Take the position plane of the vibrating body when it is stationary under the action of no wave impact as the reference plane, record the vibration amplitude of the vibrating body relative to the reference plane during the vibration process, and fit the vibration amplitude to the vibration two-dimensional coordinate system, wherein the horizontal coordinate of the vibration two-dimensional coordinate system corresponds to the time node, and the vertical coordinate of the vibration two-dimensional coordinate system corresponds to the vibration amplitude, to obtain the reciprocating waveform of the vibrating body vibration.

[0012] Further, the quantification method of the observation sampling frequency of the pressure monitoring sensor in the cycle is as follows: Let any adjacent two wave peak time nodes be and , wherein i is the label of the time node, and the sampling period of the pressure monitoring sensor between the adjacent two wave peak time nodes is ; The negative envelope extraction is performed on the reciprocating waveform of the vibration of the vibration body to obtain a negative half-waveform, and the number of samples of the negative half-waveform between two adjacent peak time nodes is counted and recorded as ; Based on the sampling period and the number of samples, the observation sampling frequency of the pressure monitoring sensor in the period is calculated .

[0013] Further, the quantification method of the high-frequency energy degree of the negative half-waveform is as follows: The load pressure value collected by the pressure monitoring sensor at the xth negative half-waveform trough in the sampling period and the time node triggered by the load pressure value are recorded as and respectively, where ; A double-scale wavelet window model is set, the double-scale including a lateral scale corresponding to the time node scale and a longitudinal scale corresponding to the load pressure scale , then a rectangular window is constructed through the double-scale, and the rectangular window is mapped into the negative half-waveform, then the lateral value range and the longitudinal value range of the rectangular window mapped in the xth negative half-waveform are and respectively, obtaining the rectangular window mapped in the xth negative half-waveform ; Based on the double-scale wavelet window model, the high-frequency energy degree of the xth negative half-waveform is calculated , wherein represents the overlapping area of the rectangular window and the xth negative half-waveform, represents the area of the rectangular window , and .

[0014] Further, the quantification method of the calibration sampling frequency is as follows: Based on the high-frequency energy degree, the load sampling frequency of the pressure monitoring sensor in the sampling period is calculated ; Based on the observation sampling frequency and the load sampling frequency , an intelligent locking protection model is constructed to calibrate the sampling frequency of the pressure monitoring sensor, and the calibration sampling frequency is calculated, wherein is a preset sensitivity parameter, is the load pressure value collected by the pressure monitoring sensor in real time, is a preset reference load pressure value; the calibration sampling frequency feedback to the pressure monitoring sensor.

[0015] In the above method, the observation sampling frequency can reflect the negative vibration frequency of the vibrating body within the sampling period, represent the time domain characteristics of the vibration frequency, the load sampling frequency is weighted by the high-frequency energy degree of the time interval of the load pressure concentration area, and represents the energy domain characteristics of the vibration frequency. In the present application, the observation frequency provides a basic quantization of the vibration frequency, the load frequency provides a frequency correction under high-energy impact, and the combination of the two can simultaneously capture the vibration regularity and impact strength.

[0016] Compared with the prior art, the present application has the beneficial effects that: in the wave impact intelligent locking protection method and system based on artificial intelligence provided by the present application, wave oscillation waveform data is collected, wave peak time nodes are identified and time periods are divided, the reciprocating vibration state of the vibrating body is monitored and the waveform is fitted; the negative vibration characteristic waveform is extracted, the observation sampling frequency of the pressure monitoring sensor is quantized; the load pressure value and the time node at the negative half-wave trough are captured, and a double-scale wavelet window model is constructed to quantize the high-frequency energy degree; the load sampling frequency is calculated based on the high-frequency energy degree, and the sampling frequency is calibrated by the intelligent locking protection model and fed back to the sensor. The system includes data acquisition, feature extraction, frequency calculation and intelligent calibration control modules, and realizes dynamic monitoring and sampling strategy optimization of the load pressure of the vibrating body under wave impact. The present application combines the time domain and energy domain characteristics of the vibration frequency, and improves the adaptability and efficiency of the intelligent locking protection under wave impact environment. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.

