149 results about "Threshold function" patented technology

The invention provides an industrial production Internet of Things data anomaly detection method, medium and system, and belongs to the technical field of industrial production Internet of Things. Industrial equipment sensor data is collected and preprocessed to establish a multi-dimensional data set, and principal component analysis and a mutual information algorithm are used to construct a dimension reduction feature data set; a virtual sensor algorithm is utilized to make up for data missing to form an extended data set, a simulation statistical mechanical anomaly analysis model is established based on a statistical mechanical law to convert the microscopic state of massive high-dimensional sensor data into a macro thermodynamic parameter, and a statistical mechanical feature vector is calculated through a virtual particle ensemble simulation equipment operation state. A state evaluation model and a dynamic threshold function are adopted to identify an abnormal mode and perform grading marking, a feedback optimization mechanism is constructed to continuously improve the system performance, and the technical problem that a traditional algorithm has a curse of dimensionality and cannot perform effective anomaly detection due to extremely high data dimensionality of an industrial equipment sensor is solved.

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 relates to a transformer fault identification method based on voiceprint signals, and belongs to the technical field of cepstrum for extracting parameters in audio decoding or coding. The method comprises the following steps: setting a fault type and establishing a fault identification model for training; arranging an acoustic sensor to collect voiceprint signals of the transformer; utilizing a dream optimization algorithm to optimize the penalty factor and a successive variational mode decomposition method to decompose a plurality of mode components, and dividing the mode components into pure components and noisy components; noise reduction is carried out on the noisy component by adopting a designed threshold function in combination with wavelet threshold noise reduction, and the noisy component and the pure component after noise reduction are input into the recognition model to obtain a probability vector; and finally, fusing into a first fusion probability vector and a second fusion probability vector through fuzzy measurement, and taking the fault type corresponding to the maximum second fusion probability as the fault type of the transformer. The method can accurately capture the mapping relation between the acoustic features and the transformer fault state, and accurately identifies the transformer fault type.

The invention provides a heterogeneous multi-agent system control method under a DoS attack based on an event triggering strategy. Robust cooperative control under a complex attack environment is realized through the following steps: S1, performing state description and topological structure modeling on a heterogeneous multi-agent system; s2, carrying out approximation on an unknown nonlinear function in the system; s3, restraining the error dynamic state through a performance function, and ensuring that the system response meets the preset precision; s4, designing a time-varying threshold function, and dynamically adjusting a trigger condition to reduce communication burden; s5, designing an anti-attack controller in combination with event triggering and a Lyapunov theory; s6, designing an adaptive law to carry out online estimation on disturbance; and S7, testing synchronization precision, communication efficiency and anti-attack performance based on a physical platform. According to the method, deep fusion of the event triggering mechanism and the preset performance constraint is realized for the first time, and the self-adaptive dynamic threshold strategy is introduced, so that malicious interference of DoS attacks on a communication link is effectively dealt with, and it is ensured that the heterogeneous nodes can still meet preset performance indexes in an attack scene.

The invention discloses a data processing method based on an ultrahigh frequency original signal and a storage medium, and relates to the field of data processing, and the method comprises the steps: carrying out the adaptive gain adjustment collection of the ultrahigh frequency original signal, and enabling the signal amplitude to be stabilized in a preset effective interval through the dynamic adjustment of a gain parameter, acquiring initial acquisition data containing complete discharge characteristics; performing time-frequency domain combined noise reduction preprocessing on the initially acquired data, constructing an adaptive threshold function based on signal frequency domain distribution characteristics, and filtering background noise and non-discharge interference signal components; according to the method, the signal amplitude is dynamically stabilized through adaptive gain adjustment, it is ensured that initial collection data completely contains discharge characteristics, the problem of information loss caused by signal fluctuation is solved, the adaptive threshold is constructed based on frequency domain energy characteristics through time-frequency domain combined noise reduction, background noise and interference components are accurately filtered out, and the signal purity is improved.

