42 results about "Ship noise" patented technology

The invention discloses a deep convolutional network-based ship noise identification and classification method, and belongs to the field of underwater ship noise identification. A solution is proposedmainly for the problems of small file processing quantity, unremarkable extracted features and easy falling into local optimal solution in a BP neural network. According to the method, firstly according to MFCC, noises are removed in an original sound; corresponding effective features are extracted; and in the process, the noises with high interference are removed. A sound file processed throughthe MFCC is converted to be in a format receivable for a deep convolutional network. The effective features are extracted in a multilayer manner through the deep convolutional neural network; and theextracted features are higher in validity and higher in universality. A large amount of sound data in reality is identified and classified, so that the manual intervention degree is reduced, differentship noises can be better distinguished, and the purpose of identification is achieved.

The invention discloses an installation method of a main engine with an elastic support, the main engine comprises a main engine carrier block and a main engine base, the main engine carrier block is connected with the main engine base through the elastic support, and the elastic support comprises a vertical shock absorbing block, a side shock absorbing block and a longitudinal shock absorbing block. The elastic support is arranged between the main engine carrier block and the base, and the vibration generated during the operation of the main engine is absorbed by a rubber damping device of the rubber shock absorbing block, so that the vibration and the noise are weakened. The use of the installation method can shorten the overall installation period of the ship main engine and greatly enhance the progress. The method is applicable to ships with higher requirements on the ship noise and the vibration indicators, so as to obtain good working and living environment of the ships and the comfort of passengers.

The hydroacoustic noise detecting and distinguishing method based on wavelet domain includes the first mean value-eliminating and amplitude normalizing treatment of ship noise and marine environment noise, and the subsequent multiple scale domain decomposing of the wavelet domain and establishing implicit wavelet domain Markov model. Based on the characteristic difference between marine environment noise model and ship radiation noise model, one detection method based on Nyman-Pearson criterion is proposed. The likelihood value between the ship noise and the marine environment noise HMT model is used as the inspection statistic for binary hypothesis inspection to raise the ship noise detecting capacity. One characteristic vector describing the detection sample is formed for judging ship noise type, resulting in raised identification rate and robustness.

The invention provides an underground ship noise characteristic extracting method based on IMF energy entropy and PCA. The method comprises steps of extracting time domain characteristic parameters of the average, the variance, the peak value and the skewness of ship noise signals and constructing a time domain characteristic matrix; using the EEMD method to decompose the original ship noise signals to obtain IMF components, converting the IMF components into energy characteristic vectors, thereby observing change of frequency band energy characteristics, and constructing a frequency domain characteristic variable to form a characteristic matrix A together with the time domain characteristic vectors; using the PCA dimension reduction method to carry out dimension reduction processing on the characteristic matrix A, and mapping the characteristic matrix with the high dimension into characteristic matrix with the low dimension to serve as a new characteristic so as form a characteristic matrix B; and training LSSVM parameters, according the risk minimization principle, adjusting the parameter gamma and the kernel width sigma, finally inputting the characteristic matrix B subjected to the dimension reduction into a classifier and testing a classification result. According to the invention, time domain information can be combined with the frequency domain information and information complementation is formed, so quite complete characteristic information can be provided and classification precision can be improved.

The invention discloses an autonomous extracting method for a non-stable low-frequency line spectrum of ship noise. The method comprises the following steps of (1), performing ship noise long-integration power spectrum line spectrum first detection, and obtaining a suspected line spectrum; (2), for aiming at the suspected line spectrum, setting the number of segments in segmented power spectrum analysis and the interval length of the integration domain; (3), for aiming at the suspected line spectrum, setting the integration domain starting point in analysis of each segmented power spectrum; (4), performing ship noise secondary power spectrum estimation according to the preset integration domain starting point and the interval length; and (5), performing continuous spectrum automatic offsetand line spectrum autonomous identification extraction. The method aims at a difficulty of autonomous extraction for the non-stable low-frequency line spectrum of the ship noise; through automatic analysis for a non-stable characteristic of the low-frequency line spectrum of the ship noise, adaptive adjustment is performed on the number of segments in segmented power spectrum analysis, the integration domain starting point and the interval length, thereby improving tolerance of the power spectrum to the line spectrum non-stable characteristic, improving signal-to-noise ratio of the line spectrum, and realizing autonomous extraction to the non-stable low-frequency line spectrum of the ship noise through automatic continuous spectrum offset and autonomous line spectrum identification.

