485 results about "Noise level" patented technology

The invention provides a self-adaptive bearing fault diagnosis method based on multi-base wavelet fusion. The objective of the invention is to solve the problems of noise reduction, insufficient feature extraction and low diagnosis precision under noise conditions. A Kaisixi University bearing public data set is used as original data, and Gaussian noise with different SNRs is superposed to simulate various noise intensities. And uniformly carrying out length alignment, down-sampling, equal-length segmentation, division and normalization preprocessing. Then, wavelet bases such as sym4, db4, coif5 and the like are adopted for parallel multi-scale decomposition and reconstruction; and adaptively determining the number of decomposition layers and a threshold strategy according to the noise level, and generating a de-noising branch. And performing weighted fusion on the denoising results of the branches, and performing iterative denoising on the residual error. Signals subjected to noise reduction processing are sent to a double-branch convolution-cycle-attention network, a convolution layer extracts features, an LSTM and a self-attention module capture time sequence changes, and accurate recognition of various bearing faults is achieved. The training adopts a segmented attenuation learning rate and an early stop strategy, and the robustness and generalization ability of different SNR working conditions are improved.

The invention discloses a self-adaptive threshold SAMP reconstruction method for a power quality disturbance signal. According to the method, a compression observation value is obtained by constructing a random Gaussian observation matrix, and sparse representation is carried out on an original signal by using discrete Fourier transform. In the iterative reconstruction process, the spectrum amplitude difference is introduced for the first time to serve as an adaptive termination basis, and automatic adaptation of different noise levels and different disturbance characteristics is achieved in combination with a dynamic threshold update function. According to the method, the problems of traditional SAMP sparseness overestimation and redundant iteration are effectively avoided, and the calculation load is remarkably reduced. Compared with an OMP method, an original SAMP method and the like, the method has the advantages that the number of iterations can be reduced by 30%-60%, the reconstruction signal-to-noise ratio is increased by 2-5 dB, the root-mean-square error is reduced by 10%-25%, higher robustness and real-time performance are achieved in power quality disturbance signal reconstruction, and the method is quite suitable for scenes such as compressed sampling, edge calculation and high-speed signal reconstruction in a power quality monitoring system.

The invention relates to the technical field of ultrasonic signal processing, in particular to a wavelet and MAD adaptive threshold combined laser ultrasonic signal denoising method. The method comprises the following steps: firstly, preprocessing an acquired trigger channel signal and an ultrasonic channel signal, determining a signal starting point through differential positioning, and intercepting an effective signal segment; carrying out multilayer wavelet decomposition on the effective signal segment to obtain a wavelet coefficient of each layer; extracting a detail coefficient of the highest decomposition layer, and adaptively estimating a noise standard deviation based on a median absolute deviation criterion; calculating an adaptive threshold according to the noise standard deviation and the signal length, and processing each layer of wavelet coefficient by adopting a hard threshold function; and finally, carrying out wavelet inverse transformation reconstruction to obtain a denoised signal. According to the method, prior noise information is not needed, the noise level can be adaptively estimated, the optimal threshold value can be determined, the signal features are reserved while noise is effectively suppressed, and the signal-to-noise ratio is remarkably improved.

The invention discloses a non-convex contiguous group sparse variational filtering hyper-parameter adaptive calculation method, which comprises the following steps of: selecting a reference signal according to ultrasonic signal characteristics, and determining the size of a group sparse group through a specific formula in combination with a sampling frequency and a probe frequency; after interference removal is carried out on the ultrasonic signal, a scattering noise standard deviation is calculated based on the reference signal, and a regularization parameter is obtained in combination with a formula; calculating a penalty factor; when the detection sound path distance is less than 1000mm, adopting global filtering; when the sound path distance is detected to be larger than 1000 mm, local filtering is adopted, a punishment intensity reference value is estimated based on the noise level, the reference value is corrected in combination with the signal-to-noise ratio, and then a formula is substituted to calculate a punishment factor coefficient and a final punishment factor; in the iteration process, the relative variable quantity of the cost function is used as a convergence criterion, iteration is stopped when the variable quantity is smaller than a preset threshold value, and efficient and stable filtering is ensured; the method completely meets the engineering application requirements of industrial ultrasonic detection.