The method comprises the following steps: obtaining a power quality signal containing noise; selecting permutation entropy as an adaptive function of genetic algorithm, calling variational mode decomposition through genetic algorithm, and iteratively optimizing a penalty factor α and a decomposition mode number k of the variational mode decomposition to determine optimal parameters; decomposing signal data into k mode components through the variational mode decomposition, and determining effective mode components and noise mode components through a correlation coefficient; for improved wavelet threshold, a parameter-adjustable threshold function is proposed, and the concept of wavelet energy entropy is introduced into the threshold function; the noise mode components are denoised through the improved wavelet threshold, and the effective mode components and the denoised noise mode components are reconstructed to obtain a denoised power quality disturbance signal. The method can effectively remove noise interference while retaining singular information of mutation points of the collected signal, and provides help for subsequent analysis and treatment of the power quality disturbance signal.

The invention relates to the field of deep learning, and provides a container ship loading and unloading operation time prediction method and system for solving the problems that an existing container ship operation time estimation method is rough, and operation data has noise in a real loading and unloading operation scene. On the basis of a deep residual shrinkage network DRSN-CW (Deep Residual Shrinkage Net) model with feature extraction and denoising capabilities under a ResNet-18 architecture, a fixed soft threshold function is changed into a more flexible adaptive threshold function, so that the noise reduction capability is more flexible while the feature learning capability is kept. According to the scheme of the application, the request of real-time prediction of the quay crane operation duration in the berth planning process can be well met with relatively high average detection precision and robustness.

The invention discloses a power distribution network reliability intelligent analysis method and analysis system, particularly relates to the technical field of power system fault detection, and is used for solving the problems of misjudgment and missed judgment caused by fault feature weakening and harmonic interference in a bidirectional power flow scene in an existing method. The method comprises the following steps: firstly, synchronously collecting fundamental wave and harmonic components of nodes of a power distribution network, extracting a current amplitude change rate and a phase offset by adopting time-frequency domain conjoint analysis, and generating dynamic distribution characteristics by combining harmonic energy clustering; identifying high-frequency interference components in a bidirectional power flow mode by quantizing a phase deviation direction and harmonic energy aggregation characteristics; based on a harmonic frequency band coupling effect and a current dynamic association relationship, an adaptive detection threshold function is constructed, and accurate separation of fault features is realized; and finally, combining a real-time fault judgment result with historical repair data to generate a reliability evaluation parameter. And the sensitivity of fault detection and the rationality of reliability evaluation in a complex operation environment are remarkably improved.

The invention discloses a seismic signal denoising method based on EMD combined improved wavelet transform, and the method comprises the following steps: reading an original seismic signal, and carrying out the EMD decomposition, and obtaining a plurality of IMF components; calculating a kurtosis value of the IMF so as to distinguish the IMF containing noise and the IMF containing effective signals; discrete wavelet transform is carried out on the noisy IMF, and a detail coefficient and an approximation coefficient are calculated; selecting a proper wavelet basis and a proper number of decomposition layers, and processing a wavelet coefficient based on an improved threshold function to reduce noise interference; and recovering the de-noised IMF by using a reconstruction formula, and combining the de-noised IMF with the unprocessed low-frequency IMF to obtain a final de-noised signal. According to the method, the threshold function of wavelet transformation is improved, the calculation efficiency is improved, the signal loss is reduced, the signal-to-noise ratio is remarkably improved, the denoising stability is ensured, and the method has important significance on seismic data processing in a complex noise environment.

The invention discloses a wavelet noise reduction multi-path extraction method for improving a threshold function, which can be used for effectively extracting and weakening multi-path errors of a global navigation satellite system (GNSS), solves the problem that signals of wavelet coefficients are discontinuous or distorted at a preset threshold, improves the traditional wavelet noise reduction effect and multi-path error modeling precision, and improves the accuracy of multi-path error modeling. Therefore, the influence of multi-path errors on GNSS positioning and orbit determination precision is effectively weakened, a feasible solution is provided for realizing high-precision positioning and orbit determination in a complex environment, and a solid foundation is laid for establishing 1mm space-time reference in the future.

