Marine launching safety real-time prediction method

By establishing a dynamic model and proxy model of the offshore launch system and using sensor data for frequency and phase analysis, the problem of complexity of the dynamic effect of the offshore launch platform is solved, real-time safety prediction of different platforms and device locations is achieved, and the improvement of Safety of launch missions.

CN119004984APending Publication Date: 2024-11-22BEIJING INST OF TECH
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411105158.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The six-degree of freedom motion of the offshore launch platform leads to a complex coupling dynamic effect, affecting the emission safety characteristic value. The existing technology is difficult to adapt to different launch platforms and device locations, and real-time safety prediction cannot be achieved.

Method used

Establish a dynamic model of the offshore emission system, obtain ship motion data through sampling, train agent models, use sensors to collect motion data, perform fast Fourier transformation, extract frequency, period and phase, judge the emission safety characteristic value, and realize the use of different platforms. and device location adaptation and real-time security prediction.

Benefits of technology

Real-time and accurate safety judgment of sea launches is achieved, the safety prediction accuracy of the launch mission is improved, and different launch platforms and device locations are adapted to ensure the safety of the launch process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119004984A_ABST
    Figure CN119004984A_ABST
Patent Text Reader

Abstract

The invention provides an offshore launching safety real-time prediction method, and belongs to the technical field of launching. The specific process of the method is as follows: establishing a marine launching system dynamics model, obtaining a sample database, training an agent model, collecting ship motion data and performing safe launching prediction, when the sample database is obtained, a sampling range is determined, sampling is performed from the sampling range to obtain sampling points, and a ship motion result of each sampling point is calculated; and taking the ship motion result as an input load boundary condition of the marine launching system dynamic model, obtaining the output of each sampling point to the model, and obtaining a sample database.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of launch technology, and in particular relates to a real-time prediction method for offshore launch safety. Background Art

[0002] Sea-based launch has the advantage of flexible site selection. It can reduce the fuel required for the launched system, the fuel consumption for satellite orbit adjustment, and avoid scattered objects falling in residential areas by selecting the launch site.

[0003] However, the launch platform will produce six-degree-of-freedom motion with periodic characteristics under the action of waves, which will cause complex coupled dynamic effects in the launch process and affect the characteristic value of launch safety. Different launch platforms have different rolling and pitching centers and the locations of launch devices on the launch platforms are also different. This leads to the lack of universality of motion application based on the rolling center when modeling the dynamics of the offshore launch system. The safety of offshore launch needs to be predicted based on the real-time status of the offshore launch system. Therefore, a technical solution is developed that can adapt to different launch platforms and launch device installation locations, and can predict the safety of offshore launch in real time. Summary of the invention

[0004] The purpose of the present invention is to provide a real-time prediction method for the safety of offshore launch, which can realize real-time and accurate judgment on the safety of offshore launch.

[0005] The technical solution for implementing the present invention is as follows:

[0006] A real-time prediction method for marine launch safety, the specific process is as follows:

[0007] Establish a dynamic model of the sea launch system: Establish a dynamic model of the sea launch system, set the model input load boundary conditions, and the model output is the launch safety characteristic value when the launched system completely leaves the launch tube;

[0008] Obtaining a sample database: determining a sampling range, sampling from it to obtain sampling points, and calculating the ship motion results at each sampling point; using the ship motion results as input load boundary conditions of the marine launch system dynamics model, obtaining the output of the model for each sampling point, and obtaining a sample database;

[0009] Training the proxy model: using the sample database, training the proxy model between the launch device motion boundary amplitude, frequency, phase and launch safety characteristic value;

[0010] Collecting ship motion data: Using sensors installed on the ship to collect the ship's motion data, extract the motion boundary amplitude, frequency and phase, and make validity judgments;

[0011] Safe launch prediction: Input the valid characteristic value into the proxy model to obtain the launch safety characteristic value of the launched system, and judge whether the launch safety characteristic value exceeds the safe launch range, so as to realize safe launch prediction.

