41 results about "Akaike information criterion" patented technology

The invention discloses a cascade water-wind-light multi-power system modeling method based on time-space correlation, and the method comprises the steps: building an autoregression integral moving average-generalized autoregression condition heterovariance model of wind power photovoltaic power and cascade hydropower station runoff, building an R-vine, C-vine and D-vine Copula function spatial correlation model through employing a generalized autoregression condition heterovariance value, and carrying out the modeling of a cascade water-wind-light multi-power system. And selecting an optimal spatial correlation model according to a Bayesian information criterion, an akaike information criterion and a logarithm likelihood value goodness of fit index. Based on the optimal spatial correlation model, constructing a scene generation method of Latin hypercube-Copula function sampling, performing scene generation on the wind power photovoltaic power and the runoff volume of the cascade hydropower station, and constructing a scene reduction method of contour coefficient optimization Kmeans clustering to reduce the generated scene. The invention provides a time-space correlation cascade water-wind-light multi-power system modeling method, and aims to analyze water-wind-light multi-energy complementary operation characteristics and provide reference for a scheduling mode among water-wind-light multi-energy.

The invention belongs to the technical field of ecological restoration, and provides a mountainous river-oriented fish diversified habitat natural shaping restoration method, which comprises the following steps of: determining fishes with a relative important index greater than 5% in a river as target fishes; the habitat factors of the river are screened, the habitat factors with colinearity are divided into a factor group, and only one habitat factor is reserved in each factor group to serve as a representative habitat factor; carrying out modeling on the resource distribution and representative habitat factors of the target fishes by adopting a generalized additive model; screening to obtain a generalized additive model with a minimum akaike information criterion, and determining an optimal habitat factor combination; and determining a restoration scheme according to the habitat factor combination, and adjusting the habitat of the target fish. According to the restoration method, the natural process is simulated to remodel the diversity of the mountain river habitat, the key problems of hydrological connectivity fracture, habitat simplification, engineering and ecological contradiction and the like are solved, and fish population restoration and ecological system self-maintenance are achieved.

The application discloses a site micro-vibration vibration source identification method based on machine learning, signals measured by acceleration sensors at the edge of a measured site and acceleration sensors at the center of the measured site are processed in sequence, including denoising, transient impact signal extraction, feature extraction, clustering processing of dynamic time warping distance of the transient impact signal based on a feature matrix, GMM modeling, obtaining a model, verifying the number of independent Gaussian components of the model, selecting the number of components with the minimum value as the optimal model according to the Akaike information criterion value, and performing maximum a posteriori probability identification on the transient impact signals captured by the sensors; GMM parameter comparison is performed on sets that may belong to the same category, two sets with highly overlapped models are merged into the same set, and the set is used as vibration data sets generated around the site and having an impact on the center position of the site.

The invention discloses a distribution network line fault trip prediction method and device, a medium and a product, and relates to the field of power distribution network operation and maintenance, and the method comprises the steps: obtaining fault data of a distribution network line; constructing a fault data set; according to the fault data set, constructing a trip prediction model based on an akaike information criterion; an exogenous variable corresponding to the tripping prediction model is determined; constructing a limit gradient lifting model based on a grid search method according to the exogenous variables and the historical fault trip times; constructing a fusion model according to the trip prediction model and the limit gradient lifting model; respectively determining prediction errors of the tripping prediction model and the fusion model, and determining a final prediction model according to the prediction errors; determining the number of fault tripping times of the month to be predicted by using the final prediction model; and regularly obtaining fault data of the distribution network line to regularly update the final prediction model. The monthly fault trip times of the distribution network line can be efficiently predicted, and the operation and maintenance reliability of the distribution network is improved.

