A method and system for reducing vibration and noise of ships

Abstract

The present invention proposes a method and system for ship vibration reduction and noise reduction, which obtains real-time ship operation data and historical ship operation data, performs feature analysis respectively, obtains a first feature set and a second feature set, obtains key factor factors through key factor extraction and factor association, sets a number of different working condition training channels for a preset ship operation simulation model, performs simulation training for different working conditions, obtains vibration responses and noise responses under different working conditions, updates the key factor factors until the vibration responses and noise responses under different working conditions meet the preset operation requirements, obtains target vibration responses and target noise responses, generates target ship operation strategies and adjusts ship operation parameters. The present invention solves the problem that the prior art lacks simulation analysis of ship vibration data and noise data under different working conditions, resulting in poor reliability of ship vibration reduction and noise reduction optimization schemes. The present invention improves the reliability of ship vibration reduction and noise reduction optimization schemes through multi-level data simulation analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration reduction and noise reduction, and in particular to a method and system for vibration reduction and noise reduction of ships. Background Art

[0002] During the operation of a ship, noise and vibration are generated by the operation of mechanical equipment such as the main engine, auxiliary engine and propeller, as well as the interaction between the water flow and the hull. Large vibrations and noises generated during the operation of a ship not only affect the health of the crew and the water environment, but also cause unpredictable damage to the ship itself.

[0003] At present, traditional vibration reduction and noise reduction technologies are usually active control, passive control and hybrid control. Active control is to dynamically adjust the system structure according to the set control law and actively add dynamic vibration absorbers to the vibration source, thereby reducing high-frequency factors to achieve vibration reduction and noise reduction effects; passive control is to use conventional sound insulation technology, sound absorption technology and vibration isolation technology combined with specific materials to reduce low-frequency factors to achieve vibration reduction and noise reduction effects; hybrid control is to combine active control and passive control to achieve control of high-frequency factors and low-frequency factors, and achieve vibration reduction and noise reduction effects. It can be seen that the existing technology lacks simulation analysis of ship vibration data and noise data under different working conditions, resulting in the problem of poor reliability of ship vibration reduction and noise reduction optimization solutions. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a method and system for ship vibration and noise reduction, which can realize simulation analysis of ship vibration data and noise data under different working conditions and improve the reliability of ship vibration and noise reduction optimization scheme.

[0005] To achieve the above-mentioned purpose, an embodiment of the present invention provides a ship vibration reduction and noise reduction method, comprising:

[0006] Obtain real-time ship operation data and historical ship operation data;

[0007] Based on a preset feature analysis algorithm, feature analysis is performed on the real-time operation data of the ship and the historical operation data of the ship to obtain a first feature set and a second feature set;

[0008] Based on a preset factor identification algorithm, extract key factors and associate factors from the first feature set and the second feature set to obtain key factor factors;

[0009] A plurality of different working condition training channels are set for the preset ship operation simulation model, and the key factors are input into the preset ship operation simulation model to perform simulation training under different working conditions, so as to obtain vibration response and noise response under different working conditions;

[0010] Updating the key factor factors until the vibration response and the noise response under the different working conditions meet the preset operation requirements, and obtaining the target vibration response and the target noise response;

[0011] Based on the target vibration response and the target noise response, a target ship operation strategy is generated and ship operation parameters are adjusted to reduce vibration and noise of the ship.

[0012] The embodiment of the present invention proposes a ship vibration reduction and noise reduction method, which increases the data basis and ensures the applicability of the optimization scheme by acquiring real-time operation data and historical operation data and performing feature analysis respectively, and associates the real-time operation data with the historical operation data by combining key factor extraction and joint technology, thereby improving the reliability of data analysis; then, by simulating the operation of the ship under different working conditions, the vibration source and noise source under different working conditions are analyzed at multiple levels, so that targeted optimization schemes can be made, and the reliability of the vibration reduction and noise reduction optimization scheme is improved; and by continuously updating key factors, the operation data is continuously iterated, the vibration and noise responses of the target are screened, and the target operation strategy is generated based on the vibration and noise responses of the target, and the operation parameters of the ship are adjusted, so that the ship can achieve vibration reduction and noise reduction during operation; therefore, the present invention provides multi-scale data feature analysis combined with key factors, simulates the operation of the ship under different working conditions, generates a target operation strategy, and improves the reliability of the ship vibration reduction and noise reduction optimization scheme.

[0013] Furthermore, the acquisition of real-time ship operation data and historical ship operation data includes:

[0014] The first sensor data is acquired in real time by the vibration sensor, the first sensor data is traversed, and the data that meets the preset vibration value interval is marked as normal data, the data that does not meet the preset vibration value interval is marked as abnormal data, and the data with the same value as the last traversed data is marked as duplicate data, the abnormal data and duplicate data are eliminated, and the normal data is integrated to obtain the real-time vibration monitoring data of the ship;

[0015] The second sensing data is acquired in real time by the noise sensor, and a noise spectrum is constructed based on the second sensing data, prominent points and blank points in the spectrum are removed, and the processed spectrum is converted into corresponding data values ​​to obtain real-time noise monitoring data of the ship;

[0016] Integrate the ship's real-time vibration monitoring data and the ship's real-time noise monitoring data to obtain the ship's real-time operation data;

[0017] Based on the ship's historical operation database, the ship's historical vibration monitoring data and the ship's historical noise monitoring data are read to obtain the ship's historical operation data.