The invention provides a system and a method for supplying natural energy as power to an electric ship. The embodiment of the invention transforms solar energy and wind energy into electrical energy through a wind-light mixed power station, and is used by a wharf energy-adding station and used for charging a storage battery container hold of the electric ship; at the same time, the system is also connected with a commercial-power electric-fence; when the weather is fine in the daytime and the wind is heavy, the generated energy of the wind-light mixed power station is sufficient, the excess electric quantity can be merged into the commercial-power electric-fence for the use of other loads; when the weather is wet at night or in the daytime and the wind is light, the alternating current of the commercial-power electric-fence can be transformed into direct current so as to be used by the wharf energy-adding station and also used for charging the storage battery container hold of the electric ship; the wind-light mixed power station and the commercial-power electric-fence complement each other, so the natural energy supplied as power to the electric ship stably and reliably, the dependence on non-renewable energy resources such as petroleum, coals, etc. as well as the ship noise pollution can be reduced, energy crisis is relieved, and the atmosphere and the marine environment are protected.

The invention discloses a ship classification method based on vibration noise identification, and the method comprises the following steps: 1, collecting ship vibration noise, and converting the shipvibration noise into a time domain signal; 2, performing improved ensemble empirical mode decomposition on the ship vibration signals; 3, performing Hilbert-Huang transform on the signal after modal decomposition; 4, performing Hilbert spectrum analysis on ship radiation noise; 5, extracting ship noise characteristics; 6, classifying the ship noise characteristics by using a ship classifier basedon a support vector machine; 7, verifying the accuracy of the ship classification result, and if the result is correct, ending ship classification; and if not, entering the ship classifier for re-classification. The ship classification method based on vibration noise identification is realized, and a new way for marine traffic management personnel, scientific research personnel and military personnel to acquire ship types is provided.

The invention relates to the technical field of shock and noise reduction for ships, and more specifically, a floating floor, which is simple in design, convenient in construction, effectively shortens the shipbuilding cycle, and has various technical indicators that fully meet various indicators such as ship vibration and noise reduction, fire prevention, etc. Requirements: The noise index of the crew cabin and working area is lower than the maximum allowable noise level, the vibration is lower than the maximum allowable vibration level, the comfort level can reach the first grade, and the surface material meets the requirements of the SOLAS international convention for low flame spread, According to the special requirements of different parts of the ship, the fire rating can reach A60. On the deck, there are damping layer, constraining layer, rock wool II, galvanized steel plate I, damping layer, galvanized steel plate II and base dressing in order from bottom to top; Above, rock wool I is in contact with the damping layer, constraining layer, rock wool II, galvanized steel plate I, damping layer, galvanized steel plate II and the base dressing, and the water-blocking flat steel is in contact with rock wool I.

The invention relates to a ship noise identification method based on holographic spectrum and deep learning, and belongs to the field of signal processing. The method comprises the steps of firstly, carrying out standardization on an extracted signal, and then carrying out calculation through employing an empirical mode decomposition method to obtain an amplitude modulation holographic spectrum of the standardized signal; taking the amplitude modulation holographic spectrogram as feature input of a convolutional neural network classification model, and training the classification model; and using the trained classification model to classify and identify the ship noise. According to the method, the amplitude modulation holographic spectrogram is used as the feature of the ship noise, and the amplitude modulation information in the ship noise is obtained through the amplitude modulation holographic spectrum. The method is more stable in the complex and changeable marine environment.

The invention discloses a test method for simulating influence of underwater noise vibration of ships and water transport engineering on fishes, and belongs to the technical field of ecological environment influence. Underwater noise vibration caused by ship noise and water transport engineering construction such as piling is simulated by a vibration sounding device and is transferred to a fish culture tank, and then the type and degree of influence on fishes by noise vibration are observed and analyzed. The main test method comprises the following steps: preparing a noise vibration source, testing a noise vibration propagation process, selecting a noise vibration influence receptor, and carrying out test analysis. The method can provide technical support for guiding shipbuilding and navigation and water transport engineering design and construction to adopt appropriate fish influence mitigation countermeasures and evaluating ecological environment influence.