[0018] Figure 1 is a step schematic diagram of a wave impact intelligent locking protection method based on artificial intelligence of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0020] In the first embodiment: a wave impact intelligent locking protection system based on artificial intelligence is provided, which comprises: a data acquisition module, a feature extraction module, a frequency calculation module and an intelligent calibration control module. The data acquisition module is configured to acquire wave oscillation waveform data and a reciprocating vibration state of the vibrating body, generate a reciprocating waveform of the vibrating body vibration, and acquire a load pressure value of the vibrating body acting on the base through the pressure monitoring sensor. The data acquisition module includes a wave monitoring unit, a vibration monitoring unit, and a pressure monitoring unit. The wave monitoring unit is configured to acquire wave oscillation waveform data, identify periodic wave peak time nodes, and divide sampling periods. The vibration monitoring unit is configured to monitor the reciprocating vibration state of the vibrating body within the sampling period, take the impact-free static position as a reference plane, and fit the vibration amplitude to form a reciprocating waveform. The pressure monitoring unit is configured to acquire a load pressure value of the vibrating body acting on the base. The feature extraction module is configured to capture the load pressure value and time node at the negative half-waveform trough based on the negative vibration characteristic waveform, construct a bi-scale wavelet window model, and quantify the high-frequency energy degree of each negative half-waveform. The feature extraction module includes a negative half-wave processing unit, a bi-scale window construction unit, and an energy degree calculation unit. The negative half-wave processing unit is configured to perform negative envelope extraction on the reciprocating waveform to obtain a negative half-waveform and count the number of negative half-waveforms within the sampling period. The bi-scale window construction unit is configured to set the time node scale and the load pressure scale, construct a rectangular window, map it to the negative half-waveform, and determine the time and pressure value range of the load pressure value within the window. The energy degree calculation unit is configured to calculate the ratio of the overlapping area of the rectangular window and the negative half-waveform to the total area of the window to obtain a high-frequency energy degree to represent the energy distribution characteristics of the negative half-waveform. The frequency calculation module is configured to calculate an observation sampling frequency based on the reciprocating waveform and a load sampling frequency based on the high-frequency energy degree. The frequency calculation module includes an observation frequency calculation unit and a load frequency calculation unit. The observation frequency calculation unit is configured to calculate the observation sampling frequency based on the sampling period length and the number of negative half-waveforms. The load frequency calculation unit is configured to calculate the load sampling frequency based on the high-frequency energy degree and the weighted adjacent trough time interval. The intelligent calibration control module is configured to construct an intelligent locking protection model to calculate a calibration sampling frequency and feed back to the pressure monitoring sensor to adjust the sampling strategy. The intelligent calibration control module includes a weight self-adaptive unit, a fusion calibration unit, and a feedback adjustment unit. The weight self-adaptive unit is configured to dynamically allocate the fusion weight of the observation sampling frequency and the load sampling frequency based on the real-time load pressure value and the reference load pressure value. The fusion calibration unit is configured to dynamically adjust the calibration sampling frequency by the intelligent locking protection model. The feedback adjustment unit is configured to feed back the calibration sampling frequency to the pressure monitoring sensor, and drive the pressure monitoring sensor to collect the load pressure value at the calibration sampling frequency.

[0021] Please refer to Figure 1 In the second embodiment, a wave impact intelligent locking protection method based on artificial intelligence is provided, which is applicable to the first embodiment. The method comprises the following steps: S1: Collecting the oscillation waveform data of the wave and identifying the time nodes representing the periodic wave peak characteristics in the oscillation waveform data; dividing the time period by adjacent wave peak time nodes, continuously monitoring the reciprocating vibration state of the vibrating body caused by the wave in each time period, and fitting the reciprocating waveform of the vibrating body vibration; For example, the fitting method of the reciprocating waveform of the vibrating body vibration is as follows: Collecting the oscillation waveform of the wave and marking the time node of each wave peak in the oscillation waveform; Recording the reciprocating behavior of the vibrating body vibration between the adjacent two wave peak time nodes, and representing the reciprocating behavior of the vibrating body vibration in the vibration two-dimensional coordinate system: Taking the position plane of the vibrating body when it is stationary under the action of the wave impact as the reference plane, recording the vibration amplitude of the vibrating body relative to the reference plane during the vibration, and fitting the vibration amplitude to the vibration two-dimensional coordinate system, wherein the horizontal coordinate of the vibration two-dimensional coordinate system corresponds to the time node, and the vertical coordinate of the vibration two-dimensional coordinate system corresponds to the vibration amplitude, to obtain the reciprocating waveform of the vibrating body vibration.