[0018] Furthermore, based on the preset feature analysis algorithm, feature analysis is performed on the real-time ship operation data and the historical ship operation data to obtain a first feature set and a second feature set, including:

[0019] respectively adding time weight values ​​to the real-time operation data of the ship and the historical operation data of the ship, extracting real-time operation feature data and historical operation feature data corresponding to the real-time operation data of the ship and the historical operation data of the ship within a plurality of preset time intervals, and constructing a first operation data feature vector and a second operation data feature vector;

[0020] Eliminate feature vectors whose contribution rate to the first operation data feature vector and the second operation data feature vector is less than or equal to a preset evaluation threshold value by using a principal component analysis method to obtain a first decomposed feature vector and a second decomposed feature vector;

[0021] Based on a preset sliding window, slidingly traverse the first decomposed feature vector and the second decomposed feature vector with a first unit length to obtain several groups of first feature comprehensive data and several groups of second feature comprehensive data;

[0022] Based on several groups of first feature comprehensive data and several groups of second feature comprehensive data, a first feature set and a second feature set are constructed.

[0023] Furthermore, based on the preset factor identification algorithm, the first feature set and the second feature set are subjected to key factor extraction and factor association to obtain key factor factors, including:

[0024] Based on a preset factor identification algorithm, identifying key factor labels corresponding to each feature data in the first feature set and the second feature set, to obtain a first key factor set and a second key factor set;

[0025] Calculate the correlation coefficient between the first key factor set and the second key factor set, and obtain the key factor with the largest correlation coefficient as the key factor factor; wherein the correlation coefficient calculation formula is as follows:

[0026]

[0027] In the formula, R is the correlation coefficient, Y min is the minimum difference, Y max is the maximum difference, Y real is the actual difference; the actual difference Y real is the difference between the kth data point of the i-th data set in the first key factor set and the kth data point of the i-th data set in the second key factor set. The minimum difference Y min and the maximum difference Y max is the minimum and maximum difference between the data points corresponding to all data sets of the first key factor set and the second key factor set, It is the sensitivity coefficient, and its value range is between 0 and 1.

[0028] Furthermore, the preset ship operation simulation model is provided with a plurality of different working condition training channels, and the key factors are input into the preset ship operation simulation model for different working condition simulation training to obtain vibration response and noise response under different working conditions, including:

[0029] Inputting the key factors into a preset ship operation simulation model; wherein the preset ship operation simulation model is provided with a plurality of training channels of different working conditions;

[0030] Based on the preset convolution algorithm and attention mechanism algorithm, simulations under different working conditions are carried out through the preset ship operation simulation model to obtain the vibration response and noise response under different working conditions.

[0031] Furthermore, based on the preset convolution algorithm and the attention mechanism algorithm, the simulation under different working conditions is performed through the preset ship operation simulation model to obtain the vibration response and noise response under different working conditions, including:

[0032] Based on a multi-column convolution algorithm, a training channel is matched with several different working conditions; wherein the training channel includes: a convolution layer, a function activation layer, a simulation training layer, an attention mechanism algorithm layer and an output layer;

[0033] Based on the convolution layer, feature recognition is performed on the key factor to obtain first training feature data;

[0034] Based on the function activation layer, a weight coefficient is calculated for the first training feature data by a preset activation function, and the calculated weight coefficient is multiplied by the key factor to obtain second training feature data;

[0035] Based on the attention mechanism algorithm layer, the product of the second training feature data and the weight values ​​of different operating conditions is calculated to obtain the third training feature data;

[0036] The simulation training layer is used to perform running simulation training on the third training feature data, and the vibration response and noise response under different working conditions are output through the output layer.

[0037] Further, updating the key factor factors until the vibration response and noise response under the different working conditions meet the preset operation requirements, and obtaining the target vibration response and target noise response, including:

[0038] Based on a preset update time interval, the real-time operation data of the ship is updated in real time, and the key factors are updated synchronously;

[0039] Whenever the vibration response is lower than the preset vibration frequency and the noise response is lower than the preset decibel value, the currently updated key factor is recorded as the target key factor;

[0040] The corresponding vibration response and noise response generated by simulation training of the target key factor are output as the target vibration response and the target noise response.

[0041] Further, based on the target vibration response and the target noise response, a target ship operation strategy is generated and the target ship operation parameters are adjusted to reduce vibration and noise of the ship, including:

[0042] Based on the target vibration response and the target noise response, the corresponding target key factor is identified, and the real-time ship operation data and the historical ship operation data corresponding to the target key factor are obtained;

[0043] Map the historical operation parameters corresponding to the historical operation data of the ship to the real-time operation data of the ship, and formulate the target operation strategy based on the historical operation parameters;

[0044] Adjust the ship's target operating parameters based on the target operation strategy.