The invention discloses an unbalanced power distribution network risk analysis method based on a multistage sparse polynomial, and belongs to the technical field of power distribution network risk analysis. The method comprises the steps of establishing a power distribution network input-output model; approximating the input and output model of the power distribution network as a linear regression model; calculating an ANCOVA index and a P value of the linear regression model, and sparsifying the input and output model of the power distribution network to obtain a preliminary sparsifying input and output model of the power distribution network; restraining a high-order cross term of the preliminary rarefaction power distribution network input and output model according to a preset truncation rule, shrinking a coefficient of the preliminary rarefaction power distribution network input and output model according to the optimized polynomial chaos expansion coefficient, and taking a preset residual error akaike information criterion as a model balance criterion to obtain a final rarefaction power distribution network input and output model; and obtaining a power distribution network risk analysis result according to the final sparse power distribution network input and output model. The method can provide a quantitative basis for operation and control of the power distribution network.

The invention discloses a partial discharge positioning detection method and system based on acoustoelectric joint detection and multi-dimensional feature fusion, and belongs to the technical field of intelligent power grid state detection. According to the method, an ultra-high frequency (UHF) signal is used as a discharge absolute time reference, an akaike information criterion (AIC) is adopted to adaptively pick up an ultrasonic weak wave head, and the problem of failure of a traditional threshold value method under a low signal-to-noise ratio is solved; meanwhile, a four-dimensional state vector containing space coordinates and equivalent sound velocity is constructed, a weighted Gaussian-Newton iterative algorithm is used for solving, a temperature field sound velocity correction model and a grid search initial value estimation algorithm are introduced, and the problems of iterative divergence and local optimal solution are solved. In addition, the method integrates algorithm modules of intelligent noise reduction, physical field adaptive correction, sound intensity assisted positioning and the like, and realizes high-precision and high-robustness positioning of the partial discharge source.

This invention relates to a method for highlighting spike discharge data of intracranial electroencephalogram (EEG) signals. The method involves collecting background noise data from the patient's brain without neuronal discharges, preprocessing the background noise data, automatically selecting the optimal order of an autoregressive (AR) model using the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC), estimating the AR model coefficients using the Yul-Walker equation, and constructing a background noise model. The intracranial EEG signals to be processed are then subjected to high-pass filtering. A short-time Fourier transform (STFT) and window-based frame-by-frame processing strategy are used to subtract the spectrum of the signal from the noise. By adjusting the parameters of the spectral subtraction, noise removal and preservation of neuronal signal features are achieved.

The invention relates to the technical field of global navigation satellite system data processing, and discloses a GARCH-based GNSS cycle slip detection threshold real-time determination method, which comprises the following steps: acquiring GNSS double-frequency observation data, constructing a Morse-Wiberner and geometric distance-free combination observation value, and carrying out differential processing to obtain a noise time sequence. Secondly, traversal calculation is carried out based on an akaike information criterion, and the optimal order and the sliding window length of the GARCH model are determined; then, a current epoch condition standard deviation is predicted using the model, and a multiple factor is adjusted in combination with an environmental state to construct a dynamic detection threshold. And finally, comparing an absolute value of the noise sequence with a threshold value to judge a cycle slip, and calculating an integer solution by constructing an observation equation to complete repair. According to the method, the heterovariance characteristic of the observation noise is modeled by using the GARCH model, so that the adaptive adjustment of the detection threshold along with the environmental noise level is realized, the misjudgment rate in a complex environment is effectively reduced, and the detection accuracy is improved.

This invention discloses a multi-information source mechanism calculation method based on earthquake early warning station network observations. The method includes selecting velocity and acceleration recorded waveforms to form an input database; obtaining the P-wave initial motion polarity through manual picking and template matching techniques; calculating the source mechanism using a grid scanning method, and using solutions with an inconsistency ratio below a threshold as preliminary solutions; manually screening P-wave and S-wave phase waveforms; and jointly using initial motion polarity, S / P amplitude ratio, and body wave waveform as constraints to calculate the probability value of each preliminary solution. The probability distribution of clustered solutions is obtained through the Akaike information criterion and cluster analysis to obtain the optimal source mechanism calculation result. This invention, based on joint inversion calculations using acceleration records from early warning stations and velocity records from seismic stations, lowers the lower limit of the magnitude of earthquakes with computable source mechanisms. The comprehensive use of three types of information constraints improves the reliability of the results, and the calculation of the probability distribution of clustered solutions reduces the adverse effects of multiple solutions, thus improving the stability of the results.