The invention belongs to the field of computer graphics and computer vision, and provides an explicit volume rendering primitive densification method based on detail perception gradient, which is suitable for a three-dimensional reconstruction task under a sparse view angle or sparse point cloud condition. According to the method, a detail perception gradient index is introduced, the fitting capability of a current model to a fuzzy region in an image is dynamically evaluated, and whether point refinement operation is executed or not is judged according to the fitting capability. Compared with a traditional densification strategy depending on position gradient, the index provided by the invention can more accurately identify a detail missing region, and the rationality of density distribution is improved. Furthermore, according to the method, visual angle correlation and gradient normalization information are combined, a threshold function is set to split and reconstruct Gaussian primitives, and low-overhead and controllably-distributed point set enhancement is achieved. While the compactness of the whole model is kept, the reconstruction quality of a complex structure and an edge region is effectively improved, and the method can be widely applied to various application scenes such as real-time rendering, underwater perception and sparse reconstruction.

The invention relates to the technical field of multi-agent control, and discloses a multi-agent dynamic event triggering preset time bipartite tracking control method and system, and the method comprises the steps: obtaining a position state of an agent in real time through bipartite tracking control, and constructing a symbol communication topology with a balanced structure and a leader following multi-agent system model; determining a bipartite tracking control target in preset time; then, two bounded time-varying scale functions are designed; a dynamic event triggering mechanism is introduced, a measurement error, an event triggering function and a dynamic event triggering threshold function are constructed, dynamic event triggering conditions are defined, and it is ensured that updating operation is conducted only when control input meets the dynamic event triggering conditions. Based on a bipartite tracking framework, the method meets the requirements of grouping confrontation and task difference in multi-group collaboration, enhances the flexibility and applicability of the system, effectively reduces the updating frequency and communication burden while ensuring the system performance, and is suitable for a distributed control scene of a large-scale complex multi-agent system.

The invention discloses an abnormal sample detection and identification method and system based on evolutionary computation and multi-modal consistency constraint, and belongs to the field of network security and artificial intelligence security. The method comprises the following steps: S1, acquiring and preprocessing input data; s2, multi-modal feature extraction and unified characterization are carried out; s3, optimization feature selection and weight self-adaption are carried out; s4, constructing a cross-modal fusion detection model; s5, abnormal sample risk identification; and S6, result judgment output: generating a final detection label according to a threshold function. The method shows high accuracy and strong robustness in counterfeit image and video detection and abnormal sample elimination, and can be widely applied to the fields of multimedia authentic identification, intelligent security and protection and AI content traceability.

The invention discloses a self-adaptive noise reduction method for a vibration signal of a filling pipeline in the technical field of signal denoising. The self-adaptive noise reduction method comprises the following steps: decomposing the vibration signal of an original pipeline by adopting an optimized VMD to obtain a plurality of IMF components; identifying a signal dominant component and a noise dominant component in the IMF component; noise reduction is carried out on the noise dominant component by adopting an improved wavelet threshold function; and reconstructing the noise dominant component and the signal dominant component after noise reduction to obtain a denoised filling pipeline vibration signal. According to the method, the optimal modal number K and the penalty factor alpha of the VMD are adaptively determined by adopting an NSGA-II multi-target optimization algorithm, the subjectivity of parameter adjustment by traditional artificial experience is overcome, and meanwhile, the problems that a traditional single-target optimization method is prone to falling into local optimum, weak in global search capability and one-sided in optimization are solved; and meanwhile, a wavelet threshold function is cooperatively improved, so that medium-high-frequency noise can be eliminated, low-frequency interference can be suppressed, and the signal quality and the feature identification degree are greatly improved.

The present application relates to the technical field of transformer fault diagnosis, and in particular to a transformer body vibration signal analysis and fault diagnosis method, device and medium. The present application combines the kurtosis characteristics of the vibration signal, adopts a semi-soft threshold function wavelet denoising method based on the threshold selection method of the 3sigm rule, and achieves better denoising effect. Through the fault diagnosis model of the organic fusion of the two improved HHT transforms-mobilenetV2 model, combined with different feature extraction methods, it is more conducive to retaining the effective features of the vibration signal; the improved mobilenetV2 model designs a multi-scale deep convolution model, introduces a channel attention mechanism before the channel-by-channel convolution, and after multi-scale deep convolution feature extraction, a multi-source data attention mechanism is introduced; without affecting the safe and reliable operation of the transformer, the intelligent diagnosis of the vibration state fault of the transformer is realized.