[0045] The embodiment of the present invention also provides a ship vibration reduction and noise reduction system, comprising: a data acquisition module, a feature analysis module, a factor identification module, a working condition simulation module, a factor update module and a vibration reduction and noise reduction module;

[0046] The data acquisition module is used to acquire real-time ship operation data and historical ship operation data;

[0047] The feature analysis module is used to perform feature analysis on the real-time ship operation data and the historical ship operation data based on a preset feature analysis algorithm to obtain a first feature set and a second feature set;

[0048] The factor identification module is used to extract key factors and associate factors on the first feature set and the second feature set based on a preset factor identification algorithm to obtain key factor factors;

[0049] The working condition simulation module is used to set a number of different working condition training channels for the preset ship operation simulation model, input the key factors into the preset ship operation simulation model to perform simulation training under different working conditions, and obtain vibration response and noise response under different working conditions;

[0050] The factor updating module is used to update the key factor factors until the vibration response and noise response under the different working conditions meet the preset operation requirements, thereby obtaining the target vibration response and the target noise response;

[0051] The vibration reduction and noise reduction module is used to generate a target ship operation strategy and adjust ship operation parameters based on the target vibration response and the target noise response to reduce vibration and noise of the ship.

[0052] The embodiment of the present invention proposes a ship vibration reduction and noise reduction system, which acquires real-time operation data and historical operation data through a data acquisition module and a feature analysis module and performs feature analysis respectively, thereby increasing the data basis and ensuring the applicability of the optimization scheme. The real-time operation data is associated with the historical operation data through a factor identification module combined with key factor extraction and joint technology, thereby improving the reliability of data analysis. The ship operation under different working conditions is simulated through the factor identification module, and the vibration source and noise source under different working conditions are analyzed at multiple levels, so that a targeted optimization scheme can be made, thereby improving the reliability of the vibration reduction and noise reduction optimization scheme. The key factor factors are continuously updated through the factor update module, so that the operation data is continuously iterated, the vibration and noise response of the target are screened, and the target operation strategy is generated based on the vibration and noise response of the target through the vibration reduction and noise reduction module, and the ship's operation parameters are adjusted, so that the ship can achieve vibration reduction and noise reduction during operation. Therefore, the present invention provides multi-scale data feature analysis combined with key factors, simulates the ship operation under different working conditions, generates a target operation strategy, and improves the reliability of the ship vibration reduction and noise reduction optimization scheme.

[0053] Furthermore, the data acquisition module is used to acquire real-time ship operation data and historical ship operation data; including:

[0054] A real-time vibration data acquisition unit, a real-time noise data acquisition unit, a real-time operation data construction unit and a historical operation data construction unit;

[0055] The real-time vibration data acquisition unit is used to acquire first sensor data in real time through a vibration sensor, traverse the first sensor data, mark the data that meets the preset vibration value interval as normal data, mark the data that does not meet the preset vibration value interval as abnormal data, mark the data with the same value as the last traversed data as duplicate data, remove the abnormal data and duplicate data, and integrate the normal data to obtain the real-time vibration monitoring data of the ship;

[0056] The real-time noise data acquisition unit is used to acquire the second sensing data in real time through the noise sensor, and construct a noise spectrum based on the second sensing data, remove prominent points and blank points in the spectrum, and then convert the processed spectrum into corresponding data values ​​to obtain real-time ship noise monitoring data;

[0057] The real-time operation data construction unit is used to integrate the real-time vibration monitoring data of the ship and the real-time noise monitoring data of the ship to obtain the real-time operation data of the ship;

[0058] The historical operation data construction unit is used to read the historical vibration monitoring data and the historical noise monitoring data of the ship based on the historical operation database of the ship to obtain the historical operation data of the ship. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic flow chart of the steps of a method for reducing vibration and noise of a ship provided by a certain embodiment of the present invention;

[0060] Figure 2 A schematic diagram of a sliding window processing data flow of a ship vibration reduction and noise reduction method provided by an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of a data processing flow under different working conditions of a ship vibration and noise reduction method provided by an embodiment of the present invention;

[0062] Figure 4 A schematic diagram of the module structure of a ship vibration and noise reduction system provided by an embodiment of the present invention;

[0063] Figure 5 A schematic diagram of the structure of a data acquisition module of a ship vibration and noise reduction system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Example 1

[0066] See also Figure 1 , Figure 1 A schematic flow chart of the steps of a ship vibration reduction and noise reduction method provided by an embodiment of the present invention; Figure 1 As shown, the embodiment of the present invention provides a method for reducing vibration and noise of a ship, including steps 101 to 106, each of which is specifically as follows:

[0067] Step 101, obtaining real-time ship operation data and historical ship operation data;

[0068] As an example of this embodiment, the first sensor data is acquired in real time by a vibration sensor, and the data that meets the preset vibration value range is traversed through the first sensor data, and the data that does not meet the preset vibration value range is marked as normal data, and the data with the same value as the last traversed data is marked as duplicate data, and the abnormal data and duplicate data are eliminated, and the normal data are integrated to obtain the real-time vibration monitoring data of the ship; the second sensor data is acquired in real time by a noise sensor, and a noise spectrum graph is constructed based on the second sensor data, and the prominent points and blank points in the spectrum graph are eliminated, and the processed spectrum graph is converted into corresponding data values ​​to obtain the real-time noise monitoring data of the ship; the real-time vibration monitoring data of the ship and the real-time noise monitoring data of the ship are integrated to obtain the real-time operation data of the ship; based on the ship historical operation database, the historical vibration monitoring data of the ship and the historical noise monitoring data of the ship are read to obtain the historical operation data of the ship. Specifically, the vibration data of the ship mainly comes from hydrodynamic, mechanical vibration and propeller vibration. The first sensor data is obtained in real time through the vibration sensor preset on the hull. The vibration sensor can adopt the existing conventional piezoelectric sensor, electric sensor, eddy current sensor, inductive sensor and capacitive sensor. In this embodiment, the collected first vibration data are marked one by one, and the data greater than or less than the preset vibration value interval is marked as abnormal data, and the data in the preset vibration value interval is marked as normal data. At the same time, when marking the data one by one, the previous and next data values ​​of each data are also compared. If the values ​​are the same, the data traversed later are marked as duplicate data. For an explanation of the data marking method, see Table 1. Assuming that the preset vibration value interval is 5~10,