[0012] Furthermore, the present invention establishes a dynamic model of the sea launch system, sets the model input load boundary conditions and model output, and the specific process is as follows:

[0013] Establish a collective model of the sea launch system, perform meshing, and set the parameters required for model dynamics calculations;

[0014] Model load setting: apply thrust to the bottom of the launched system, and apply displacement boundary conditions to the base and the part of the launcher that contacts the ship (i.e. the launch pad base and the crater);

[0015] Model contact setting: setting contact between the launched system and the launcher, between the launched system and the adapter, between the launched system and the base, and between the adapter and the guide rail on the launcher;

[0016] Model output settings: output the forces on the launched system during the launch process, displacement, velocity, acceleration in three directions, angular displacement, angular velocity, and angular acceleration in three directions; stress and strain of the adapter, launcher, and base; all contact forces.

[0017] Furthermore, the sampling range described in the present invention is: according to the maximum amplitude and maximum period of the six degrees of freedom motion of the ship corresponding to the maximum sea conditions that can safely launch at sea, the amplitude range during sampling is determined to be 0 to the maximum amplitude * expansion factor, and the sampling period is 0 to the maximum period * expansion factor. The expansion factor makes the sampling range include the maximum amplitude and maximum period that may be encountered during actual launch; the sampling phase range is determined to be 0 to 2π.

[0018] Furthermore, the sampling points of the present invention are composed of amplitude, period and phase, and the ship motion result of each sampling point is calculated. t is time, i takes values ​​of 1, 2, 3, 4, 5, 6, representing the six degrees of freedom of the ship, Ais represents the sampling amplitude, Tis represents the sampling period, Indicates the sampling phase.

[0019] Furthermore, the present invention obtains the output of the model corresponding to each sampling point, and extracts the contact force between the launched system and the launching tube in the output result. If the contact force is not equal to 0, it indicates that a collision has occurred, which is recorded as a collision situation. If the contact force is equal to 0, the launch safety characteristic value is extracted. The launch safety characteristic value is the angular velocity and angular displacement in three directions, the displacement in three directions, and the speed of the launched system when the launched system is completely out of the tube, thereby obtaining a sample library.

[0020] Furthermore, the launch safety characteristic values ​​of the present invention also include: the minimum gap between the launched system and the launch tube during the launch process, the force applied to the launched system during the launch process, and the stress and strain of the adapter, launch frame, and base in the launch system.

[0021] Furthermore, when training the proxy model described in the present invention, the R-squared, mean square error and / or mean absolute error method is used to judge the correctness of the proxy model. When it is judged that the proxy model training does not meet the correctness requirements, the number of sampling points is increased and further training is performed.

[0022] Furthermore, the specific process of collecting ship motion data in the present invention is as follows:

[0023] S4-1 uses three sets of displacement sensors and three sets of angle sensors arranged on the launch pad base or crater for continuous measurement;

[0024] S4-2 Fourier transform of motion curve in the past 3T seconds

[0025] At each moment, the measurement data of the past 3T seconds are subjected to fast Fourier transformation, where T is the period of the ship's maximum degree of freedom motion under the current sea conditions, to obtain the amplitude spectrum, power spectrum, phase spectrum, and intercept the signal with a frequency of 0-5 Hz;

[0026] S4-3 Past T Launch Fourier transform of motion curve in time

[0027] To the past T Launch Fast Fourier transform is performed in seconds to obtain the amplitude-frequency curve, power spectrum, phase spectrum, and signal with intercept frequency of 0-5 Hz. Launch It is the time from ignition of the launched system to its exit from the launch tube;

[0028] S4-4 Determination of Predictive Validity

[0029] Determine the relationship between the corresponding curves obtained by the two fast Fourier transforms. If the corresponding curves are similar, it is considered that the currently collected data is valid, and proceed to step S4-5;

[0030] S4-5 Data feature value extraction

[0031] The data characteristic values ​​are obtained by extracting the maximum amplitude, the frequency and the phase corresponding to the maximum amplitude from the amplitude-frequency curve of the six-degree-of-freedom motion measured by the sensor, thereby obtaining the predicted amplitude, phase and frequency of the motion in the six degrees of freedom directions.