[0022] S2: Based on the time nodes of adjacent wave peaks in the wave oscillation waveform data, analyzing the reciprocating waveform of the vibrating body vibration, extracting the waveform representing the negative vibration characteristics of the vibrating body, and quantifying the observation sampling frequency of the pressure monitoring sensor in the cycle, the pressure monitoring sensor is used to collect the load pressure value generated when the vibrating body acts on the vibrating body base under the action of the wave impact; For example, the quantification method of the observation sampling frequency of the pressure monitoring sensor in the cycle is as follows: Let any two adjacent wave peak time nodes be and , wherein i is the label of the time node, and the sampling period of the pressure monitoring sensor between the adjacent two wave peak time nodes is ; The negative envelope of the reciprocating waveform of the vibrating body vibration is extracted to obtain the negative half waveform, and the number of samples of the negative half waveform between the adjacent two wave peak time nodes is counted, denoted as ; Based on the sampling period and the sampling number, the observation sampling frequency of the pressure monitoring sensor in the period is calculated .

[0023] S3: Based on the waveform of the negative vibration feature, the load pressure value collected by the pressure monitoring sensor at the negative half-wave trough and the time node triggered by the load pressure value are captured, and a double-scale wavelet window model is constructed to quantify the high-frequency energy degree of each negative half-wave; For example, the quantification method of the high-frequency energy degree of the negative half-wave is as follows: Capture the load pressure value collected by the pressure monitoring sensor at the xth negative half-wave trough in the sampling period , and the time node triggered by the load pressure value, which are sequentially recorded as and , respectively, where ; Set the double-scale wavelet window model, which includes the lateral scale and the longitudinal scale. The lateral scale corresponds to the time node scale , and the longitudinal scale corresponds to the load pressure scale . Then a rectangular window is constructed through the double scale, and the rectangular window is mapped into the negative half-wave. Then the lateral value range and the longitudinal value range of the rectangular window mapped in the xth negative half-wave are sequentially recorded as and , respectively, to obtain the rectangular window mapped in the xth negative half-wave ; Based on the double-scale wavelet window model, the high-frequency energy degree of the xth negative half-wave is calculated , where represents the overlapping area of the rectangular window and the xth negative half-wave, represents the area of the rectangular window , and .

[0024] S4: Based on the high-frequency energy degree, the load sampling frequency of the pressure monitoring sensor in the period is quantified; an intelligent locking protection model is constructed, and based on the observation sampling frequency and the load sampling frequency, the calibration sampling frequency is calculated and fed back to the pressure monitoring sensor; For example, the quantification method of the calibration sampling frequency is as follows: Based on the high-frequency energy degree, the load sampling frequency of the pressure monitoring sensor in the sampling period is calculated ; ; Based on the observation sampling frequency and the load sampling frequency , an intelligent locking protection model is constructed to calibrate the sampling frequency of the pressure monitoring sensor, and the calibration sampling frequency is calculated, where a sensitivity parameter preset, a load pressure value collected by the pressure monitoring sensor in real time, a reference load pressure value preset; the calibration sampling frequency is fed back to the pressure monitoring sensor; For example, for high load impact, much greater than At this time can approach 0, and the load sampling frequency dominates (focuses on high-frequency impact signals), At the same time for calibrating the load sampling frequency, the damping coefficient of the vibration body base can be instructed to increase to achieve impact resistance while achieving high-frequency sampling. For low load impact, much less than , can approach 1, and the observation sampling frequency dominates (optimizes low-frequency scene sampling efficiency), At the same time for calibrating the observation sampling frequency, the damping coefficient of the vibration body base can be instructed to decrease.

