CN114386175A - Ship local fatigue hotspot damage rate forecasting method - Google Patents
Ship local fatigue hotspot damage rate forecasting method Download PDFInfo
- Publication number
- CN114386175A CN114386175A CN202210041700.XA CN202210041700A CN114386175A CN 114386175 A CN114386175 A CN 114386175A CN 202210041700 A CN202210041700 A CN 202210041700A CN 114386175 A CN114386175 A CN 114386175A
- Authority
- CN
- China
- Prior art keywords
- ship
- fatigue
- motion
- spectrum
- motion response
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000006378 damage Effects 0.000 title claims abstract description 65
- 238000013277 forecasting method Methods 0.000 title claims abstract description 6
- 230000033001 locomotion Effects 0.000 claims abstract description 113
- 230000004044 response Effects 0.000 claims abstract description 100
- 238000001228 spectrum Methods 0.000 claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 38
- 238000012544 monitoring process Methods 0.000 claims abstract description 18
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 14
- 238000003062 neural network model Methods 0.000 claims abstract description 13
- 230000003938 response to stress Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000013461 design Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 239000006185 dispersion Substances 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 abstract description 9
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 230000036541 health Effects 0.000 abstract description 4
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000005314 correlation function Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for forecasting a local fatigue hotspot damage rate of a ship, which comprises the following steps: s1, respectively acquiring ship motion response as an input database and hotspot fatigue damage as an output database based on a spectrum analysis principle; s2, constructing an artificial neural network model corresponding to the fatigue hot spot, and establishing a relation between ship motion response and hot spot fatigue damage; s3, acquiring motion data through a real ship monitoring system, acquiring a wave energy spectrum by adopting a correlation function, and acquiring a ship motion response input parameter by adopting a spectrum analysis principle; s4, fatigue damage prediction based on ship real ship monitoring motion data is achieved by means of the neural network model. The invention provides a ship local fatigue hot spot damage forecasting method based on ship motion monitoring data. By utilizing the method, the indirect forecast of the fatigue damage of the position which is not directly monitored can be realized, and the method is a beneficial improvement on the monitoring function of the existing ship structure health monitoring system.
Description
Technical Field
The invention relates to the field of ship and ocean engineering, in particular to a method for forecasting local fatigue hotspot damage rate of a ship.
Background
The fatigue strength of a ship is an important aspect of the strength problem of a ship structure, and the accumulation of fatigue damage may lead to catastrophic failure of the ship structure. The fatigue strength evaluation of the ship in the design stage only considers the sea condition information given by the classification society, but for the same ship, different degrees of fatigue damage can be generated due to different operation experiences, and the fatigue strength evaluation of the design stage does not consider the difference of the operation experiences of the ship. Although the ship structure monitoring system can acquire the structural stress response in the ship operation process, the stress monitoring at a few positions can be generally realized due to the limitation of economic factors, and the full-coverage acquisition of the structural stress response of all fatigue hot spots cannot be realized.
Disclosure of Invention
The invention provides a method for forecasting a local fatigue hot spot damage rate of a ship, which aims to overcome the problems.
The invention comprises the following steps:
s1, based on the spectrum analysis principle, acquiring ship motion response as an input database, and acquiring hot spot fatigue damage as an output database;
s2, taking ship motion response data under short-term sea conditions as input data and fatigue damage as output data, constructing a neural network model corresponding to a fatigue hot point, and establishing a relation between ship motion response and hot point fatigue damage; the short-term sea state is a wave load model set according to experience in a set time;
s3, according to the ship motion data, a wave energy spectrum is obtained by adopting a fast Fourier transform method, and a ship motion response input parameter is obtained by adopting a spectrum analysis principle:
measuring the motion time history response of the ship by using a motion sensor arranged on the ship; decomposing continuous motion response according to experience and a certain time length, wherein each time period is the motion response time history of the ship under a short-term sea condition, and calculating a motion response spectrum by adopting a fast Fourier transform method; calculating the motion response zero-order moment and the second-order moment of the ship under the short-term sea condition according to the motion response spectrum;
s4, the fatigue damage forecast based on the ship real ship monitoring motion data is realized by utilizing a neural network model:
the ship motion response zero-order moment and the second-order moment under the short-term sea condition are used as input, and a neural network model is utilized to realize rapid prediction of different hot point fatigue injuries; fatigue damage under each short-term sea condition included in the set time period t can be linearly superposed, and the fatigue damage data of each hot spot of the ship corresponding to the operation experience of the ship can be forecasted.