[0032] Furthermore, the basis for judging whether the curves described in the present invention are similar is:

[0033] Calculate r based on the frequency of the two curves,

[0034]

[0035] Where n is the total number of discrete frequencies, y i3T is the i-th value of the amplitude curve obtained in step S4-2, y iTL is the i-th value of the amplitude curve obtained in step S4-3;

[0036] When r is less than the set threshold, it is considered that the corresponding curves are similar and the prediction is valid.

[0037] Furthermore, when r is less than 0.2, the present invention considers that the corresponding curves are similar and the prediction is valid.

[0038] Beneficial effects:

[0039] First, the present invention uses the establishment of a dynamic model of the sea launch system to obtain corresponding sample data, trains the proxy model, and uses the trained model to perform safety prediction. It can adapt to different launch platforms and launch device installation locations and can predict the safety of sea launch in real time, providing a reference for on-site launch missions.

[0040] Second, when establishing the sea launch dynamics model, the present invention applies displacement boundary conditions to the base and the part of the launch frame that contacts the ship, and uses a sine function to simulate. The periodicity of the sine function and the periodicity of the launch platform motion are matched. After that, the collected random motion of the launch platform is Fourier transformed to obtain information such as the amplitude-frequency curve. That is, which simple harmonic functions constitute the random motion. In this way, the simulated boundary and the actual measured motion boundary correspond to each other, thereby ensuring that the judgment result of this method has higher accuracy.

[0041] Third, when collecting ship motion data, the invention innovatively adopts a motion data feature extraction method, that is, converting from the time domain to the frequency domain and then extracting the frequency, period, and phase. There is no need to consider the relative position of the launch device and the launch platform, and it can adapt to different launch platforms.

[0042] Fourth, when extracting motion data features, the measured launch system motion data is subjected to fast Fourier transform to obtain the frequency domain information of the launch system motion data. By comparing the difference in frequency domain information over a long period of time and a short period of time, it is determined whether the launch platform motion has frequency domain stability and whether it has predictive value. Based on the above process, the accuracy of the launch safety characteristic value can be ensured.

[0043] Fifth, in the prior art, boundary conditions are imposed with the help of a rocking center. The rocking center is the point around which the launch platform rocks, but the rocking center of the roll and pitch is not in one place. Using a rocking center is a wrong method. And it is actually difficult to get an accurate rocking center. Correct simulation cannot be achieved. The present invention directly grasps the characteristics of the launch platform motion having a certain periodicity. By applying motion under the rocket / missile (i.e., the launch pad or crater), the correspondence between the simulation and the actual measurement points can be easily achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 is a schematic diagram of a sea launch system of a launched system according to an embodiment of the present application;

[0046] Figure 2 is a cross-sectional view of a launching device in a sea launching system of a launched system according to an embodiment of the present application;

[0047] Among them: 1-launching device; 2-launching platform; 3-launched system; 4-launching tube; 5-adapter; 6-support ring; 7-base. DETAILED DESCRIPTION

[0048] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0049] It should be noted that the following embodiments and features in the embodiments may be combined with each other in the absence of conflict; and, based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making any creative work are within the scope of protection of the present disclosure.