[0025] It should be noted that, in this document, the terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0026] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An artificial intelligence-based wave impact intelligent locking protection method, characterized in that, The method comprises the following steps: S1: Collecting oscillation waveform data of waves, and identifying time nodes representing periodic wave peak characteristics in the oscillation waveform data; dividing time periods by adjacent wave peak time nodes, continuously monitoring the reciprocating vibration state of the vibrating body caused by wave action in each time period, and fitting the reciprocating waveform of the vibration of the vibrating body; S2: Based on the time nodes of adjacent wave peaks in the oscillation waveform data of waves, analyzing the reciprocating waveform of the vibration of the vibrating body, extracting a waveform representing the negative vibration characteristics of the vibrating body, and quantifying the observation sampling frequency of the pressure monitoring sensor in the cycle, the pressure monitoring sensor is used to collect the load pressure value generated when the vibrating body acts on the vibrating body base under the impact of waves; S3: Based on the waveform of the negative vibration characteristics, capturing the load pressure value collected by the pressure monitoring sensor at the negative half-wave valley and the time node triggered by the load pressure value, and constructing a double-scale wavelet window model to quantify the high-frequency energy degree of each negative half-wave; S4: Based on the high-frequency energy degree, the load sampling frequency of the pressure monitoring sensor in the cycle is quantified; an intelligent locking protection model is constructed, and the calibration sampling frequency is calculated based on the observation sampling frequency and the load sampling frequency and fed back to the pressure monitoring sensor.

2. The wave impact intelligent locking protection method based on artificial intelligence according to claim 1, characterized in that, The fitting method of the reciprocating waveform of the vibration of the vibrating body is as follows: Collect the oscillation waveform of the wave, and mark the time node of each wave peak in the oscillation waveform; Record the reciprocating behavior of the vibrating body between the two adjacent wave peak time nodes, and represent the reciprocating behavior of the vibrating body in the vibration two-dimensional coordinate system: Taking the position plane of the vibrating body when it is stationary under the action of no wave impact as the reference plane, record the vibration amplitude of the vibrating body relative to the reference plane during the vibration process, and fit the vibration amplitude to the vibration two-dimensional coordinate system, wherein the horizontal coordinate of the vibration two-dimensional coordinate system corresponds to the time node, and the vertical coordinate of the vibration two-dimensional coordinate system corresponds to the vibration amplitude, to obtain the reciprocating waveform of the vibration of the vibrating body.

3. The intelligent wave impact locking protection method based on artificial intelligence according to claim 1, characterized in that, The quantification method of the observation sampling frequency of the pressure monitoring sensor in the cycle is as follows: Any two adjacent wave crest time nodes are respectively denoted as and wherein i is the label of the time node, and a sampling period of the pressure monitoring sensor is formed between the two adjacent wave crest time nodes, denoted as ; The negative half wave form is obtained by extracting a negative envelope of the reciprocating wave form of the vibration of the vibration body, and the number of samples of the negative half wave form between two adjacent wave peak time nodes is counted and recorded as ; Based on the sampling period and the number of samples, calculate the observation sampling frequency of the pressure monitoring sensor within the period .

4. The wave impact intelligent locking protection method based on artificial intelligence according to claim 3, characterized in that, The quantification method of the high-frequency energy degree of the negative half-wave is as follows: captured in a sampling period The load pressure value collected by the pressure monitoring sensor at the trough of the xth negative half-wave form and the time node triggered by the load pressure value are sequentially recorded as and wherein, ; A double scale wavelet window model is set, the double scale including a transverse scale and a longitudinal scale, the transverse scale corresponding to a time node scale , and the longitudinal scale corresponding to a load pressure scale , a rectangular window is constructed through the double scale, and the rectangular window is mapped into a negative half wave form, the transverse value range and the longitudinal value range of the rectangular window mapped in the xth negative half wave form are respectively and , and the rectangular window mapped in the xth negative half wave form is obtained . Based on the two-scale wavelet window model, the high frequency energy degree of the xth negative half wave form is calculated , wherein denotes a rectangular window , wherein denotes a rectangular window , and wherein .