[0069] Table 1 Data processing table of the data marking method of this embodiment

[0070]

[0071] By eliminating abnormal and duplicate data, the normal data obtained is integrated into real-time ship vibration monitoring data.

[0072] For the noise data, the collected second sensor data is converted into a noise spectrum using fast Fourier transform (FFT), and the abnormal prominent values ​​and blank values ​​in the spectrum are deleted to obtain an optimized noise spectrum. Figure 1 The corresponding data are converted into data values ​​to obtain the real-time noise monitoring data of the ship; finally, the real-time operation data of the ship is constructed through the real-time vibration monitoring data and the real-time noise monitoring data of the ship. The real-time operation data of the ship will be continuously recorded in the ship operation database, and the data will be extracted from the database when needed to obtain the historical operation data of the ship.

[0073] Step 102, based on a preset feature analysis algorithm, feature analysis is performed on the real-time ship operation data and the historical ship operation data to obtain a first feature set and a second feature set;

[0074] As an example of this embodiment, time weight values ​​are added to the real-time operation data of the ship and the historical operation data of the ship respectively, and the real-time operation feature data and the historical operation feature data corresponding to the real-time operation data of the ship and the historical operation data of the ship within a number of preset time intervals are extracted to construct a first operation data feature vector and a second operation data feature vector; the feature vectors whose contribution rates of the first operation data feature vector and the second operation data feature vector are less than or equal to a preset evaluation threshold are eliminated through the principal component analysis method to obtain a first decomposition feature vector and a second decomposition feature vector; based on a preset sliding window, the first decomposition feature vector and the second decomposition feature vector are traversed by sliding with a first unit length to obtain several groups of first feature comprehensive data and several groups of second feature comprehensive data; based on the several groups of first feature comprehensive data and several groups of second feature comprehensive data, a first feature set and a second feature set are constructed. A specific implementation method is to perform time series analysis on the real-time operation data of the ship and the historical operation data of the ship respectively through time series analysis and data mining technology, and then extract valid data in a specific time interval (equivalent to real-time operation feature data and historical operation feature data), construct a first operation data feature vector and a second operation data feature vector, and then use a feature decomposition method such as principal component analysis (PCA) to score the contribution rate of each of the first operation data feature vector and the second operation data feature vector, perform linear transformation on the first operation data feature vector and the second operation data feature vector, project the original data into a new coordinate system, the principal component represents the direction with the largest contribution rate score in the data, extract the first few principal components as decomposition feature data, obtain the first decomposition feature vector and the second decomposition feature vector, use a sliding window or recursive segmentation technology to perform multi-scale time series segmentation on the decomposition feature data, set multiple time windows of different sizes, slide the feature data, and extract the data fragments in each window. Figure 2 , Figure 2 A schematic diagram of a sliding window processing data flow of a ship vibration reduction and noise reduction method provided by an embodiment of the present invention; Figure 2 As shown, in one of the examples of this embodiment, the length of the first unit is set to 3 data units, the data sequence is traversed one by one, and a new data is obtained after every three data are extracted (equivalent to a group of first feature comprehensive data and several groups of second feature comprehensive data). By changing the length of the time window, several groups of first feature comprehensive data and several groups of second feature comprehensive data are obtained to construct the first feature set and the second feature set.

[0075] Step 103, based on a preset factor identification algorithm, extract key factors and associate factors from the first feature set and the second feature set to obtain key factor factors;

[0076] As an example of this embodiment, based on a preset factor identification algorithm, the key factor labels corresponding to each feature data in the first feature set and the second feature set are identified to obtain the first key factor set and the second key factor set; the correlation coefficient between the first key factor set and the second key factor set is calculated, and the key factor with the largest correlation coefficient is obtained as the key factor factor; wherein the correlation coefficient calculation formula is as follows:

[0077]

[0078] In the formula, R is the correlation coefficient, Y min is the minimum difference, Y max is the maximum difference, Y real is the actual difference; the actual difference Y real is the difference between the kth data point of the i-th data set in the first key factor set and the kth data point of the i-th data set in the second key factor set. The minimum difference Y min and the maximum difference Y max is the minimum and maximum difference between the data points corresponding to all data sets of the first key factor set and the second key factor set, is a sensitivity coefficient, ranging from 0 to 1, and usually taking a value of 0.5. A specific implementation method is to identify key factors for the key factor labels corresponding to each feature data in the first feature set and the second feature set, and the key factor labels include: active vibration, propeller vibration, wave and water flow excitation, propeller rotation, mechanical operation, etc. After identifying the key factors for each feature data, the correlation coefficient is calculated to obtain the correlation coefficient between the first key factor set and the second key factor set, and then the key factor factor is extracted.