[0050] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0051] like Figure 1-2 As shown, the embodiment of the present application discloses a real-time prediction method for marine launch safety, comprising the following steps:

[0052] Establish a dynamic model of the sea launch system: Establish a dynamic model of the sea launch system, set the model input load boundary conditions, and the model output is the launch safety characteristic value when the launched system completely leaves the launch tube;

[0053] Obtaining a sample database: determining a sampling range, sampling from it to obtain sampling points, and calculating the ship motion results at each sampling point; using the ship motion results as input load boundary conditions of the marine launch system dynamics model, obtaining the output of the model for each sampling point, and obtaining a sample database;

[0054] Training the proxy model: using the sample database, training the proxy model between the launch device motion boundary amplitude, frequency, phase and launch safety characteristic value;

[0055] Collecting ship motion data: Using sensors installed on the ship to collect the ship's motion data, extract the motion boundary amplitude, frequency and phase, and make validity judgments;

[0056] Safe launch prediction: Input the valid characteristic value into the proxy model to obtain the launch safety characteristic value of the launched system, and judge whether the launch safety characteristic value exceeds the safe launch range, so as to realize safe launch prediction.

[0057] In the specific implementation of this embodiment, the dynamic modeling of the sea-launching system is carried out by using dynamic calculation software, applying boundary conditions, simulating the sea-launching process, and obtaining the launch safety characteristic value when the launched system completely leaves the launch tube. The motion boundary conditions can be approximated by simple harmonic functions and implemented by applying them to the bottom of the crater or launch pad.

[0058] In the specific implementation of this embodiment, the sampling objects are the amplitude, phase, and frequency of the six-degree-of-freedom motion input as the motion boundary conditions in the dynamic modeling of the marine launch system. The sample library is established by substituting the sampled values ​​into the marine launch system dynamic model for calculation to obtain the launch safety characteristic value corresponding to the sampling result;

[0059] In the specific implementation of this embodiment, the proxy model is established as a proxy model of the amplitude, phase, frequency and launch safety characteristic values ​​of the motion boundary conditions based on the sample library. The sampling range includes all possible situations of the launch mission, and the launch safety characteristic values ​​corresponding to the sampling points are calculated by the dynamics simulation of the marine launch system.

[0060] In the specific implementation of this embodiment, the motion data collection method refers to placing sensors on the launch platform or crater to measure the six-degree-of-freedom motion of the launch system.

[0061] The following is a detailed description of the specific process of each part:

[0062] S1. Dynamics modeling of sea launch system

[0063] According to the actual sea launch system, a dynamic simulation model is established to simulate the sea launch process of the launched system. The specific steps are as follows:

[0064] S1-1. Geometric modeling

[0065] According to the actual size, the geometric model of the sea launch system is established using 3D modeling software. The sea launch system includes a base, a launch rack, a launched system, and an adapter. The model can be reasonably simplified according to the actual structure and relative position of the launch device.

[0066] S1-2. Grid division

[0067] Use meshing software to mesh the geometric model. Thin structures such as steel plates and reinforcement plates are simulated using shell elements; solid block structures such as adapters and support columns are simulated using entity elements; and the launched system is simulated using rigid body elements. The mesh can be reasonably simplified based on the actual structure of the launcher and stress estimation.

[0068] S1-3. Input of parameters required for dynamic calculation

[0069] Definition of quality characteristics: Set the density, Young's modulus, Poisson's ratio and other parameters for the base, launcher and adapter respectively. The adapter can be simulated with rubber material and the corresponding values ​​can be input; set the center of gravity position and moment of inertia parameters for the launched system.

[0070] S1-4. Contact and load settings

[0071] Load: Apply thrust to the bottom of the launched system, and apply displacement boundary conditions to the base and the part of the launcher that contacts the ship (i.e., the launch pad and the crater), that is, on the launch pad or, simulated by a sinusoidal function.

[0072] Contact: contact is set between the launched system and the launcher, between the launched system and the adapter, between the launched system and the base, and between the adapter and the guide rail on the launcher;

[0073] S1-5. Solution and output settings

[0074] Set the calculation time and result saving frequency, and output the launch safety characteristic value when the launched system completely leaves the launch tube; in this embodiment, the force of the launched system during the launch process, displacement, velocity, acceleration in three directions, angular displacement, angular velocity, angular acceleration in three directions; stress and strain of the adapter, launch rack, and base; all contact forces are output.