The invention discloses a movable lift quad-hull type underwater measuring platform, which belongs to the technical field of marine monitoring and underwater acoustic measurement. The platform comprises pontoon modules, an upper frame, a hydrophone mounting bracket, underwater winches, a varying-voltage cabinet, a control cabinet, an attitude course pressure sensor and anchoring modules. Four pontoon modules are respectively symmetrically arranged at four corners of the bottom of the upper frame. Four underwater winches are respectively connected with the underwater winch mounting base of the pontoon modules and are respectively connected with four anchoring modules. The varying-voltage cabinet and the control cabinet are connected with the buoyant adjustment pontoon of a pontoon module respectively. The hydrophone mounting bracket is arranged at the top surface of the upper frame. The attitude course pressure sensor is arranged at the bottom of the beam of the H-shaped frame of the upper frame. According to the invention, deployment and recycling of predetermined depth of water in a predetermined sea area can be completed; the platform can maintain a stable attitude under the water; and ship noise measurement is smoothly completed by carrying an acoustic measuring device.

The invention relates to a robust ship noise identifying method, and belongs to the field of marine signal processing. The method comprises the following steps: firstly an extracted signal is standardized, and then filter and Fourier transform combined processing is used to obtain a modulation spectrum of the signal; secondly, a modulation significance statistic quantity is calculated by dividinga value of a spectrum line with the maximum amplitude in the modulation spectrum by an average amplitude of the modulation spectrum; and finally, the modulation significance statistic quantity is compared with a threshold, if the modulation significance statistic quantity is larger than the threshold, a ship is considered to appear, and if the modulation significance statistic quantity is continuously smaller than the threshold in a period of time, no ship is considered to appear around. According to the method disclosed by the invention, a more stable physical variable, namely, a modulation characteristic of ship noise is utilized; and compared with a mode of using signal power to detect the ship noise, the method is more robust and better in anti-interference performance.

The invention provides a ship noise spectrum modulation method and device. The method comprises the following steps: receiving a noise signal collected by a noise measuring instrument disposed on a ship, the noise signal comprises noise emitted by the ship sailing in a target sea area and noise emitted by a noise jammer installed on the ship in the target sea area; obtaining spectrum data of the noise signal; and under the condition that the frequency spectrum data is not matched with the standard frequency spectrum of the environmental noise of the target sea area, adjusting the emission parameters of the noise jammer according to the standard frequency spectrum until the frequency spectrum data is matched with the standard frequency spectrum.

The invention relates to the technical field of ship noise forecasting methods, and concretely relates to a ship propeller noise forecasting method considering fluid-structure interaction. The noise forecasting method is characterized by comprising the following steps: 1, establishing a propeller model and a ship appearance model according to a ship body design drawing, a profile drawing and a propeller profile value table; 2, combining the propeller model and the ship body shape model, and performing non-fluid-structure interaction transient flow field calculation; 3, performing mesh generation, material attribute endowing and boundary condition setting on the ship body structure model; 4, calculating a fluid-structure interaction transient structure of the ship body and the propeller; and 5, considering the noise forecasting under the fluid-structure interaction condition. The noise forecasting method is adopted to achieve the advantages of being simple, small in noise forecasting error, high in noise forecasting effect and the like.

The invention relates to a sub-space direction estimation method based on a linear ambient noise model. For an actual noise environment, because of influence of wind-generated noise and ship noise, spatial noise intensity distribution is non-uniform. A linear noise model is used for modeling. Fourier series expansion is performed on a spatial noise power distribution function. A finite item Fourier grade number is used for approximately fitting an ambient noise model. A signal subspace in a linear noise model condition is extracted. Orthogonality between the signal sub-space and the noise sub-space is utilized for obtaining direction estimation of a target.

The invention relates to the technical field of ship transportation, in particular to a ship noise control method. A waste gas header pipe is installed between a cylinder waste gas outlet of a diesel engine and a supercharger, low-frequency noise is weakened, high-frequency noise is weakened through a silencer below a deck, noise current is transmitted to the seabed in a radiation mode through a transmission antenna, the probability of being detected is reduced, parameter signals are compared with ship body noise, error noise is obtained, and the noise is detected. The transducer is driven to emit error noise, noise around the ship body below the water surface is counteracted, and acoustic stealth of the ship is achieved.