[0079] Step 104, setting a plurality of different working condition training channels for the preset ship operation simulation model, inputting the key factors into the preset ship operation simulation model to perform simulation training under different working conditions, and obtaining vibration responses and noise responses under different working conditions;

[0080] As an example of this embodiment, the key factor is input into a preset ship operation simulation model; wherein the preset ship operation simulation model sets several training channels for different working conditions; based on a preset convolution algorithm and an attention mechanism algorithm, simulations under different working conditions are performed through the preset ship operation simulation model to obtain vibration responses and noise responses under different working conditions.

[0081] As another example of this embodiment, based on a multi-column convolution algorithm, training channels for several different working conditions are matched; wherein, the training channel includes: a convolution layer, a function activation layer, a simulation training layer, an attention mechanism algorithm layer and an output layer; based on the convolution layer, feature recognition is performed on the key factor to obtain first training feature data; based on the function activation layer, a weight coefficient is calculated for the first training feature data through a preset activation function, and the calculated weight coefficient is multiplied by the key factor to obtain second training feature data; based on the attention mechanism algorithm layer, the product of the second training feature data and the operating weight values ​​of different working conditions is calculated to obtain third training feature data; through the simulation training layer, the third training feature data is subjected to operation simulation training, and the vibration response and noise response under different working conditions are output through the output layer. For a specific possible implementation method, see. Figure 3 , Figure 3 A schematic diagram of data processing flow under different working conditions of a ship vibration reduction and noise reduction method provided by an embodiment of the present invention; Figure 3 As shown, the key factors are input into several different working conditions for training. In each channel, the input data is identified by the convolution layer to obtain the corresponding first training feature data. After convolution, it is input into the function activation layer to calculate the weight coefficient and multiplied with the key factors to obtain the corresponding second training feature data. A set of feature maps are constructed, and the feature maps are input into the attention mechanism algorithm layer. In this embodiment, the ECA attention mechanism is used for interpretation. The feature maps are subjected to secondary feature extraction through the ECA attention mechanism, that is, the product of the second training feature data and the weight value of the operation under different working conditions is calculated to obtain the corresponding third training feature data, which is then input into the simulation training layer to simulate the operation status of the ship under the real operation conditions, obtain the operation parameters of the ship, obtain the vibration response and noise response under different working conditions and output them through the output layer.

[0082] Step 105, updating the key factor factors until the vibration response and noise response under the different working conditions meet the preset operation requirements, and obtaining the target vibration response and target noise response;

[0083] As an example of this embodiment, based on a preset update time interval, the real-time operation data of the ship is updated in real time, and the key factors are updated synchronously; whenever the vibration response is lower than the preset vibration frequency and the noise response is lower than the preset decibel value, the key factors currently updated are recorded as the target key factors; the corresponding vibration response and noise response generated by the simulation training of the target key factors are output as the target vibration response and the target noise response. A specific implementation method is to update the real-time operation data of the ship based on a preset update time interval, such as 5s, repeat steps 101 to 103, update the key factors, compare each updated key factor with the preset vibration frequency and the preset decibel value, and whenever the vibration response is lower than the preset vibration frequency and the noise response is lower than the preset decibel value, record the key factors updated at the current moment as the target key factors; through continuous updating, a number of target key factors are obtained, and step 104 is performed on the several target key factors in each update interval to obtain the target vibration response and the target noise response through simulation training.

[0084] Step 106: Based on the target vibration response and the target noise response, a target ship operation strategy is generated and ship operation parameters are adjusted to reduce vibration and noise of the ship.

[0085] As an example of this embodiment, based on the target vibration response and the target noise response, the corresponding target key factor is identified, and the real-time operation data of the ship and the historical operation data of the ship corresponding to the target key factor are obtained; the historical operation parameters corresponding to the historical operation data of the ship are mapped to the real-time operation data of the ship, and a target operation strategy is formulated through the historical operation parameters; based on the target operation strategy, the target operation parameters of the ship are adjusted. A specific explanation is that by mapping the operation strategy corresponding to the historical operation data of the ship to the real-time operation data of the ship with high correlation, a target operation strategy is formulated, and the target operation parameters of the ship are adjusted by the target operation strategy; more specifically, based on the target vibration response and target noise response, the natural frequency or excitation frequency of the structure can be changed to prevent the occurrence of resonance. For example, this can be achieved by adjusting the design and structure of the ship to ensure that the natural frequency of the ship is different from the external excitation frequency such as waves, thereby avoiding the amplification of oscillation and noise; then dynamic balancing or structural modification can be performed to reduce the excitation amplitude, such as including optimizing the design of the ship's power machinery and auxiliary machinery to reduce unbalanced forces, thereby reducing the generation of vibration and noise; the transmission of oscillation and noise can also be reduced by increasing the damping force. For example, this can be achieved by adding damping materials or devices to the ship structure to effectively absorb and consume vibration energy, thereby reducing noise and oscillation. ‌‌

[0086] The embodiment of the present invention proposes a ship vibration reduction and noise reduction method, which increases the data basis and ensures the applicability of the optimization scheme by acquiring real-time operation data and historical operation data and performing feature analysis respectively, and associates the real-time operation data with the historical operation data by combining key factor extraction and joint technology, thereby improving the reliability of data analysis; then, by simulating the operation of the ship under different working conditions, the vibration source and noise source under different working conditions are analyzed at multiple levels, so that targeted optimization schemes can be made, and the reliability of the vibration reduction and noise reduction optimization scheme is improved; and by continuously updating key factors, the operation data is continuously iterated, the vibration and noise responses of the target are screened, and the target operation strategy is generated based on the vibration and noise responses of the target, and the operation parameters of the ship are adjusted, so that the ship can achieve vibration reduction and noise reduction during operation; therefore, the present invention provides multi-scale data feature analysis combined with key factors, simulates the operation of the ship under different working conditions, generates a target operation strategy, and improves the reliability of the ship vibration reduction and noise reduction optimization scheme.