[0075] S2. Sampling and sample library establishment

[0076] S2-1 Determine the sampling range

[0077] According to the maximum amplitude and maximum period of the six degrees of freedom of the ship motion corresponding to the maximum sea condition for safe launch at sea, the amplitude range of sampling is determined to be 0 to the maximum amplitude * expansion factor, and the sampling period is determined to be 0 to the maximum period * expansion factor. The expansion factor enables the sampling range to include the maximum amplitude and maximum period that may be encountered during actual launch. The sampling phase range is determined to be 0 to 2π.

[0078] S2-2 sampling to obtain sample points

[0079] Sampling methods such as stratified sampling and Latin hypercube can be used to obtain sampling points, and the information of each sampling point includes amplitude, period and phase.

[0080] S2-3 Constructing a sample library based on kinetic simulation results

[0081] Calculate the sampled ship motion results, that is t is time, i takes values ​​of 1, 2, 3, 4, 5, 6, representing the six degrees of freedom of the ship, A is Represents the sampled amplitude, T is represents the sampling period, Indicates the sampling phase. The ship motion results are used as boundary inputs in boundaries S1-4 to calculate and obtain the dynamic simulation calculation results of all sampling points.

[0082] The contact force between the launched system and the launch tube is extracted from the results. If the contact force is not equal to 0, it means a collision has occurred, which is recorded as a collision situation. If the contact force is equal to 0, the launch safety characteristic value is extracted. The launch safety characteristic value is the angular velocity and angular displacement in three directions of the launched system, the displacement in three directions, and the speed when the launched system is completely out of the tube. Thus, a sample library is obtained.

[0083] S3, establishment of agent model;

[0084] S3-1 takes 70% of the sample library obtained from S2 as a training set, 20% as a validation set, and 10% as a test set. By using the Kriging model, response surface model, or machine learning method, a proxy model between the amplitude, frequency, phase of the motion boundary of the launch device used in the simulation and the launch safety characteristic value is established.

[0085] S3-2 can use R-squared, mean square error, and mean absolute error to judge the correctness of the proxy model.

[0086]

[0087] Among them, y i is the true value, i.e. the actual value of the observed data; Predicted value, which is the data value predicted by the model The average value of the true value. 2 When the value of is close to 0.85 or meets the accuracy requirement, the proxy model is successfully established. Otherwise, return to S2 to increase the sampling points.

[0088] S4, collection and processing of real-time motion data of the launch device;

[0089] S4-1 uses a total of three groups of displacement sensors and three groups of angle sensors arranged at the base of the launch pad for continuous measurement, or arranged at a location that can be regarded as a rigid connection with the base of the launch pad (such as the surrounding deck), that is, the location of the crater. The angle sensor is required to be able to measure the result relative to the horizontal state. The sampling frequency is greater than 1000Hz. The position of the above arrangement can be regarded as a rigid structure, that is, a structure with almost no deformation. For this kind of structure, if the movement of one point is known, the movement of other points on the rigid component can be calculated through kinematics in theoretical mechanics. Therefore, measurements and conversions can also be made at other rigid connections.

[0090] S4-2 Fourier transform of motion curve in the past 3T seconds

[0091] At each moment, the measurement data of the past 3R seconds is fast Fourier transformed, T is the period of the ship's maximum degree of freedom motion under the current sea conditions, and the amplitude spectrum, power spectrum, phase spectrum, and signal with a frequency of 0-5hz are intercepted. According to the periodic characteristics of ship motion, its amplitude-frequency curve increases first and then decreases, and will present a unique maximum value.

[0092] S4-3 Past T Launch Fourier transform of motion curve in time

[0093] To the past T Launch Fast Fourier transform is performed in seconds to obtain the amplitude-frequency curve, power spectrum, phase spectrum, and signal with interception frequency of 0-5 Hz. According to the periodicity of ship motion, its amplitude-frequency curve increases first and then decreases, and will present a unique maximum value. Launch It is the time from ignition of the launched system to leaving the launch tube.