Further, S1 includes the steps of:
s11, solving motion response and wave load data of the ship based on a three-dimensional linear frequency domain potential flow theory;
s12, determining the position of the fatigue hot spot needing to be concerned on the ship structure according to the design specification of the ship and the experience of a user;
s13, establishing a ship structure numerical model by adopting a finite element technology, and solving the amplitude-frequency response of each fatigue hot spot stress by loading the obtained ship wave load;
s14, determining short-term sea conditions for calculation according to the sea wave statistical data of the ship navigation area;
s15, solving ship motion response data and fatigue damage data of each hot spot under different short-term sea conditions by adopting a spectrum analysis method, and constructing a database containing ship motion response and fatigue damage of each hot spot.
Further, the fatigue damage data in S15 is calculated based on the following strategy:
S15A1, calculating a stress response spectrum of the fatigue hot spot:
GXX-p(ωe,θ)＝|Hp(ωe,θ)|2Gηη(ωe,θ) (1)
wherein the encounter frequency of the ship and the waves is omegaeThe course angle of the ship is theta, | Hp(ωe,θ)|2The stress amplitude-frequency response of the fatigue hot spot; gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-p(ωeθ) is the stress response spectrum of the calculated hot spot;
S15A2, by a spectral analysis method, the calculation formula of the fatigue damage data of the ship under the short-term sea condition is as follows:
wherein D is the fatigue damage in the short-term sea state, TdIs the duration of the short-term sea state,
and m is the S-N curve parameter corresponding to the fatigue hotspot analyzed, v0Is the mean zero crossing rate of the alternating stress process
Is a gamma function, m0And m2Respectively, a zero order moment and a second order moment of the fatigue hot spot stress response spectrum.
Further, the zeroth moment m of the stress response spectrum in S15A20And second moment m2The calculation formula of (2) is as follows:
wherein f (β) is the wave dispersion function; n is the order moment of the stress response spectrum, and n is 0 and 2.
Further, in S15, the motion response data is the zero-order moment m of the response spectrum of the motion of the ship in three degrees of freedom, namely heave, pitch and roll0And second moment m2；
The motion response data is obtained based on the following strategies:
S15B1, establishing a motion response spectrum of the vessel in motions of three degrees of freedom in the directions of heaving, pitching and rolling in the waves:
GXX-v(i)(ωe,θ)＝|Hv(i)(ωe,θ)|2Gηη(ωe,θ) (4)
wherein, i is 3,4,5 respectively corresponding to the motions of three degrees of freedom of the ship such as heave, pitch and roll in the waves, omegaeThe encounter frequency of the ship and the waves, theta is the ship course angle, | Hv(i)(ωe,θ)|2For corresponding amplitude-frequency response of motion in degrees of freedom, Gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-v(i)(ωeθ) is a motion response spectrum obtained by each degree;
S15B2, calculating the zeroth order moment m of the three-freedom-degree motion response spectrum0And second moment m2：
Wherein, i is 3,4 and 5 respectively corresponding to the three conditions of the ship such as heave, pitch and rollZero order moment m of motion response spectrum of individual degree of freedom0And second moment m2(ii) a k is the order moment of the motion response spectrum, k is 0, 2.
Further, the calculation formula of the motion response spectrum F (ω) in S4 is:
wherein, F (omega) is a motion response spectrum, F (t) is a motion response time history, and omega is the encounter frequency of the ship and the waves; e.g. of the type-iωtIs a complex variable function; t is a set time period.