The invention discloses a design method for ISTA sparse channel estimation based on a single-transmitting and single-receiving OFDM (Orthogonal Frequency Division Multiplexing) system. According to the method, the signal can be recovered under the condition of unknown sparseness, accurate reconstruction of the signal is realized, and a sparse channel is reconstructed through a small amount of training data, so that the training overhead is reduced, the spectrum efficiency is improved, and signal reconstruction is carried out by utilizing gradient updating and a soft threshold threshold function. A simulation experiment shows that the algorithm can realize better channel reconstruction, and meanwhile, the algorithm also has higher convergence speed and lower calculation complexity.

The invention provides a self-adaptive denoising method based on Meyer wavelets and an improved smooth threshold function in order to solve the problem that a transformer partial discharge high-frequency current sensor field calibration signal is seriously polluted by field electromagnetic noise interference. The method comprises the following steps: firstly, arranging a standard sensor according to an optimal cross-core distance to obtain an original noisy signal; then adaptively determining a decomposition series according to a data length, selecting Meyer wavelets to perform multi-scale decomposition, performing smooth contraction on wavelet coefficients by adopting a continuously derivable, progressive and unbiased improved threshold function, and finally reconstructing a high signal-to-noise ratio verification signal through wavelet inverse transformation; the method does not need manual intervention of the threshold value, keeps orthogonality and high-order derivability in the whole process, and is beneficial to improving the accuracy of the verification result of the high-frequency sensor to be tested.

The invention discloses a lightweight non-repudiation model fingerprint efficient traceability method, and belongs to the related technical fields of cryptography application, digital content traceability and the like. The method comprises the following specific steps: inputting an original image data set into a computer system; training and learning a lightweight encoder and a decoder, wherein the encoder maps a binary fingerprint and an image to generate a residual error; the residual error is constrained through a Sigmoid function to generate a fingerprint-containing image, and the model is optimized through dual loss; generating metadata of a PIT parameter binding salt value and a commitment value, and embedding the fingerprint into the image; during detection, a decoder is loaded to extract a decoded fingerprint, binaryzation is carried out through a threshold function, and dual verification is realized through fingerprint polynomial comparison and commitment value verification. According to the method, a fingerprint verification mode is converted into polynomial verification through a polynomial equivalent detection technology, the communication complexity is reduced to sub-linear overhead, lightweight efficient verification is realized, and the method is suitable for efficient traceability and authenticity verification of deep forged images.

The invention discloses a power servo knife rest health state assessment method based on an integrated multi-scale convolution attention network, and belongs to the technical field of power servo knife rest health state assessment, the method adopts a base classifier architecture of the multi-scale convolution attention network, and the base classifier architecture is composed of a multi-scale convolution noise reduction module and an attention enhancement module; the multi-scale convolution noise reduction module extracts rich discrimination features through a multi-scale kernel; the noise is further filtered through a soft threshold function, the attention enhancement module adopts a sparse improved channel attention mechanism to adaptively filter redundant features, important channel information is concerned, and the anti-noise performance and generalization of the model are improved; an interactive joint training strategy is adopted, a basic classifier automatically pays attention to a category with poorer classification through a designed Recall guide loss function, a balance training subset is constructed for each basic classifier, weighted voting is adopted to integrate the trained basic classifiers, and the robustness of an integrated model and the evaluation performance under data imbalance are improved.

The application discloses a water acoustic signal time-frequency spectrum line spectrum extraction method and system, and belongs to the technical field of signal processing. A noisy time domain signal is generated through simulation, and a mask label of a time-frequency spectrum of the noisy time domain signal belonging to a line spectrum is generated through a soft threshold function; a denoising model is trained according to the time-frequency spectrum and the mask label; a target water acoustic signal is acquired, the time-frequency spectrum of the target water acoustic signal is input into the denoising model, and a mask label corresponding to the time-frequency spectrum input is output through inference; the time-frequency spectrum of the denoised target water acoustic signal is acquired according to the mask label and the time-frequency spectrum; an initial candidate point set of the time-frequency spectrum of the denoised target water acoustic signal is acquired, and an initial candidate point of a current frame time-frequency spectrum in the initial candidate point set is acquired; a correlation cost matrix is constructed, the initial candidate point and a trajectory are correlated and matched with the minimum difference as a target, and a line spectrum of the trajectory and the candidate point dynamic correlation is acquired. The method can balance denoising fidelity, detection accuracy and real-time performance.