[0094] S4-4 Determination of Predictive Validity

[0095] The relationship between the corresponding curves obtained by two fast Fourier transforms can be judged by methods such as mean square error and Pearson correlation coefficient. If the corresponding curves are similar, it means that the spectrum of the ship's motion has remained stable in the past period of time, and the time of the future launch process is short, which will not have a great impact on the spectrum. Therefore, the fast Fourier transform results of the past time period can be used to predict the future launch process. In this embodiment, a new method for judging the similarity of two discrete curves is proposed and adopted.

[0096] Since the sensor frequency is the same, the discrete frequency values ​​in the results after fast Fourier transform are the same, so the i-th frequency of the two curves is equal.

[0097]

[0098] Where n is the total number of discrete frequencies, y i3T is the i-th value of the amplitude curve obtained in step S4-2, y iTL is the ith value of the amplitude curve obtained in step S4-3. When r is less than 0.2, the prediction is considered to be valid.

[0099] S4-5 Data feature value extraction

[0100] The data feature values ​​are the frequency, amplitude, and phase extracted from the curve of the six degrees of freedom motion measured by the sensor. The maximum amplitude, the frequency and phase corresponding to the maximum amplitude in the amplitude-frequency curve are taken as the data feature values. Thus, the predicted amplitude, phase, and frequency of the six degrees of freedom motion are obtained, a total of 18 values.

[0101] S5. Extract features from the collected data and input them into the proxy model to obtain emission safety feature values.

[0102] The data characteristic value obtained in S4-5 is input into the proxy model to obtain the launch safety characteristic value of the launched system, and it is determined whether the launch safety characteristic value exceeds the safe launch range.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A real-time prediction method for marine launch safety, characterized in that: The specific process is: Establish a dynamic model of the sea launch system: Establish a dynamic model of the sea launch system, set the model input load boundary conditions, and the model output is the launch safety characteristic value when the launched system completely leaves the launch tube; Obtaining a sample database: determining a sampling range, sampling from it to obtain sampling points, and calculating the ship motion results at each sampling point; using the ship motion results as input load boundary conditions of the marine launch system dynamics model, obtaining the output of the model for each sampling point, and obtaining a sample database; Training the proxy model: using the sample database, training the proxy model between the launch device motion boundary amplitude, frequency, phase and launch safety characteristic value; Collecting ship motion data: Using sensors installed on the ship to collect the ship's motion data, extract the motion boundary amplitude, frequency and phase, and make validity judgments; Safe launch prediction: Input the valid characteristic value into the proxy model to obtain the launch safety characteristic value of the launched system, and judge whether the launch safety characteristic value exceeds the safe launch range, so as to realize safe launch prediction.

2. The method for real-time prediction of marine launch safety according to claim 1, characterized in that: The specific process of establishing the sea launch system dynamics model and setting the model input load boundary conditions and model output is as follows: Establish a collective model of the sea launch system, perform meshing, and set the parameters required for model dynamics calculations; Model load setting: apply thrust to the bottom of the launched system, and apply displacement boundary conditions to the base and the part of the launcher that contacts the ship; Model contact setting: setting contact between the launched system and the launch frame, between the launched system and the adapter, between the launched system and the base, and between the adapter and the guide rail on the launch frame; Model output settings: output the forces of the launched system during the launch process, displacement in three directions, velocity, acceleration, angular displacement in three directions, angular velocity, and angular acceleration; Stress and strain in adapter, launcher, base; all contact forces.