The application discloses a water engineering scheduling runoff intelligent simulation method, comprising: edge distribution function fitting: using the matrix method to calculate the mean, dispersion coefficient and skewness coefficient, and adopting a P-III type curve to construct a daily runoff frequency curve; wisdom knowledge set construction: using a time series model to establish wisdom knowledge sets of each parameter in a mixed Copula set, using a cooperative search algorithm, taking the minimum of the Akaike information criterion of the theoretical distribution and the empirical distribution as the target, and calibrating the wisdom knowledge set parameters; daily runoff simulation: based on the Bayesian rule, a mixed conditional Copula set is established, and Latin hypercube sampling is used to generate a long sequence of daily runoff. The method reduces the difficulty of daily runoff simulation by the mixed Copula, reduces the number of parameters, improves the fitting precision of the mixed Copula, speeds up the construction speed of the mixed Copula set, and provides a valuable technical tool for basin management.

The present application belongs to the field of blast furnace system health management and prediction, and specifically discloses a blast furnace degradation model calibration and residual life prediction method under a noise environment. The method is based on a fractional Brownian motion with historical dependence, constructs an initial degradation model of the blast furnace system, estimates the model parameters by the maximum likelihood estimation method, and determines the optimal model structure of the blast furnace system temperature change by the Akaike information criterion. In addition, the present application also designs a model calibration trigger mechanism; when the mechanism is not triggered, the Bayesian fusion particle filter is used to estimate the potential degradation state of the system and update the model parameters; when the mechanism is triggered, a prediction error model is constructed and the degradation model structure is calibrated to adapt to complex working conditions. Finally, the probability distribution function of the residual life of the blast furnace system is derived, and the residual life prediction of the blast furnace system is carried out according to the calibrated degradation model.

The invention discloses an eVTOL collaborative design optimization method based on improved genetic programming, and the method comprises the steps: firstly constructing pneumatic, propulsion and dynamics high-fidelity models of an eVTOL takeoff stage, building an original collaborative optimization problem, and solving the problem through a direct transcription method; secondly, multi-target genetic programming fusing a root mean square error, an akaike information criterion, a predicted residual sum of squares and a Bayesian information criterion is creatively adopted, and an agent model for calculating thrust is constructed to replace a high-fidelity thrust model to solve and obtain optimal design parameters. The error between the optimal solution obtained by the method and the optimal solution obtained by using the high-fidelity model is very small, the calculation cost is remarkably reduced, and multidisciplinary efficient collaborative optimization is realized.

The invention relates to the technical field of paving thickness measurement, in particular to an ultrasonic paving thickness detection method and system, and the method comprises the steps: employing an Akaike information criterion for an echo measurement sequence obtained through collection of a paving road, and obtaining an AIC value of all elements of the echo measurement sequence based on the numerical distribution of AIC values of all elements of the echo measurement sequence; an echo positioning sequence and an echo AIC statistical sequence are obtained; element distribution and change characteristics in the echo positioning sequence are analyzed, and the change trend of each element in the echo AIC statistical sequence is obtained to obtain a signal mutation trend degree; obtaining an echo development trend degree based on a straight line fitting result of each element in the echo AIC statistical sequence; and confirming a primary echo moment, and calculating the paving thickness by combining the propagation speed of the ultrasonic waves in the paving mixture. The invention aims to improve the accuracy of paving thickness measurement.