The invention discloses a riverway ship traffic analysis system and a riverway ship traffic analysis method. The system comprises an underwater sound sensor, a transmission unit, a processing unit andan image acquisition unit, the underwater sound sensor is arranged on a riverbed, the image acquisition unit is arranged at the upstream and the downstream of the position where the underwater soundsensor is located, and the image acquisition unit and the underwater sound sensor are connected with the processing unit through the transmission unit. The riverway ship traffic analysis method basedon the system comprises the steps: detecting and transmitting ship noise to the processing unit; converting data into an oscillogram; intercepting a detection waveform; selecting a plurality of detection time points, and intercepting ship noise; establishing a curve graph; and S3, calulating the number of wave crests in each curve graph, obtaining an average ship traffic number, and returning to the step S3. The system has the advantages that the image acquisition unit only detects whether a ship enters and exits or not, and the requirement for the visibility of a river channel is low; image acquisition and sound detection are combined, and the ship traffic detection accuracy is high.

The invention discloses a ship noise reduction device. The ship noise reduction device is characterized by comprising a carrying board 1, a hollow damping mass block 2, a sound absorption material 3 and a band board 4. The hollow damping mass block 2 is installed on the carrying board 1. The middle portion of the carrying board is broken and the hollow damping mass block is fixed to the position between the two parts of the carrying board. The band board 4 is fixed to the hollow damping mass block 2. The sound absorption material 3 is installed on the band board 4. By the adoption of the ship noise reduction device, vibration and sound radiation in an engine compartment section of a ship are effectively reduced, the benefit ratio is high, economic cost is saved, the quality is guaranteed, and the production yield is increased.

The invention discloses a ship noise forecasting method and system based on an equivalent surface source and an additional point source and relates to the related technical field of ship overall acoustic design. The method comprises steps of determining a virtual boundary line, determining a virtual sound source on the virtual boundary line based on a wave superposition theory; respectively constructing a surface sound source sound pressure model and a point sound source sound pressure model; constructing a surface sound source-point sound source sound pressure model according to the surface sound source sound pressure model and the point sound source sound pressure model; obtaining the optimal number and the optimal position data of the additional point sources; and forecasting the self-noise of the broadside platform according to the obtained relationship between the sound pressure and the normal component of the vibration velocity of the ship shell. Through the combination of the equivalent surface source and the additional point source, calculation accuracy is improved, the calculation amplitude is reduced, and calculation efficiency is improved; through learning of the neural network, the capability of quickly forecasting the self-noise of the ship shell sonar platform area is improved, and an important method support is provided for developing the self-noise control of the ship side sonar platform area.

The invention discloses a ship noise reduction detachable ventilator, and relates to the technical field of ship ventilators, the ship noise reduction detachable ventilator comprises a fixed cylinder, an air cap cylinder arranged on the top of the fixed cylinder and an air cap cover arranged on the top of the air cap cylinder, the bottom of the air cap cylinder is connected with a vertical air inlet pipe, the top of the air cap cover is provided with a ventilation pipe, and the upper part of the fixed cylinder is internally provided with a fan; a silencing assembly is arranged in the fixing cylinder, and a ventilation mechanism is arranged on the outer wall of the stabilizing cylinder. A plugging mechanism for closing the air inlet pipe is arranged in the middle of each ventilation opening; through cooperation of the sealing cover, the plugging mechanism and the ventilation mechanism, the ship can have two different ventilation operation modes during ventilation operation, different ventilation operations can be switched conveniently according to different weather conditions, the protection effect of the ship during ventilation operation is effectively improved, double sealing operation can be conducted, and the ventilation effect of the ship is improved. And through the arrangement of the silencing assembly and the silencing cotton, the quietness of the device during use is effectively improved.

The invention relates to a ship noise identification method based on holographic spectrum and deep learning, which belongs to the field of signal processing. The method firstly standardizes the extracted signal, and then uses the empirical mode decomposition method to calculate the amplitude modulation hologram of the standardized signal. Spectrum. The amplitude-modulated hologram is used as the feature input of the convolutional neural network classification model to train the classification model. Use the trained classification model to classify and identify ship noise. The method of the invention uses the amplitude modulation hologram as the characteristic of the ship noise, and the amplitude modulation hologram obtains the amplitude modulation information in the ship noise, and is more stable in the complex and changeable ocean environment.