[0087] Example 2

[0088] See also Figure 4 , Figure 4 The following is a schematic diagram of the module structure of a ship vibration reduction and noise reduction system provided by an embodiment of the present invention. Figure 4 As shown, an embodiment of the present invention provides a ship vibration reduction and noise reduction system, comprising:

[0089] Data acquisition module 401, feature analysis module 402, factor identification module 403, working condition simulation module 404, factor update module 405 and vibration reduction and noise reduction module 406;

[0090] The data acquisition module 401 is used to acquire real-time ship operation data and historical ship operation data;

[0091] As an example of this embodiment, see Figure 5 , Figure 5 A schematic diagram of a data acquisition module structure of a ship vibration and noise reduction system provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, the data acquisition module 401 is used to acquire real-time ship operation data and historical ship operation data; including:

[0092] A real-time vibration data acquisition unit 501, a real-time noise data acquisition unit 502, a real-time operation data construction unit 503 and a historical operation data construction unit 504;

[0093] The real-time vibration data acquisition unit 501 is used to acquire first sensor data in real time through a vibration sensor, traverse the first sensor data, mark the data that meets the preset vibration value interval as normal data, mark the data that does not meet the preset vibration value interval as abnormal data, mark the data with the same value as the last traversed data as duplicate data, remove the abnormal data and duplicate data, and integrate the normal data to obtain the real-time vibration monitoring data of the ship;

[0094] The real-time noise data acquisition unit 502 is used to acquire the second sensing data in real time through the noise sensor, and construct a noise spectrum based on the second sensing data, remove prominent points and blank points in the spectrum, and then convert the processed spectrum into corresponding data values ​​to obtain real-time ship noise monitoring data;

[0095] The real-time operation data construction unit 503 is used to integrate the ship real-time vibration monitoring data and the ship real-time noise monitoring data to obtain the ship real-time operation data;

[0096] The historical operation data construction unit 504 is used to read the historical vibration monitoring data and the historical noise monitoring data of the ship based on the historical operation database of the ship to obtain the historical operation data of the ship.

[0097] The feature analysis module 402 is used to perform feature analysis on the real-time ship operation data and the historical ship operation data based on a preset feature analysis algorithm to obtain a first feature set and a second feature set;

[0098] The factor identification module 403 is used to extract key factors and associate factors on the first feature set and the second feature set based on a preset factor identification algorithm to obtain key factor factors;

[0099] The working condition simulation module 404 is used to set a number of different working condition training channels for the preset ship operation simulation model, input the key factors into the preset ship operation simulation model to perform simulation training under different working conditions, and obtain vibration response and noise response under different working conditions;

[0100] The factor updating module 405 is used to update the key factor factors until the vibration response and noise response under the different working conditions meet the preset operation requirements, and obtain the target vibration response and target noise response;

[0101] The vibration reduction and noise reduction module 406 is used to generate a target ship operation strategy and adjust ship operation parameters based on the target vibration response and the target noise response to reduce vibration and noise of the ship.

[0102] The embodiment of the present invention proposes a ship vibration reduction and noise reduction system, which acquires real-time operation data and historical operation data through a data acquisition module and a feature analysis module and performs feature analysis respectively, thereby increasing the data basis and ensuring the applicability of the optimization scheme. The real-time operation data is associated with the historical operation data through a factor identification module combined with key factor extraction and joint technology, thereby improving the reliability of data analysis. The ship operation under different working conditions is simulated through the factor identification module, and the vibration source and noise source under different working conditions are analyzed at multiple levels, so that a targeted optimization scheme can be made, thereby improving the reliability of the vibration reduction and noise reduction optimization scheme. The key factor factors are continuously updated through the factor update module, so that the operation data is continuously iterated, the vibration and noise response of the target are screened, and the target operation strategy is generated based on the vibration and noise response of the target through the vibration reduction and noise reduction module, and the ship's operation parameters are adjusted, so that the ship can achieve vibration reduction and noise reduction during operation. Therefore, the present invention provides multi-scale data feature analysis combined with key factors, simulates the ship operation under different working conditions, generates a target operation strategy, and improves the reliability of the ship vibration reduction and noise reduction optimization scheme.