The invention provides a ship local fatigue hot spot damage forecasting method based on ship motion monitoring data. By utilizing the method, the indirect forecast of the fatigue damage of the position which is not directly monitored can be realized, and the method is a beneficial improvement on the monitoring function of the existing ship structure health monitoring system.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a hydrodynamic grid diagram according to the present invention;
FIG. 3 is a graph of the amplitude-frequency response of the heave motion of the present invention;
FIG. 4 is a graph of the amplitude frequency response of the pitching motion of the present invention;
FIG. 5 is a graph of the roll motion amplitude frequency response of the present invention;
FIG. 6 is a finite element model diagram of the whole ship of the present invention;
FIG. 7 is a diagram of a finite element mesh model of hotspot 1 of the present invention;
FIG. 8 is a diagram of a hotspot 2 finite element mesh model of the present invention;
FIG. 9 is a magnitude-frequency response graph of the hotspot 1 stress response of the present invention;
FIG. 10 is a graph of the magnitude-frequency response of the hotspot 2 stress response of the present invention;
FIG. 11 is a diagram of a neural network architecture in accordance with the present invention;
FIG. 12 is a graph of the convergence results of hotspot 1 of the present invention;
FIG. 13 is a graph of the convergence results of hotspot 2 of the present invention;
FIG. 14 is a graph of the mean square error of the loss rate of the present invention;
FIG. 15 is a graph of the agreement between the neural network prediction data and the raw data according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention comprises the following steps:
s1, based on the spectrum analysis principle, acquiring ship motion response as an input database, acquiring hotspot fatigue damage as an output database:
s11, solving motion response and wave load data of the ship based on a three-dimensional linear frequency domain potential flow theory;
specifically, when wave load data is calculated, the wave load data is divided into long-term forecasting and short-term forecasting, waves corresponding to the short-term forecasting are described as short-term sea conditions, the time range of the short-term sea conditions is half an hour to several hours, and the short-term sea conditions can be regarded as a stable normal random process with the mean value of zero.
S12, determining the position of the fatigue hot spot needing to be concerned on the ship structure according to the design specification of the ship and the experience of a user;
specifically, the fatigue hot spots of the ship are different according to different ship models and different use places, so that the setting of the fatigue hot spots is obtained according to the design specifications of the ship, and the setting is added through the experience of a user.
S13, establishing a ship structure numerical model by adopting a finite element technology, and solving the amplitude-frequency response of each fatigue hot spot stress by loading the obtained ship wave load.
And S14, determining short-term sea conditions for calculation according to the sea wave statistics of the ship navigation area.
S15, solving ship motion response data and fatigue damage data of each hot spot under different short-term sea conditions by adopting a spectrum analysis method, and constructing a database containing ship motion response and fatigue damage of each hot spot.
Preferably, the fatigue damage data in S15 is calculated based on the following strategy:
S15A1, calculating a stress response spectrum of the fatigue hot spot:
GXX-p(ωe,θ)＝|Hp(ωe,θ)|2Gηη(ωe,θ) (1)
wherein the encounter frequency of the ship and the waves is omegaeThe course angle of the ship is theta, | Hp(ωe,θ)|2The stress amplitude-frequency response of the fatigue hot spot; gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-p(ωeθ) is the stress response spectrum of the calculated hot spot;
S15A2, by a spectral analysis method, the calculation formula of the fatigue damage data of the ship under the short-term sea condition is as follows:
wherein D is the fatigue damage in the short-term sea state, TdIs the duration of the short-term sea state,
and m is the S-N curve parameter corresponding to the fatigue hotspot analyzed, v0Is the mean zero crossing rate of the alternating stress process
Is a gamma function, m0And m2Respectively, a zero order moment and a second order moment of the fatigue hot spot stress response spectrum.
In particular, in marine and oceanographic engineering, waves are generally regarded as a smooth and normal random process. Accordingly, the alternating fatigue stress process of the ship under the short-term sea condition can be regarded as a steady normal process with a zero mean value, and the stress peak distribution of the stress alternating process in the short-term sea condition is subjected to Rayleigh distribution.
Preferably, the zero order moment m of the stress response spectrum in S15A20And second moment m2The calculation formula of (2) is as follows:
wherein f (β) is the wave dispersion function; n is the order moment of the stress response spectrum, and n is 0 and 2.