The invention discloses a self-adaptive filtering method and system for an intelligent electric energy meter, and relates to the technical field of power grid signal processing, and the method comprises the steps: obtaining a power grid voltage and current signal set collected by a target intelligent electric energy meter historically, and obtaining a signal noise spectrum data set and a noise level evaluation result through spectrum analysis and noise level evaluation; if the result exceeds a preset noise threshold value, identifying an interference type and building a noise adaptive filtering multi-channel, integrating and deploying the noise adaptive filtering multi-channel into a target intelligent electric meter to complete signal filtering test and effect evaluation, and dynamically optimizing and generating an electric energy meter adaptive filtering multi-channel based on the obtained multi-channel filtering effect index parameter; and finally, signal filtering closed-loop control is realized. The technical problem that a traditional electric energy meter is weak in filtering pertinence and cannot adapt to dynamic noise of a power grid in electric power related scenes with complex dynamic interference is solved, and the technical effects that the electric variable measurement accuracy of the electric energy meter is improved, and the adaptive capacity of the electric energy meter to different interference scenes is enhanced are achieved.

The invention discloses an intelligent noise reduction method and system based on a mine earphone, and the method comprises the following steps: S1, obtaining an original noise signal in a mine operation environment, and carrying out the preprocessing of the original noise signal; s2, designing an adaptive compensation mechanism according to the reflection characteristics of the closed space of the mine, calculating a sound wave propagation attenuation coefficient and reflection delay, constructing a roadway reverberation model, and generating advanced compensation sound waves; and S3, carrying out collaborative fusion on the advanced compensation sound wave and a main noise reduction signal to generate a composite noise reduction signal. The invention relates to the technical field of intelligent noise reduction. According to the intelligent noise reduction method and system based on the mine earphone, the noise level in a mine operation environment is effectively reduced, the communication efficiency and the working comfort of operators are improved, potential damage of noise to hearing of the operators is reduced, and therefore the overall safety and the production efficiency of mine operation are improved.

The invention belongs to the field of man-machine interaction, and particularly discloses a man-machine interaction method for electronic screen control, which comprises the following steps: acquiring touch track data of a user according to a touch sensor, and performing feature extraction on the touch track data to obtain a touch feature vector; the method comprises the following steps: acquiring gesture action data of a user according to image acquisition equipment, and performing key point detection on the gesture action data to obtain a gesture feature vector; the method comprises the following steps: acquiring voice instruction data of a user according to an audio acquisition device, and performing acoustic feature extraction on the voice instruction data to obtain a voice feature vector; acquiring current environment parameters in real time through an environment detection module, wherein the environment parameters comprise illumination intensity, environment noise level and distance between a user and a screen; the objective of the invention is to solve the problem of insufficient reliability of a man-machine interaction mode in a complex environment in the prior art.

The invention discloses a composite board production quality management and control system, and relates to the technical field of composite board production. The multi-modal sensing fusion module is used for performing weighted fusion on detection data output by the geometric detection unit, the interlayer and defect detection unit and the process parameter detection unit, and dynamically adjusting weights of various data according to the production line speed, the environment temperature and the sensor noise level so as to improve the accuracy of defect identification and process abnormality judgment; and the core algorithm and control logic layer is used for analyzing and processing detection data in the production process of the composite board and outputting a control instruction. The multi-modal sensing fusion module is arranged to carry out centralized processing on data of various types of detection units, so that multi-source information is coordinated and cross-modal data fusion is carried out, the recognition precision of complex defects is improved, and the detection stability is ensured.

The invention discloses an acoustic excitation digital modulation noise reduction method and system for laser acoustic vibration detection and application. Comprising the following steps: step 1, acquiring single-point measurement time sequence parameters of laser vibration measurement equipment, step 2, measuring response characteristics of an audio excitation device, and determining pre-stabilization time required by an audio excitation signal from starting to reaching a stable amplitude; step 3, constructing an on-off keying modulation function, generating a digital gating function according to the time sequence parameter and the pre-stabilization time, and performing time domain multiplication on the audio excitation signal and the digital gating function to generate a modulated intermittent audio excitation signal; and 4, controlling the audio excitation device to play intermittent audio excitation signals, and synchronously triggering the laser vibration measurement equipment to scan and collect the measurement points on the surface of the object to be measured according to the time sequence parameters. Through the on-off keying digital modulation technology, audio excitation can be started only in the effective detection time period of laser vibration measurement, and therefore the average noise level and equipment power consumption are remarkably reduced.