[0103] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

[0104] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0105] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

Claims

1. A method for reducing vibration and noise of a ship, characterized in that: include: Obtain real-time ship operation data and historical ship operation data; Based on a preset feature analysis algorithm, feature analysis is performed on the real-time operation data of the ship and the historical operation data of the ship to obtain a first feature set and a second feature set; Based on a preset factor identification algorithm, extract key factors and associate factors from the first feature set and the second feature set to obtain key factor factors; A plurality of different working condition training channels are set for the preset ship operation simulation model, and the key factors are input into the preset ship operation simulation model to perform simulation training under different working conditions, so as to obtain vibration response and noise response under different working conditions; Updating the key factor factors until the vibration response and the noise response under the different working conditions meet the preset operation requirements, and obtaining the target vibration response and the target noise response; Based on the target vibration response and the target noise response, a target ship operation strategy is generated and ship operation parameters are adjusted to reduce vibration and noise of the ship; The acquisition of real-time ship operation data and historical ship operation data includes: The first sensor data is acquired in real time by the vibration sensor, the first sensor data is traversed, and the data that meets the preset vibration value interval is marked as normal data, the data that does not meet the preset vibration value interval is marked as abnormal data, and the data with the same value as the last traversed data is marked as duplicate data, the abnormal data and duplicate data are eliminated, and the normal data is integrated to obtain the real-time vibration monitoring data of the ship; The second sensing data is acquired in real time by the noise sensor, and a noise spectrum is constructed based on the second sensing data, prominent points and blank points in the spectrum are removed, and the processed spectrum is converted into corresponding data values ​​to obtain real-time noise monitoring data of the ship; Integrate the ship's real-time vibration monitoring data and the ship's real-time noise monitoring data to obtain the ship's real-time operation data; Based on the ship's historical operation database, read the ship's historical vibration monitoring data and the ship's historical noise monitoring data to obtain the ship's historical operation data; The method of performing feature analysis on the real-time ship operation data and the historical ship operation data based on a preset feature analysis algorithm to obtain a first feature set and a second feature set includes: respectively adding time weight values ​​to the real-time operation data of the ship and the historical operation data of the ship, extracting real-time operation feature data and historical operation feature data corresponding to the real-time operation data of the ship and the historical operation data of the ship within a plurality of preset time intervals, and constructing a first operation data feature vector and a second operation data feature vector; Eliminate feature vectors whose contribution rate to the first operation data feature vector and the second operation data feature vector is less than or equal to a preset evaluation threshold value by using a principal component analysis method to obtain a first decomposed feature vector and a second decomposed feature vector; Based on a preset sliding window, slidingly traverse the first decomposed feature vector and the second decomposed feature vector with a first unit length to obtain several groups of first feature comprehensive data and several groups of second feature comprehensive data; Based on the plurality of groups of first feature comprehensive data and the plurality of groups of second feature comprehensive data, construct a first feature set and a second feature set; The step of extracting key factors and associating factors from the first feature set and the second feature set based on a preset factor identification algorithm to obtain key factor factors includes: Based on a preset factor identification algorithm, identifying key factor labels corresponding to each feature data in the first feature set and the second feature set, to obtain a first key factor set and a second key factor set; Calculate the correlation coefficient between the first key factor set and the second key factor set, and obtain the key factor with the largest correlation coefficient as the key factor factor; wherein the correlation coefficient calculation formula is as follows: ； In the formula, R is the correlation coefficient, Y min is the minimum difference, Y max is the maximum difference, Y real is the actual difference; the actual difference Y real is the difference between the kth data point of the i-th data set in the first key factor set and the kth data point of the i-th data set in the second key factor set. The minimum difference Y min and the maximum difference Y max is the minimum and maximum difference between the data points corresponding to all data sets of the first key factor set and the second key factor set, It is the sensitivity coefficient, and its value range is between 0 and 1.

2. A ship vibration reduction and noise reduction method according to claim 1, characterized in that: The preset ship operation simulation model is set with a plurality of different working condition training channels, and the key factors are input into the preset ship operation simulation model for different working condition simulation training to obtain vibration response and noise response under different working conditions, including: Inputting the key factors into a preset ship operation simulation model; wherein the preset ship operation simulation model is provided with a plurality of training channels of different working conditions; Based on the preset convolution algorithm and attention mechanism algorithm, the simulation under different working conditions is carried out through the preset ship operation simulation model to obtain the vibration response and noise response under different working conditions.

3. A ship vibration reduction and noise reduction method according to claim 2, characterized in that: Based on the preset convolution algorithm and attention mechanism algorithm, the simulation under different working conditions is performed through the preset ship operation simulation model to obtain the vibration response and noise response under different working conditions, including: Based on a multi-column convolution algorithm, a training channel is matched with several different working conditions; wherein the training channel includes: a convolution layer, a function activation layer, a simulation training layer, an attention mechanism algorithm layer and an output layer; Based on the convolution layer, feature recognition is performed on the key factor to obtain first training feature data; Based on the function activation layer, a weight coefficient is calculated for the first training feature data by a preset activation function, and the calculated weight coefficient is multiplied by the key factor to obtain second training feature data; Based on the attention mechanism algorithm layer, the product of the second training feature data and the weight values ​​of different operating conditions is calculated to obtain the third training feature data; The simulation training layer is used to perform running simulation training on the third training feature data, and the vibration response and noise response under different working conditions are output through the output layer.