Preferably, in S15, the motion response data is the zero-order moment m of the response spectrum of the motion of the ship in three degrees of freedom, namely heave, pitch and roll0And second moment m2；
The motion response data is obtained based on the following strategies:
S15B1, establishing a motion response spectrum of the vessel in motions of three degrees of freedom in the directions of heaving, pitching and rolling in the waves:
GXX-v(i)(ωe,θ)＝|Hv(i)(ωe,θ)|2Gηη(ωe,θ) (4)
wherein, i is 3,4,5 respectively corresponding to the motions of three degrees of freedom of the ship such as heave, pitch and roll in the waves, omegaeThe encounter frequency of the ship and the waves, theta is the ship course angle, | Hv(i)(ωe,θ)|2For corresponding amplitude-frequency response of motion in degrees of freedom, Gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-v(i)(ωeθ) is obtained from the respective degreesA motion response spectrum;
S15B2, calculating the zeroth order moment m of the three-freedom-degree motion response spectrum0And second moment m2：
Wherein, i is 3,4 and 5 respectively corresponding to the zero-order moment m of the ship motion response spectrum in three degrees of freedom of heave, pitch and roll0And second moment m2(ii) a k is the order moment of the motion response spectrum, k is 0, 2.
Specifically, the motion characteristics of the ship in waves can directly reflect the sea conditions encountered by the ship, the three-degree-of-freedom motions of heave (i-3), pitch (i-4) and roll (i-5) with restoring force or playback moment are selected, two relevant parameters about a stress response spectrum adopted in the calculation of fatigue damage are referred, and a zero-order moment m is also adopted in the motion response spectrum0And second moment m2As motion input data.
S2, taking ship motion response data under short-term sea conditions as input data and fatigue damage as output data, constructing a neural network model corresponding to a fatigue hot point, and establishing a relation between ship motion response and hot point fatigue damage; the short-term sea state is a wave load model set according to experience in a set time;
specifically, the neural network model is built by taking motion response as input data and fatigue damage as output data, the technology used for building the neural network model is the existing neural network technology, and the relation between the motion and the fatigue damage is built only by means of the neural network model.
Specifically, taking a container ship as an example, a fatigue damage prediction model based on motion data is established (corresponding to S1 and S2 in the present invention), fig. 2 is a hydrodynamic grid of a target ship, and a motion response and a wave load of the target ship can be solved based on a three-dimensional linear frequency domain potential flow theory. Fig. 3-5 are amplitude-frequency responses of heave, pitch and roll motions, respectively, of a target vessel. Wherein, the frequency omega has a variation range of 0.1-2.0rad/s and an increment of 0.1rad/s, the course angle theta is 0-180 DEG because of the symmetry of the ship, and the increment is 30 deg.
Fig. 6 is a ship-wide finite element model of a container ship. There are many fatigue hot spots on the actual ship, and two typical corner positions are selected as analysis objects, wherein the hot spot 1 is a position where the end of the cabin aggregate is connected with the cabin wall, and the hot spot 2 is a corner of the second deck in the stern cargo compartment area, as shown in fig. 7 and 8. By loading the obtained wave load of the ship, the amplitude-frequency response of the fatigue hot spot stress can be obtained, and the amplitude-frequency response is shown in the graph 9 and the graph 10.
Selecting the long-term statistical result of ocean waves of North Atlantic, wherein the height H of a sense wavesAnd average over zero period TzIs two key parameters for characterizing the wave form, and a series of waves with short-term sea states can be combined by different values of the sense wave height and the zero crossing period, wherein 17H wavessThe variation range is 0.5m-16.5m, the interval is 1m, and 18TzThe variation range is 1.5s-18.5s, and the interval is 1 s. And (4) solving the ship motion response and the fatigue damage of each hot spot under different short-term sea conditions based on the step 1.5, and constructing a database containing the ship motion response and the fatigue damage of each hot spot.
Normalizing the input and output data, building a neural network model (figure 11), and randomly selecting 75% of original data as training data and the other 25% of original data as verification data. The core algorithm is selected as an error back propagation algorithm, and training of the neural network model is carried out by adjusting the neural network structure, the iteration times and the learning rate.
The mean-square error is selected as an evaluation standard, mean-square error (MSE) is a convenient method for measuring the mean error, the MSE can evaluate the change degree of data, and the smaller the value of the MSE is, the better accuracy of the prediction model description experiment data is shown. The expression is as follows:
wherein MSE is a mean square error function; k is the number of samples;
the method is characterized in that the method is a predicted value output by the neural network, y is a true value of input data, i represents ith data, and the numeric area of i is 1 to n.