An air-moving device may include an aerodynamic rotor in the form of a bladed disk. The aerodynamic rotor may include a physical geometry that constrains a noise level of the air-moving device.

Route planning for a hybrid electric vehicle (HEV) includes obtaining respective engine activation actions for at least some road segments of a route between an origin and a destination by optimizing for at least one of a noise level or energy consumption of an engine of the HEV that is used to charge a battery of the HEV. The HEV is then controlled to follow the at least some of the road segments of the route and to activate the engine according to the respective engine activation actions. Controlling the HEV to follow the at least some of the road segments includes masking at least one of the respective engine activation actions for a current road segment by increasing a volume of an entertainment system of the HEV.

Embodiments disclose a method and a system for robust transformation of input with a neural network. The method comprises processing the input data with a variational neural network (VNN) trained with ML to produce static parameters including noise level for the input data, injecting a set of random noises sampled on a probabilistic distribution according to the statistic parameters defined by the VNN to produce a set of perturbed input samples. The method comprises processing each of the set of perturbed input samples with a transformation neural network to produce a set of transformations and outputting a combination of the set of transformations as the robust transformation of the input data. Some embodiments consider training the variational neural network and transformation neural network by using adversarial examples from an attack model via alternating, explicit, and implicit gradient frameworks.

The invention discloses an Internet of Things networking scheduling method for railway tunnel disaster prevention illumination and evacuation indication, and relates to the technical field of railway tunnel disaster prevention. Comprising the following steps: S1, tunnel end monitoring equipment actively initiates environmental channel detection through a power line carrier communication network, scans impedance characteristics, noise levels and attenuation characteristics of different line sections in a tunnel, establishes a dynamic channel fingerprint database, and records an optimal communication frequency point and a channel quality threshold; and S2, performing real-time linkage between the local center monitoring main station and a fire alarm system, and after receiving a fire alarm signal, immediately generating a dynamic evacuation path through a disaster evolution prediction model according to a fire alarm position, a smoke diffusion state and a real-time channel state. The power line carrier communication module is utilized to actively perform channel detection and electrical parameter acquisition, a state channel fingerprint database is established, and fault diagnosis and state grading are performed based on the state channel fingerprint database, so that the overall reliability and maintainability of the scheme are improved, and the availability of equipment in a disaster is ensured.

Herein are described methods and systems for characterizing noise and noise level in pressure measurements in a downhole pressure testing operation in formation testing while drilling. The pressure measurements are acquired using a formation testing tool disposed on a drill string. The methods include drilling a borehole, extending the one or more probes into an inner surface of the borehole, performing a pressure testing operation with a first mud pump rate, wherein the pressure testing operation comprises a drawdown period and a build up period, decreasing the first mud pump rate to a second mud pump rate while measuring pressure during the build up period, decreasing the second mud pump rate to a third mud pump rate while measuring pressure during the build up period, and calculating a noise associated with pore pressure measurements during the first pressure testing operation.

The invention discloses a multi-level diffraction grating precision displacement sensor and a displacement measurement method, and relates to the field of displacement sensors. The grating precision displacement sensor comprises a laser, a half-wave plate, a first non-polarization splitting prism, a + / -2-level diffraction interference grating precision displacement sensor module and a + / -3-level diffraction interference grating precision displacement sensor module. And a final displacement measurement result is calculated after weighting by taking the noise levels of the + / -2-level diffraction interference grating precision displacement sensor module and the + / -3-level diffraction interference grating precision displacement sensor module as weights. The multi-level diffraction grating precision displacement sensor provided by the invention improves the utilization rate of effective light by using the polarization characteristic of light, reduces the interference of environmental stray light on measurement signals, and improves the accuracy and stability of the whole system.

A multi-stage deep learning method is provided, comprising training a deep learning network using a set of training pairs interchangeably including input data from a first noisy dataset with a first noise level and target data from a second noisy dataset with a second noise level, and input data from the second noisy dataset and target data from the first noisy dataset; generating an ultra -low noise data equivalent based on a low noise data fed into the trained deep learning network; and retraining the deep learning network on the set of training pairs using the target data of the set of training pairs in a first retraining step, and using the ultra-low noise data equivalent as target data in a second retraining step.