3. The real-time prediction method for marine launch safety according to claim 1, characterized in that: The sampling range is: according to the maximum amplitude and maximum period of the six degrees of freedom motion of the ship corresponding to the maximum sea conditions for safe launch at sea, the amplitude range during sampling is determined to be 0 to the maximum amplitude * expansion factor, and the sampling period is 0 to the maximum period * expansion factor. The expansion factor makes the sampling range include the maximum amplitude and maximum period that may be encountered during actual launch; the sampling phase range is determined to be 0 to 2π.

4. The method for real-time prediction of marine launch safety according to claim 3, characterized in that: The sampling points are composed of amplitude, period and phase, and the ship motion results of each sampling point are calculated. t is time, i takes values ​​of 1, 2, 3, 4, 5, 6, representing the six degrees of freedom of the ship, A is Represents the sampled amplitude, T is represents the sampling period, Indicates the sampling phase.

5. The method for real-time prediction of marine launch safety according to claim 4, characterized in that: The output of the model corresponding to each sampling point is obtained, and the contact force between the launched system and the launching tube is extracted from the output result. If the contact force is not equal to 0, it indicates that a collision has occurred, which is recorded as a collision situation. If the contact force is equal to 0, the launch safety characteristic value is extracted. The launch safety characteristic value is the angular velocity and angular displacement in three directions of the launched system, the displacement in three directions, and the speed when the launched system is completely out of the tube, thereby obtaining a sample library.

6. The method for real-time prediction of marine launch safety according to claim 1, characterized in that: The launch safety characteristic values ​​also include: the minimum gap between the launched system and the launch tube during the launch process, the force applied to the launched system during the launch process, and the stress and strain of the adapter, launch frame, and base in the launch system.

7. The method for real-time prediction of marine launch safety according to claim 1, characterized in that: When the proxy model is trained, the R-squared, mean square error and / or mean absolute error method is used to judge the correctness of the proxy model. When it is judged that the proxy model training does not meet the correctness requirements, the number of sampling points is increased and further training is performed.

8. The method for real-time prediction of marine launch safety according to claim 1, characterized in that: The specific process of collecting ship motion data is as follows: S4-1 uses three sets of displacement sensors and three sets of angle sensors arranged on the launch pad base or crater for continuous measurement; S4-2 Fourier transform of motion curve in the past 3T seconds At each moment, the measurement data of the past 3T seconds are subjected to fast Fourier transformation, where T is the period of the ship's maximum degree of freedom motion under the current sea conditions, to obtain the amplitude spectrum, power spectrum, phase spectrum, and intercept the signal with a frequency of 0-5 Hz; S4-3 Past T Launch Fourier transform of motion curve in time To the past T Launch Fast Fourier transform is performed in seconds to obtain the amplitude-frequency curve, power spectrum, phase spectrum, and signal with intercept frequency of 0-5 Hz. Launch It is the time from ignition of the launched system to its exit from the launch tube; S4-4 Determination of Predictive Validity Determine the relationship between the corresponding curves obtained by the two fast Fourier transforms. If the corresponding curves are similar, it is considered that the currently collected data is valid, and proceed to step S4-5; S4-5 Data feature value extraction The data characteristic values ​​are obtained by extracting the maximum amplitude, the frequency and the phase corresponding to the maximum amplitude from the amplitude-frequency curve of the six-degree-of-freedom motion measured by the sensor, thereby obtaining the predicted amplitude, phase and frequency of the motion in the six degrees of freedom directions.

9. The method for real-time prediction of marine launch safety according to claim 8, characterized in that: The basis for judging that the curves are similar is: Calculate r based on the frequency of the two curves, Where n is the total number of discrete frequencies, y i3T is the i-th value of the amplitude curve obtained in step S4-2, y iTL is the i-th value of the amplitude curve obtained in step S4-3; When r is less than the set threshold, it is considered that the corresponding curves are similar and the prediction is valid.

10. The method for real-time prediction of marine launch safety according to claim 9, characterized in that: When r is less than 0.2, the corresponding curves are considered to be close and the prediction is valid.

Citation Information

Cited By

  • Method for improving high dynamic launching capability of load

    CN122433484A