The neural network data comprises: 1. training data: training Dataset; 2. and (3) verifying data: validation Dataset; 3. cycle number: epochs. The convergence results of the hot spots 1 and 2 (fig. 12 and 13) show that the MSE (ordinate loss) values of the training loss rate and the verification loss rate are both lower than 0.0002 after 600 cycles, the coincidence degree of the neural network prediction data (ANNet) and the original data (Reference) is high (fig. 14 and 15), and the requirement of predicting fatigue damage through the ship motion monitoring data in the operation process can be completely met.
S3, according to the ship motion data, a wave energy spectrum is obtained by adopting a fast Fourier transform method, and a ship motion response input parameter is obtained by adopting a spectrum analysis principle: decomposing continuous motion response according to a certain time length, wherein each time period can be regarded as the motion response time of the ship under a short-term sea condition, calculating a motion response spectrum by adopting fast Fourier transform, and further calculating the zero order moment and the second order moment of the motion response of the ship under the short-term sea condition.
A motion response time history F (t) in short-term sea states can be obtained by fourier transformation of the motion response spectrum F (ω):
wherein, F (omega) is a motion response spectrum, F (t) is a motion response time history, and omega is the encounter frequency of the ship and the waves; e.g. of the type-iωtIs a complex variable function; t is a set time period.
And S4, using the ship motion response zero-order moment and the second-order moment under the short-term sea condition as input, and using the neural network forecasting models corresponding to different fatigue hot spots to realize the rapid forecasting of the fatigue damage of different hot spots. Fatigue damage in a series of time periods is superposed, so that fatigue damage of each hot spot of the ship, which is experienced by the ship operation, can be corresponded, and safe operation and maintenance of the ship structure is guided.
The whole beneficial effects are as follows:
the existing ship structure health monitoring system is generally provided with a sensor at the position of a ship fatigue hot spot to forecast the fatigue damage of a monitoring position. The invention provides a ship local fatigue hot spot damage forecasting method based on ship motion monitoring data. By utilizing the method, the indirect forecast of the fatigue damage of the position which is not directly monitored can be realized, and the method is a beneficial improvement on the monitoring function of the existing ship structure health monitoring system.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. A ship local fatigue hotspot damage rate forecasting method is characterized by comprising the following steps:
s1, based on the spectrum analysis principle, acquiring ship motion response as an input database, and acquiring hot spot fatigue damage as an output database;
s2, taking ship motion response data under short-term sea conditions as input data and fatigue damage as output data, constructing a neural network model corresponding to a fatigue hot point, and establishing a relation between ship motion response and hot point fatigue damage; the short-term sea state is a wave state set according to experience in the sea state time, and the sea state time is set according to experience;
s3, according to the ship motion data, a wave energy spectrum is obtained by adopting a fast Fourier transform method, and a ship motion response input parameter is obtained by adopting a spectrum analysis principle:
measuring the motion time history response of the ship by using a motion sensor arranged on the ship; decomposing continuous motion response according to experience and a certain time length, wherein each time period is the motion response time history of the ship under a short-term sea condition, and calculating a motion response spectrum by adopting a fast Fourier transform method; calculating the motion response zero-order moment and the second-order moment of the ship under the short-term sea condition according to the motion response spectrum;
s4, the fatigue damage forecast based on the ship real ship monitoring motion data is realized by utilizing a neural network model:
the ship motion response zero-order moment and the second-order moment under the short-term sea condition are used as input, and a neural network model is utilized to realize rapid prediction of different hot point fatigue injuries; fatigue damage under each short-term sea condition included in the set time period t can be linearly superposed, and the fatigue damage data of each hot spot of the ship corresponding to the operation experience of the ship can be forecasted.
2. The method for forecasting the local fatigue hotspot damage rate of the ship according to claim 1, wherein the step S1 comprises the following steps:
s11, solving motion response and wave load data of the ship based on a three-dimensional linear frequency domain potential flow theory;
s12, determining the position of the fatigue hot spot needing to be concerned on the ship structure according to the design specification of the ship and the experience of a user;
s13, establishing a ship structure numerical model by adopting a finite element technology, and solving the amplitude-frequency response of each fatigue hot spot stress by loading the obtained ship wave load;
s14, determining short-term sea conditions for calculation according to the sea wave statistical data of the ship navigation area;
s15, solving ship motion response data and fatigue damage data of each hot spot under different short-term sea conditions according to the amplitude-frequency response of each fatigue hot spot stress by adopting a spectrum analysis method, and constructing a database containing ship motion response and each hot spot fatigue damage.