The invention discloses a self-adaptive intervention type single-array-element photoacoustic endoscopic imaging near-field noise removal method based on a spatial domain and morphological characteristics. The method comprises the following steps that 1, original data of photoacoustic endoscopic imaging are obtained; step 2, adaptively generating a patch mask area; 3, preprocessing the image in a frequency domain; step 4, executing strip interference detection based on adaptive spatial domain filtering on the non-plaque region to obtain a spatial domain noise mask; step 5, screening strip-shaped noise areas; step 6, performing interpolation restoration on the noise area; 7, calculating the convergence degree and the noise level in the iteration process; and step 8, outputting a final result after polar coordinate transformation. The intrusive single-array-element photoacoustic endoscopic imaging method can effectively inhibit strip-shaped interference signals occurring in intrusive single-array-element photoacoustic endoscopic imaging and reduce the covering influence of background noise on target signals, so that the imaging quality of a near-end area is improved, and the structural definition and contrast ratio of overall imaging are remarkably optimized.

The invention discloses an FISTA plug-and-play sparse radar imaging method based on adaptive regulation and control, which combines FISTA with deep learning prior, introduces reinforcement learning to carry out adaptive parameter regulation and control, and aims to significantly improve the automation degree, convergence rate and cross-scene robustness of an imaging process. The method is especially suitable for processing challenges of coexistence of different sampling rates, noise levels and scene complexity, and can effectively solve the problems that a traditional method depends on manual parameter adjustment, the calculation overhead is high, and the generalization ability is weak. According to the method, selection modeling of key hyper-parameters (such as step length, de-noising intensity and momentum coefficient) and a stop criterion is used as a Markov decision process, and an intelligent agent makes a decision adaptively according to a current reconstruction state, so that the problem is solved, and an advanced solution is provided for efficient and reliable radar imaging processing.

The invention relates to the technical field of air purification equipment, in particular to an unpowered public place air directional flow purification structure. The problems that in the prior art, the natural ventilation principle and the multi-stage purification function cannot be fused on the structural design level, and consequently inherent defects exist in the aspects of energy consumption, noise, maintenance cost, airflow organization stability and the like are solved. The unpowered public place air directional flow purification structure is arranged on a wall unit and comprises a micropore physical pre-filtering structure, a multi-layer labyrinth type physical filtering structure and an adsorption / deposition structure which are integrally arranged in a closed mounting cavity in the wall unit. In the operation process of the structure, no fan or motor or external power supply is needed in the whole process, air circulation is driven by completely depending on the thermal characteristics of the wall and weak thermal buoyancy formed by an indoor environment heat source, no mechanical moving part exists in the operation process, the noise level is low, and the noiseless requirement of a hospital ward area at night is met.

The invention relates to the technical field of computer vision and artificial intelligence, and discloses an industrial anomaly detection-oriented visual large model feature correction method, which comprises the following steps of: inputting a to-be-processed original input image into a pre-trained visual large model, and extracting an original feature vector; generating a plurality of noise adding variants with different noise adding degrees based on a target picture in a target data set of a certain industrial product, inputting the model to extract a feature vector, and calculating an average value to obtain a steady-state feature vector of the target picture; calculating a difference value between the steady-state feature vector and the original feature vector to obtain a deviation feature vector; selecting at least two representative pictures from the data set, extracting deviation feature vectors and calculating an average value to obtain a general feature correction vector; and in a model reasoning deployment stage, superposing the general feature correction vector to the original feature vector of the to-be-processed input picture to complete feature correction. According to the method, the accuracy and robustness of anomaly detection of the visual large model in a complex industrial environment can be effectively improved without fine tuning of the pre-trained large model.

The invention provides a segmented adaptive SG filtering method and device based on cosine weight transition. The method comprises the following steps: obtaining a noisy fault signal of a power distribution network; extracting local feature indexes of the signals, wherein the local feature indexes comprise an average gradient representing the change frequency of the signals and a composite feature index representing the noise level; dynamically determining a window length and a polynomial order of SG filtering based on the local feature indexes; the transient stage is the first 1 / 2 period of the post-fault signal, the steady-state stage is the subsequent part of the post-fault signal, an overlapping region is arranged at the junction of the transient stage and the steady-state stage, the overlapping region is a 1 / 2 period following the transient stage, and a cosine weight function is adopted to perform weighted fusion on filtering results of the two stages in the overlapping region; and splicing the filtering results of the non-overlapping region and the overlapping region of the transient stage and the non-overlapping region of the steady stage, and outputting a filtered fault feature signal.