The invention discloses an abnormal station identification method and system based on an observation value and a distance dynamic threshold value, and relates to the technical field of observation station data quality control. The method comprises the following steps: acquiring long-sequence hourly precipitation data of observation stations and position data of the observation stations which are verified to be qualified; constructing a relation function between the distance close to the observation station and the hourly precipitation difference threshold value, and a relation function between the hourly precipitation of the observation station and the hourly precipitation difference threshold value close to the observation station; on the basis, calculating the data exception probability of all observation stations in the to-be-analyzed time period; and judging the observation station level according to the data anomaly probability. According to the method, the dual-threshold function is dynamically constructed by fusing the size of the observation value and the station spacing, the limitation of a traditional fixed threshold is broken through, accurate and adaptive identification of abnormal data of the observation station can be realized, and the reliability and the service application value of ground observation data are improved.

The application discloses an abnormal station identification method and system based on observation values and distance dynamic threshold values, and relates to the technical field of observation station data quality control. The method comprises the following steps: acquiring verified qualified long sequence of hourly precipitation data of observation stations and observation station position data; constructing a relationship function of the distance of adjacent observation stations and the difference threshold value of hourly precipitation, and a relationship function of the hourly precipitation of observation stations and the difference threshold value of hourly precipitation of adjacent observation stations; on the basis, calculating the data anomaly probability of all observation stations in the analysis period; and judging the observation station grade according to the data anomaly probability. The application breaks through the limitation of traditional fixed threshold value by fusing the observation value size and the station distance to dynamically construct a double threshold function, can realize accurate and adaptive identification of abnormal data of observation stations, and improves the reliability and business application value of ground observation data.

The invention discloses an aluminum plate ultrasonic Lamb wave denoising method based on ICPO-VMD combined improvement of a wavelet threshold, and the method comprises the steps: collecting an ultrasonic echo signal of an aluminum plate, carrying out the adaptive optimization through an improved crown porcupine optimization algorithm, and obtaining an optimal parameter combination; performing variational mode decomposition on the ultrasonic echo signal by using the optimal parameter combination to obtain a plurality of intrinsic mode function components; calculating a correlation coefficient of each intrinsic mode function component and the original ultrasonic echo signal, and dividing all the components into a signal dominant component, a mixed component and a noise dominant component according to a preset correlation coefficient threshold; de-noising processing is carried out on the mixed component by adopting an improved wavelet threshold function; and reconstructing the signal dominant component and the denoised mixed component to obtain a denoised Lamb wave signal. According to the invention, high-fidelity and high-reliability de-noising processing of the ultrasonic Lamb wave signal of the aluminum plate is realized, and the accuracy and robustness of subsequent defect detection are improved.

The invention provides a karst tunnel excavation support pressure reliability design method based on threshold gradient. According to the method, aiming at three random variables in an HB constitutive model, firstly, center composite sampling is carried out in a standard normal space to obtain seven test points, coordinates of the test points are converted into an original parameter space, a failure threshold value of each test point is obtained through calculation by establishing a numerical model, then a secondary response surface agent model without cross terms is established, and the failure threshold value of each test point is calculated. And establishing a tunnel face support force dominant threshold function by solving a linear equation set and determining seven coefficients of the proxy model, on this basis, proposing a standard normal space target reliability isoline checking point iteration method, performing checking point iteration on a sphere with the radius being the target reliability, and updating checking points in a gradient manner by adopting the threshold function. And calculating the collapse load of the design point during iterative convergence, so that the collapse load meets the requirement of a target reliable index. According to the method, the design complexity is reduced, and the design efficiency is improved.

The invention relates to an image denoising method and system based on a wavelet transform algorithm, and belongs to the technical field of image processing, in each wavelet transform layer, when a first wavelet coefficient in a current layer is smaller than a wavelet threshold of the current layer, a threshold function uses a wavelet hard threshold function to carry out threshold processing on the first wavelet coefficient; when the second wavelet coefficient in the current layer is larger than the wavelet threshold value of the current layer, the threshold value function uses a preset improved threshold value function to carry out threshold value processing on the second wavelet coefficient, and a wavelet coefficient representing a non-noise signal in each wavelet transformation layer is obtained; according to the method, wavelet inverse transformation is carried out according to wavelet coefficients of non-noise signals in all wavelet transformation layers, a denoised image is obtained, the quality of the image is improved through an improved threshold function, the image is more suitable for further image analysis or computer vision tasks, the whole processing process is high in automation degree, and the processing efficiency is improved. And a satisfactory enhancement effect can be obtained without manual intervention.