3. The method for forecasting the local fatigue hotspot damage rate of the ship according to claim 2, wherein the fatigue damage data in the step S15 is calculated based on the following strategies:
S15A1, calculating a stress response spectrum of the fatigue hot spot:
GXX-p(ωe，θ)＝|Hp(ωe，θ)|2Gηη(ωe，θ) (1)
wherein the encounter frequency of the ship and the waves is omegaeThe course angle of the ship is theta, | Hp(ωe，θ)|2The stress amplitude-frequency response of the fatigue hot spot; gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-p(ωeθ) is the stress response spectrum of the calculated hot spot;
S15A2, by a spectral analysis method, the calculation formula of the fatigue damage data of the ship under the short-term sea condition is as follows:
wherein D is the fatigue damage in the short-term sea state, TdIs the duration of the short-term sea state,
and m is the S-N curve parameter corresponding to the fatigue hotspot analyzed, v0Is the mean zero crossing rate of the alternating stress process
Is a gamma function, m0And m2Respectively, a zero order moment and a second order moment of the fatigue hot spot stress response spectrum.
4. The method for forecasting the local fatigue hot spot damage rate of the ship according to claim 2, wherein the zero-order moment m of the stress response spectrum in S15A20And second moment m2The calculation formula of (2) is as follows:
wherein f (β) is the wave dispersion function; n is the order moment of the stress response spectrum, and n is 0 and 2.
5. The method for forecasting the local fatigue hot spot damage rate of the ship according to claim 2, wherein the motion response data in S15 is the zero-order moment m of the motion response spectrum of the ship in three degrees of freedom, namely heave, pitch and roll0And second moment m2；
The motion response data is obtained based on the following strategies:
S15B1, establishing a motion response spectrum of the vessel in motions of three degrees of freedom in the directions of heaving, pitching and rolling in the waves:
GXX-v(i)(ωe，θ)＝|Hv(i)(ωe，θ)|2Gηη(ωe，θ) (4)
wherein, i is 3,4,5 respectively corresponding to the motions of three degrees of freedom of the ship such as heave, pitch and roll in the waves, omegaeThe encounter frequency of the ship and the waves, theta is the ship course angle, | Hv(i)(ωe，θ)|2For corresponding amplitude-frequency response of motion in degrees of freedom, Gηη(ωeθ) is the wave spectrum of the short-term sea conditions encountered by the vessel; gXX-v(i)(ωeθ) is a motion response spectrum obtained by each degree;
S15B2, calculating the zeroth order moment m of the three-freedom-degree motion response spectrum0And second moment m2：
Wherein, i is 3,4 and 5 respectively corresponding to the zero-order moment m of the ship motion response spectrum in three degrees of freedom of heave, pitch and roll0And second moment m2(ii) a k is the order moment of the motion response spectrum, k is 0, 2.