The hydroacoustic noise detecting and distinguishing method based on wavelet domain includes the first mean value-eliminating and amplitude normalizing treatment of ship noise and marine environment noise, and the subsequent multiple scale domain decomposing of the wavelet domain and establishing implicit wavelet domain Markov model. Based on the characteristic difference between marine environment noise model and ship radiation noise model, one detection method based on Nyman-Pearson criterion is proposed. The likelihood value between the ship noise and the marine environment noise HMT model is used as the inspection statistic for binary hypothesis inspection to raise the ship noise detecting capacity. One characteristic vector describing the detection sample is formed for judging ship noise type, resulting in raised identification rate and robustness.

The invention discloses a rotary gas explosion sound source. The sound source comprises a stator, a rotor, a piston and a controller, wherein the rotor is installed in the stator and driven by the motor to rotate relative to the stator, release openings distributed in the circumferential direction are machined in the corresponding positions of the stator and the rotor, the piston is matched with an inner hole of the rotor and located in the stator, the stator is provided with an inflation inlet and an electromagnetic valve connector, the inflation inlet is communicated with an inner cavity ofthe stator, and the electromagnetic valve connector is communicated with a closed cavity defined by the piston and the stator, and the controller controls the pressure in an air cylinder and the movement position of the piston by controlling the on-off of the electromagnetic valve, so that an air gun array formed by different numbers of release openings is controlled to release high-pressure air,and different sound fields are generated to simulate ship noise.

The invention relates to the technical field of cabins, in particular to a novel sailor cabin with a shock absorption and noise reduction structure. A cabin door frame is fixedly installed on the right side of the cabin, damping devices are fixedly installed on the periphery of the inner wall of the cabin door frame, and a sound insulation device is fixedly installed on the right side of the cabin door frame. When a sailor closes the cabin door, external noise is isolated, and therefore the situation that the rest quality of the sailor is poor due to the external noise in the rest process of the sailor is avoided; meanwhile, the cabin door sound insulation mechanism and the cabin door damping component are matched to enable the cabin door to make tight contact with the cabin door sound insulation mechanism, it is guaranteed that sound in a cabin is isolated from sound outside the cabin, damping glue, a composite damping sound insulation plate and a low-frequency sound insulation and damping plate are matched, the ship noise is greatly reduced, and the comfort of sailors in life is effectively improved.

The invention relates to a subspace orientation estimation method based on a linear environmental noise model. For the actual noise environment, due to the influence of wind-generated noise and ship noise, the spatial noise intensity distribution is non-uniform, and the linear noise model is used for modeling. The spatial noise power distribution function is expanded by Fourier series, and the environmental noise model is approximated by using the finite term Fourier series. The signal subspace under the condition of linear noise model is extracted, and the orientation estimation of the target is obtained by using the orthogonality between the signal subspace and the noise subspace.

The invention provides a ship noise reduction plate and a preparation method. The ship noise reduction plate comprises a first metal plate, a first sound absorption and insulation composite plate, a second sound absorption and insulation composite plate, a third sound absorption and insulation composite plate and a second metal plate, which are sequentially connected in the longitudinal direction. The first sound absorption and insulation composite board comprises a plurality of first sound insulation parts and a plurality of first sound absorption parts; the second sound insulation composite board comprises a plurality of second sound insulation parts and a plurality of second sound absorption parts; the third sound absorption and insulation composite board comprises a plurality of third sound insulation parts and a plurality of third sound absorption parts; the first sound insulation part, the second sound absorption part and the third sound insulation part are longitudinally spliced; and the first sound absorption part, the second sound insulation part and the third sound absorption part are longitudinally spliced. The sound absorption and insulation composite board has the beneficial effects that the first sound absorption and insulation composite board, the second sound absorption and insulation composite board and the third sound absorption and insulation composite board are arranged at intervals through the sound absorption pieces and the sound insulation pieces in the transverse direction and the longitudinal direction, and independent noise reduction units are formed; according to the structure, the noise reduction effect is obviously better than that of a traditional noise reduction plate, and noise in different directions can be reduced.