The present disclosure provides a data reconstruction method based on Radon domain sparse representation and related equipment, relating to the technical field of communication. The method comprises: obtaining low-resolution Radon coefficients; inputting the low-resolution Radon coefficients into a neural network of an encoding-decoding structure to obtain first high-resolution Radon coefficients; and increasing the sparsity of the first high-resolution Radon coefficients through an adaptive soft threshold function cascaded at the back end of the neural network to obtain second high-resolution Radon coefficients. The method can reduce the data size, operation time, and operation cost of time-domain inversion of reduced-time variable Radon transform, and improve the stability and resolution of reconstructed data.

The application discloses a current signal denoising method and system for fault arc detection and a storage medium, comprising: collecting the working current of the load at the current time, and analyzing to obtain the noise estimation spectrum of the working current at the current time; wavelet decomposing the noise estimation spectrum and the working current of the load at the subsequent time respectively to obtain first wavelet coefficients and second wavelet coefficients; correcting the second wavelet coefficients based on the first wavelet coefficients to obtain third wavelet coefficients; and wavelet reconstructing the third wavelet coefficients to obtain a time domain enhanced signal. The improved wavelet threshold function improves the defects of poor continuity and constant deviation of the traditional wavelet threshold function, and the current signal processed by the denoising algorithm not only suppresses the existence of noise, but also improves the arc characteristics of the signal, so that the detection performance of the fault arc detection algorithm is significantly improved.

The application provides a video decoding method and device, a video encoding method and device, and relates to the technical field of video coding and decoding. The method comprises the following steps: decoding the encoding data of a current frame, obtaining the hyper-prior residual mean and hyper-prior residual variance of each pixel point, and residual encoding data; determining a residual feature skip threshold value, wherein the residual feature skip threshold value of the current frame is the product of a first coefficient and the value of a residual skip threshold function when the time sequence of the current frame meets a preset condition, and the residual feature skip threshold value of the current frame is the product of a second coefficient and the value of the residual skip threshold function when the preset condition is not met, and the first coefficient is smaller than the second coefficient; obtaining the reconstructed residual feature of each pixel point of the current frame; and reconstructing the current frame according to the reconstructed residual feature of each pixel point of the current frame to obtain a reconstructed frame of the current frame. The embodiment of the application is used for preventing the cumulative propagation of error information, thereby improving the encoding efficiency.

A technique to prune weights of a neural network using an analytic threshold function h(w) provides a neural network having weights that have been optimally pruned. The neural network includes a plurality of layers in which each layer includes a set of weights w associated with the layer that enhance a speed performance of the neural network, an accuracy of the neural network, or a combination thereof. Each set of weights is based on a cost function C that has been minimized by back-propagating an output of the neural network in response to input training data. The cost function C is also minimized based on a derivative of the cost function C with respect to a first parameter of the analytic threshold function h(w) and on a derivative of the cost function C with respect to a second parameter of the analytic threshold function h(w).

This invention discloses a method for online prediction of equipment faults with a small sample size, including the following prediction steps: S1, Data Processing: By optimizing the threshold function in the WTD denoising algorithm and introducing BIC to evaluate the impact of the number of decomposition layers on the complexity of WTD, an improved WTD algorithm is proposed for online filtering of noise in fault signals. This invention proposes an online prediction model for equipment faults with a small sample size, and verifies the effectiveness and reliability of the model using rolling bearing life cycle vibration data. BIC can accurately find the optimal number of decomposition layers in the WTD algorithm, providing a basis for improving the parameter settings of the WTD model. The improved WTD algorithm has excellent denoising effect, ensuring the reliability of fault data. The improved MEST algorithm and dual CSFI algorithm can effectively convert fault signals into fault degree indicators, providing high-quality data support for subsequent fault prediction.