The application discloses a fault probability monitoring and regulation system and method, relates to the technical field of fault monitoring, and collects to-be-determined related parameters in each period through a collection module, determines the optimal order of an autoregressive model of each related parameter by using the Akaike information criterion, and verifies the to-be-determined related parameters in each period, decides whether to perform secondary collection and replacement based on a verification result, generates a related parameter vector of each period, and starts a window translation along a period in a reverse order from the related parameter vector of each period through a monitoring module, calculates and arranges a comprehensive feature vector in each window, extracts and generates a core feature vector set through a time series self-encoder, trains a hidden Markov model by using a forward-backward algorithm, and obtains the fault probability of each period based on Bayesian inference, compares the fault probability with a probability threshold value to determine whether a fault exists, and executes regulation when the fault exists, so that high-precision fault monitoring and regulation based on collection self-checking, multi-dimensional data fusion and probability inference are realized.

The invention discloses a low-signal-to-noise-ratio impact signal first arrival moment extraction method, which comprises the following steps of: firstly, calculating an AIC (Akaike Information Criterion) curve of an original low-signal-to-noise-ratio impact signal according to an AIC (Akaike Information Criterion); secondly, smoothing preprocessing is conducted on the AIC curve, and then a first-order difference curve is calculated to highlight abrupt change characteristics at the first arrival moment of the signals; and finally, detecting the starting point of the first significant mutation signal through statistical analysis of the mutation characteristics of the first-order difference curve, and determining the moment corresponding to the starting point as the first arrival moment of the impact signal. According to the method, first arrival time extraction is directly carried out on the original low-signal-to-noise-ratio impact signal based on the statistical characteristics of the AIC curve of the original signal, denoising preprocessing does not need to be carried out, and the method has the technical advantages of being high in extraction precision, high in anti-interference capacity, low in calculation complexity and wide in application range; and the method shows good stability and robustness under different signal-to-noise ratio conditions.

Disclosed are methods and techniques of detecting network anomalies and responding to the anomalies once detected. The methods, for example, include receiving, by a model executed by a processor, real-time log data of an operating network; parsing, by the model executed by the processor, the log data to identify one or more metrics; determining, by the model executed by the processor, a seasonality of the one or more metrics; determining whether the model should use an autoregressive model if seasonality is detected; and on determining that an autoregressive model should be used, training a model based on determining a grid search for a parameter based on an Akaike information criterion.

The invention relates to the technical field of intelligent power grid and distributed energy management, in particular to a distributed energy collaborative management system based on edge intelligence, which comprises an end side sensing layer, an edge intelligent node, a cloud collaborative layer, a secure communication layer and a visual interaction layer. According to the scheme, the multi-scale weighted regression model is constructed, the multi-scale space-time characteristics of the distributed energy sources are analyzed, the self-adaptive weight matrix is constructed to accurately capture the regional characteristics, the scale parameters are optimized by using the corrected akaike information criterion, and wind and light curtailment and power supply shortage are remarkably reduced; according to the scheme, the optimized self-adaptive differential evolution algorithm is adopted for regional global optimization scheduling, the scaling factor and the crossover probability are dynamically adjusted based on the historical memory bank, millisecond-level response is achieved in cooperation with a linear population reduction mechanism, the system operation cost is reduced, and the energy utilization efficiency is improved.

The application discloses a signal first arrival time picking method and system, electronic equipment and a storage medium. The method comprises the following steps: obtaining a target signal by collecting and preprocessing to-be-processed data; performing amplitude-aware permutation entropy calculation on the target signal to determine a target interval containing a first arrival time; picking an improved Akaike information criterion minimum value point in the target interval as the first arrival time; and completing a geophysical data processing process according to the first arrival time. The embodiment of the application can improve the picking efficiency and the signal picking accuracy under the condition of low signal-to-noise ratio, and can be widely applied to the near-surface engineering geophysical technology field.