5. The intelligent wave impact locking protection method based on artificial intelligence according to claim 4, characterized in that, The quantification method of the calibration sampling frequency is as follows: Based on the high frequency energy measure, the sampling period is calculated Load sampling frequency of the internal pressure monitoring sensor ; Based on the observation sampling frequency And the load sampling frequency , construct an intelligent locking protection model, calibrate the sampling frequency of the pressure monitoring sensor, calculate the calibrated sampling frequency , wherein, The sensitivity parameter is a preset value, The load pressure value collected by the pressure monitoring sensor in real time, The reference load pressure value is a preset value; the calibrated sampling frequency Is fed back to the pressure monitoring sensor.

6. An artificial intelligence-based wave impact intelligent locking protection system, which executes the wave impact intelligent locking protection method according to any one of claims 1-5, characterized in that, The system comprises a data acquisition module, a feature extraction module, a frequency calculation module, and an intelligent calibration control module; The data acquisition module is used to collect wave oscillation waveform data and vibrating body reciprocating vibration state, generate reciprocating waveform of the vibration of the vibrating body, and collect load pressure value of the vibrating body acting on the base through the pressure monitoring sensor; The feature extraction module captures the load pressure value and time node at the negative half-wave valley based on the negative vibration characteristic waveform, constructs a double-scale wavelet window model, and quantifies the high-frequency energy degree of each negative half-wave; The frequency calculation module calculates the observation sampling frequency based on the reciprocating waveform, and calculates the load sampling frequency based on the high-frequency energy degree; The intelligent calibration control module is used to construct an intelligent locking protection model to calculate the calibration sampling frequency and feed back to the pressure monitoring sensor to adjust the sampling strategy.

7. The wave impact intelligent locking protection system based on artificial intelligence according to claim 6, characterized in that, The data acquisition module comprises a wave monitoring unit, a vibration monitoring unit, and a pressure monitoring unit; The wave monitoring unit is configured to collect wave oscillation waveform data, identify periodic wave peak time nodes, and divide sampling periods; The vibration monitoring unit is configured to monitor the reciprocating vibration state of the vibrating body in the sampling period, take the impact-free static position as a reference plane, and fit the vibration amplitude to form a reciprocating waveform; The pressure monitoring unit is configured to collect load pressure values of the vibrating body acting on the base.

8. The wave impact intelligent locking protection system based on artificial intelligence according to claim 6, characterized in that, The feature extraction module includes a negative half-wave processing unit, a double-scale window construction unit, and an energy degree calculation unit; The negative half-wave processing unit is configured to extract a negative envelope from the reciprocating waveform to obtain a negative half-waveform and count the number of negative half-waves in the sampling period; The double-scale window construction unit is configured to set time node scales and load pressure scales, construct a rectangular window, map it to the negative half-waveform, and determine the time and pressure value range of the load pressure values in the window; The energy degree calculation unit is configured to calculate the ratio of the overlapping area of the rectangular window and the negative half-waveform to the total area of the window to obtain a high-frequency energy degree to represent the energy distribution characteristics of the negative half-waveform.

9. The wave impact intelligent locking protection system based on artificial intelligence according to claim 6, characterized in that, The frequency calculation module includes an observation frequency calculation unit and a load frequency calculation unit; The observation frequency calculation unit is configured to calculate an observation sampling frequency based on the sampling period length and the number of negative half-waves; The load frequency calculation unit is configured to calculate a load sampling frequency based on the high-frequency energy degree and weighting adjacent trough time intervals.

10. The wave impact intelligent locking protection system based on artificial intelligence according to claim 6, characterized in that, The intelligent calibration control module includes a weight adaptive unit, a fusion calibration unit, and a feedback adjustment unit; The weight adaptive unit is configured to dynamically allocate the fusion weights of the observation sampling frequency and the load sampling frequency based on the real-time load pressure value and the reference load pressure value; The fusion calibration unit is configured to dynamically adjust the calibration sampling frequency through an intelligent locking protection model; The feedback adjustment unit is configured to feed back the calibration sampling frequency to the pressure monitoring sensor to drive the pressure monitoring sensor to collect load pressure values at the calibration sampling frequency.

Images