4. A ship vibration reduction and noise reduction method according to claim 1, characterized in that: Updating the key factor factors until the vibration response and the noise response under the different working conditions meet the preset operation requirements, and obtaining the target vibration response and the target noise response, including: Based on a preset update time interval, the real-time operation data of the ship is updated in real time, and the key factors are updated synchronously; Whenever the vibration response is lower than the preset vibration frequency and the noise response is lower than the preset decibel value at the same time, the currently updated key factor is recorded as the target key factor; The corresponding vibration response and noise response generated by simulation training of the target key factor are output as the target vibration response and the target noise response.

5. A ship vibration reduction and noise reduction method according to claim 1, characterized in that: Based on the target vibration response and the target noise response, a target ship operation strategy is generated and the target ship operation parameters are adjusted to reduce vibration and noise of the ship, including: Based on the target vibration response and the target noise response, the corresponding target key factor is identified, and the real-time ship operation data and the historical ship operation data corresponding to the target key factor are obtained; Map the historical operation parameters corresponding to the historical operation data of the ship to the real-time operation data of the ship, and formulate the target operation strategy based on the historical operation parameters; Adjust the ship's target operating parameters based on the target operation strategy.

6. A ship vibration and noise reduction system, characterized in that: A ship vibration and noise reduction method as claimed in any one of claims 1 to 5, comprising: Data acquisition module, feature analysis module, factor identification module, working condition simulation module, factor update module and vibration and noise reduction module; The data acquisition module is used to acquire real-time ship operation data and historical ship operation data; The feature analysis module is used to perform feature analysis on the real-time ship operation data and the historical ship operation data based on a preset feature analysis algorithm to obtain a first feature set and a second feature set; The factor identification module is used to extract key factors and associate factors on the first feature set and the second feature set based on a preset factor identification algorithm to obtain key factor factors; The working condition simulation module is used to set a number of different working condition training channels for the preset ship operation simulation model, input the key factors into the preset ship operation simulation model to perform simulation training under different working conditions, and obtain vibration responses and noise responses under different working conditions; The factor updating module is used to update the key factor factors until the vibration response and noise response under the different working conditions meet the preset operation requirements, thereby obtaining the target vibration response and the target noise response; The vibration reduction and noise reduction module is used to generate a target ship operation strategy and adjust ship operation parameters based on the target vibration response and the target noise response, so as to reduce vibration and noise of the ship; The data acquisition module is used to acquire real-time ship operation data and historical ship operation data; including: A real-time vibration data acquisition unit, a real-time noise data acquisition unit, a real-time operation data construction unit and a historical operation data construction unit; The real-time vibration data acquisition unit is used to acquire first sensor data in real time through a vibration sensor, traverse the first sensor data, mark the data that meets the preset vibration value interval as normal data, mark the data that does not meet the preset vibration value interval as abnormal data, mark the data with the same value as the last traversed data as duplicate data, remove the abnormal data and duplicate data, and integrate the normal data to obtain the real-time vibration monitoring data of the ship; The real-time noise data acquisition unit is used to acquire the second sensing data in real time through the noise sensor, and construct a noise spectrum based on the second sensing data, remove prominent points and blank points in the spectrum, and then convert the processed spectrum into corresponding data values ​​to obtain real-time noise monitoring data of the ship; The real-time operation data construction unit is used to integrate the real-time vibration monitoring data of the ship and the real-time noise monitoring data of the ship to obtain the real-time operation data of the ship; The historical operation data construction unit is used to read the historical vibration monitoring data and the historical noise monitoring data of the ship based on the historical operation database of the ship to obtain the historical operation data of the ship; The method of performing feature analysis on the real-time ship operation data and the historical ship operation data based on a preset feature analysis algorithm to obtain a first feature set and a second feature set includes: respectively adding time weight values ​​to the real-time operation data of the ship and the historical operation data of the ship, extracting real-time operation feature data and historical operation feature data corresponding to the real-time operation data of the ship and the historical operation data of the ship within a plurality of preset time intervals, and constructing a first operation data feature vector and a second operation data feature vector; Eliminate feature vectors whose contribution rate to the first operation data feature vector and the second operation data feature vector is less than or equal to a preset evaluation threshold value by using a principal component analysis method to obtain a first decomposed feature vector and a second decomposed feature vector; Based on a preset sliding window, slidingly traverse the first decomposed feature vector and the second decomposed feature vector with a first unit length to obtain several groups of first feature comprehensive data and several groups of second feature comprehensive data; Based on the plurality of groups of first feature comprehensive data and the plurality of groups of second feature comprehensive data, construct a first feature set and a second feature set; The step of extracting key factors and associating factors from the first feature set and the second feature set based on a preset factor identification algorithm to obtain key factor factors includes: Based on a preset factor identification algorithm, identifying key factor labels corresponding to each feature data in the first feature set and the second feature set, to obtain a first key factor set and a second key factor set; Calculate the correlation coefficient between the first key factor set and the second key factor set, and obtain the key factor with the largest correlation coefficient as the key factor factor; wherein the correlation coefficient calculation formula is as follows: ； In the formula, R is the correlation coefficient, Y min is the minimum difference, Y max is the maximum difference, Y real is the actual difference; the actual difference Y real is the difference between the kth data point of the i-th data set in the first key factor set and the kth data point of the i-th data set in the second key factor set. The minimum difference Y min and the maximum difference Y max is the minimum and maximum difference between the data points corresponding to all data sets of the first key factor set and the second key factor set, It is the sensitivity coefficient, and its value range is between 0 and 1.

Images