The invention discloses a low-noise operation control method and system for a ship cooling pipeline, and relates to the technical field of low-noise operation control, and the method comprises the steps: carrying out the flow field feature extraction of historical detection data of the ship cooling pipeline, and combining the flow field feature data with pipeline structure data to carry out the flow field distribution analysis; performing sound field distribution modeling on the flow field distribution result to obtain a sound field distribution diagram, and combining the sound field distribution diagram with the equipment parameter operation state data to perform noise attribution analysis; and real-time state monitoring data of the ship cooling pipeline are obtained, coupling evaluation of noise and cooling efficiency is carried out in combination with a noise attribution analysis result, real-time operation equipment parameters are dynamically adjusted, and low-noise operation control over the ship cooling pipeline is achieved. According to the method, the limitation of a traditional ship cooling control mode is broken through, accurate noise suppression is achieved on the premise that the cooling performance is guaranteed, and meanwhile dynamic balance between the noise level and the cooling capacity is achieved.

The invention discloses a building machine hydraulic system fault diagnosis method fusing pressure-displacement double parameters, and relates to the technical field of building machine hydraulic system fault diagnosis, and the method specifically comprises the steps: 1, determining the physical position and bearing function of a machine position, configuring an independent sensor group for each machine position, determining a collection period, and carrying out the fault diagnosis of the building machine hydraulic system; the method comprises the following steps: step 1, collecting and preprocessing multi-dimensional parameters in real time, step 2, constructing a noise level evaluation model, dynamically adjusting a filtering kernel function, extracting equipment-working condition coupling features and multi-camera cooperative features in real time, and constructing a dynamic coupling feature vector for each camera, and step 3, training an attention mechanism model based on historical fault data, calculating the similarity between the standardized core features and a fault template library, outputting the fault confidence, judging a fault mode in combination with a threshold value determined by an ROC curve, extracting a trend index by a sliding time window, matching a rule library based on a mechanism and historical data, generating an auxiliary diagnosis conclusion, and performing cross validation with a model result.

The invention relates to the technical field of alkaline water electrolysis hydrogen production, and discloses a multi-parameter fusion gas-liquid separation effect online evaluation method, which comprises the following steps: acquiring alkali liquor mass fraction concentration, fluid temperature, actual mixing density and mass flow signals on a backflow alkali liquor pipeline of a gas-liquid separator in real time; dynamically calculating the theoretical density of the bubble-free sodium carbonate solution under the current working condition according to the mass fraction concentration of the alkali solution and the fluid temperature which are collected in real time; calculating the absolute deviation between the theoretical density and the actual mixing density; performing real-time noise level analysis on the mass flow signals; according to a comparison result of the absolute deviation and a preset density deviation threshold value; and outputting a separation result judgment result and executing corresponding alarm or interlocking control. The theoretical density of the bubble-free sodium carbonate liquid is dynamically calculated in the control unit in real time according to the online measured mass fraction concentration and temperature of the alkali liquid, and reliable separation result evaluation under the real-time concentration change working condition is achieved.

This invention discloses a method and system for fitting safety monitoring data based on adaptive hybrid basis functions, belonging to the field of computer-aided data analysis technology. The method achieves fully automatic high-precision curve fitting through a six-layer adaptive architecture: it automatically analyzes the trend complexity, periodicity, noise level, and nonlinearity of the data using multi-dimensional feature recognition technology; it calculates the data complexity score based on the feature analysis results and intelligently selects basis function combinations from an extended basis function library containing 32 sub-functions across 8 categories; it constructs a dynamic weighted hybrid basis function model, achieving adaptive adjustment of model parameters through time-varying weight functions and coupling correction terms; and it employs a multi-model dynamic fusion mechanism, integrating multiple candidate models based on six-dimensional confidence evaluation. This invention innovatively introduces adaptive regularization technology and a hierarchical optimization strategy, effectively balancing fitting accuracy and generalization ability, and possesses advantages such as full automation, strong robustness, and complete uncertainty quantification.