6. The method for forecasting the local fatigue hotspot damage rate of the ship according to claim 1, wherein the formula for calculating the motion response spectrum F (ω) in S4 is as follows:
wherein, F (omega) is a motion response spectrum, F (t) is a motion response time history, and omega is the encounter frequency of the ship and the waves; e.g. of the type-iωtIs a complex variable function; t is a set time period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210041700.XA CN114386175B (en) | 2022-01-14 | 2022-01-14 | Ship local fatigue hotspot damage rate forecasting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210041700.XA CN114386175B (en) | 2022-01-14 | 2022-01-14 | Ship local fatigue hotspot damage rate forecasting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114386175A true CN114386175A (en) | 2022-04-22 |
| CN114386175B CN114386175B (en) | 2022-10-04 |
Family
ID=81202525
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210041700.XA Active CN114386175B (en) | 2022-01-14 | 2022-01-14 | Ship local fatigue hotspot damage rate forecasting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114386175B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115230910A (en) * | 2022-08-10 | 2022-10-25 | 南通中远海运川崎船舶工程有限公司 | Intelligent health monitoring system and method for ship structure based on wave radar |
| CN117909665A (en) * | 2024-03-18 | 2024-04-19 | 青岛哈尔滨工程大学创新发展中心 | Ship motion envelope forecast data processing method and system based on Fourier filtering |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378019A (en) * | 2019-07-18 | 2019-10-25 | 上海交通大学 | In conjunction with the semi-submerged platform method for estimating fatigue damages of marine actual measurement and numerical analysis |
-
2022
- 2022-01-14 CN CN202210041700.XA patent/CN114386175B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378019A (en) * | 2019-07-18 | 2019-10-25 | 上海交通大学 | In conjunction with the semi-submerged platform method for estimating fatigue damages of marine actual measurement and numerical analysis |
Non-Patent Citations (3)
| Title |
|---|
| 崔兵兵等: "基于谱分析方法的月池角隅结构疲劳强度分析", 《船舶设计通讯》 * |
| 张浩辉等: "船舶应力监测系统疲劳监测点的布置", 《船舶工程》 * |
| 肖桃云等: "基于谱分析LNG船典型热点疲劳可靠性分析", 《海洋工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115230910A (en) * | 2022-08-10 | 2022-10-25 | 南通中远海运川崎船舶工程有限公司 | Intelligent health monitoring system and method for ship structure based on wave radar |
| CN117909665A (en) * | 2024-03-18 | 2024-04-19 | 青岛哈尔滨工程大学创新发展中心 | Ship motion envelope forecast data processing method and system based on Fourier filtering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114386175B (en) | 2022-10-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114386175B (en) | Ship local fatigue hotspot damage rate forecasting method | |
| CN110750875B (en) | Structure dynamic and static parameter uncertainty quantitative analysis system only using output response | |
| CN113033011B (en) | Ship mechanical health state assessment method and system | |
| CN111582551A (en) | Method and system for predicting short-term wind speed of wind power plant and electronic equipment | |
| CN102901651A (en) | Fractional order neural network performance degradation model and service life prediction method for electronic product | |
| CN114580493B (en) | Heavy haul railway bridge health monitoring method based on AI | |
| CN114565124A (en) | Ship traffic flow prediction method based on improved graph convolution neural network | |
| Mathew et al. | Regression kernel for prognostics with support vector machines | |
| CN116933152B (en) | Wave information prediction method and system based on multidimensional EMD-PSO-LSTM neural network | |
| CN111695452A (en) | Parallel reactor internal aging degree evaluation method based on RBF neural network | |
| CN112364560A (en) | Intelligent prediction method for working hours of mine rock drilling equipment | |
| CN115422687A (en) | Service life prediction method of rolling bearing | |
| CN116595319A (en) | Prediction method and system applied to rail transit motor health state evaluation | |
| CN111506868A (en) | Ultrashort-term wind speed prediction method based on HHT weight optimization | |
| Friis-Hansen | Reliability analysis of a midship section | |
| Yu et al. | Online ship rolling prediction using an improved OS-ELM | |
| Bechrakis et al. | Wind speed prediction using artificial neural networks | |
| CN116402143A (en) | Intelligent ship index system construction and evaluation method | |
| Yazid et al. | Identification of slow drift motions of a truss spar platform using parametric Volterra model | |
| CN115392589A (en) | Sea wave height forecasting method and system | |
| CN111428420B (en) | Method and device for predicting sea surface flow velocity, computer equipment and storage medium | |
| Luo et al. | A novel method for remaining useful life prediction of roller bearings involving the discrepancy and similarity of degradation trajectories | |
| Yao et al. | Extreme motion prediction and early-warning assessment of semisubmersible platform based on deep learning method | |
| Fujikubo et al. | A digital twin for ship structures—R&D project in Japan | |
| Silva et al. | Data-driven identification of critical wave groups |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20220422 Assignee: Shandong Huihai Marine Technology Co.,Ltd. Assignor: Dalian Maritime University Contract record no.: X2023980053065 Denomination of invention: A prediction method for local fatigue hot spot damage rate of ships Granted publication date: 20221004 License type: Common License Record date: 20231219 |
|
| EE01 | Entry into force of recordation of patent licensing contract |