This invention discloses a method for generating averaged images based on dynamically weighted adaptive noise-aware frames, comprising: S10, input image sequence; S20, noise level estimation; S30, spatiotemporal consistency detection; S40, adaptive weight calculation; S50, Bayesian state update; and S60, output generation. This invention maintains a dynamic, multi-dimensional state estimate for each pixel location and, based on a Bayesian update rule, calculates an optimal fusion weight each time a new image frame is received, based on the new observation value of the pixel and its local context information, thereby updating the average estimate and confidence level of that pixel. This method abandons fixed historical frame windows or global motion models, instead employing a "soft decision" recursive update approach.

This application discloses a toilet noise control method, control device, and toilet, relating to the field of toilet control technology. The method includes the following steps: in response to a noise signal, calculating a noise level coefficient based on the current state data of the toilet; obtaining the noise level action amplitude of the toilet based on the noise level coefficient and the initial action amplitude of the toilet; and in response to a control signal, executing a toilet action corresponding to the control signal based on the noise level action amplitude of the toilet. The beneficial effects of this application are: by calculating the noise level coefficient based on the current state data of the toilet, the noise level changes with the toilet's usage state, lid-open state, etc., thereby executing actions of different amplitudes in different toilet states, resulting in a higher cleaning effect when the toilet is used for extended periods, achieving a balance between cleaning effectiveness and noise reduction.

This application provides an embodiment of a refrigerator, a refrigerator control method, and a storage medium, belonging to the field of household appliance technology. It aims to solve the technical problems of increased load and high noise levels generated by vacuum pumps in related technologies. The control module is configured to: control the vacuum pump to extract air from the storage space; obtain the current vacuum pump rate, which is determined based on the air pressure of the crisper drawer; if the current vacuum pump rate is lower than or equal to a preset rate, control the operating time of the vacuum pump as a reference time; if the current vacuum pump rate is higher than the preset rate, control the operating duration of the vacuum pump as an adjustment duration, which is less than the reference time. This refrigerator can reduce the noise generated by the vacuum pump.

The utility model relates to the technical field of intelligent fingerprint locks, in particular to a sliding cover assembly of an intelligent fingerprint lock, which comprises a sliding cover main body, the sliding cover further comprises a sliding rail, an elastic piece, a sliding cover accessory and an auxiliary structure, the sliding rail is fixed to the inner side of the sliding cover body, and the elastic piece is arranged on the inner side of the sliding cover body. Smooth and flat surfaces are guaranteed by the aid of the sliding rails and the guide strips which are manufactured by means of polishing, so that friction and noise between the sliding rails and the guide strips are reduced, abrasion of the sliding cover and the lock body can be effectively avoided by the aid of the first protection strip and the second protection strip, the service life of the fingerprint lock is prolonged, and the fingerprint lock is convenient to use. According to the lock, the sliding cover accessory is arranged, the auxiliary structures are arranged on the two sides of the sliding cover accessory, multi-stage buffering is formed through cooperation of the elastic protection pad and the multiple sets of springs, and therefore impact force and noise generated by collision are greatly reduced, the durability of the lock is further improved, the noise level is further reduced, and then the lock can provide quiet and convenient use experience for a user.

The utility model relates to the technical field of noise reduction plates, and discloses a shock absorption and noise reduction plate which comprises a shock absorption and noise reduction plate body, a shock absorption groove is formed in the outer surface of the front side of the shock absorption and noise reduction plate body, and a first damping rubber pad adheres to the inner surface of the rear side of the shock absorption groove. First rubber blocks are arranged at the positions, close to the four corners, of the outer surface of the front side of the first damping rubber pad, first clamping grooves are formed in the outer surfaces of the front sides of the four first rubber blocks, damping springs are connected to the interiors of the four first clamping grooves in a clamped mode, and a partition plate is fixedly connected to the position, close to the center, of the inner surface wall of each damping groove; the utility model solves the problem that the working efficiency of the equipment is reduced because the ideal noise reduction and sound attenuation effect cannot be achieved due to overhigh sound vibration frequency when the traditional sound attenuation plate is generally applied to places with higher noise, such as a music hall and the like.