Embodiments of the present application relate to the field of microseismic monitoring, and disclose a microseismic first arrival picking method, device, equipment and storage medium. The center time point of the time window of the microseismic signal is calculated according to the propagation path of the microseismic signal and the preset layer velocity, and the length of the time window is determined according to the center time point and the preset number of time sampling points, so as to constrain the time window of the first arrival point of the microseismic signal picked by the Akaike information criterion method, and the layer velocity is constantly updated in an iterative manner, and the time window is further updated, so as to determine the best first arrival point of the microseismic signal. Compared with the traditional Akaike information criterion method, the application has better noise immunity, can avoid the "multi-peak" phenomenon of the Akaike information criterion method under the condition of low signal-to-noise ratio, and thus avoids the misjudgment of the first arrival point.

The invention discloses an interpretable motor bearing fault diagnosis method based on hybrid feature selection. The method comprises the following steps: collecting vibration signals of a motor bearing in each operation mode; performing wavelet denoising on the vibration signals in each operation mode and extracting various features to obtain an original feature set; the multiple features comprise time domain features, frequency domain features and time-frequency domain features; performing preliminary screening on multiple features in the original feature set by using a Laplace score method to obtain a pre-selected feature set; screening features in the pre-selected feature set by using a genetic algorithm and a fitness function constructed based on AIC (Akaike Information Criterion) to obtain an optimal feature subset; constructing a belief rule base BRB fault diagnosis model according to the optimal feature subset; and obtaining an observation signal of the motor bearing, and performing fault diagnosis by using the BRB fault diagnosis model. According to the invention, a two-stage feature screening method is introduced, so that the fault feature with the best discrimination capability for the motor bearing fault can be obtained.

The present invention relates to the technical field of non-destructive testing, and provides a nonlinear system-based structural damage testing method and apparatus. The method comprises: applying an excitation force to a test object under test, so that said test object vibrates under the excitation force, performing signal acquisition on said test object by means of a laser vibrometer to obtain an excitation signal and a response signal, using the excitation signal and the response signal together with a plurality of imported order combinations to construct initial NARX models corresponding to different orders, screening out a target NARX model from among the plurality of initial NARX models on the basis of the Akaike information criterion and optimizing the target NARX model to obtain an NARX model under test, comparing order information in the NARX model under test with order information in a reference NARX model, and determining, on the basis of an order comparison result, whether said test object has a defect. A nonlinear model is constructed by means of signal data capable of expressing nonlinearity of a test object under test, parameter optimization is performed on the model, so that the model more accurately represents characteristics of said test object, and then whether said test object has a defect is determined.

This invention provides a method for visualizing the synthesized spectrum of vibration signals, comprising the following steps: First, the acceleration signal is integrated twice to obtain the velocity and displacement signals respectively; Fast Fourier Transform (FFT) is performed on the acceleration, velocity, and displacement signals to obtain the spectra of the three signals. Based on the spectra of the acceleration, velocity, and displacement signals, the cutoff frequency range of the synthesized spectrum is determined using the Akaike Information Criterion (AIC). The amplitudes of the three spectra are appropriately scaled according to the cutoff frequency range, finally yielding a synthesized spectrum containing information from the acceleration, velocity, and displacement signals. This invention overcomes the problem that analyzing a single high-frequency acceleration signal and low-frequency velocity and displacement signals makes it difficult to comprehensively capture spectral information when analyzing vibration signals. According to practical needs, the synthesized spectrum visualization provides a more comprehensive display of the spectral information of vibration signals, making vibration signal analysis more convenient and efficient.

This invention relates to a transit time estimation method based on the Akaike Information Criterion of waveform similarity, comprising the following steps: extracting effective ultrasonic waveform data; designing an FIR bandpass filter to perform zero-phase delay filtering and noise reduction processing on the effective ultrasonic waveform data; estimating transit time based on the Akaike Information Criterion of waveform similarity; and achieving super-resolution transit time correction based on phase compensation.

The invention provides a TTF early warning method and system aiming at EGFR TKIs single drug intervention NSCLC, and is applied to the technical field of medical data processing. The method comprises the following steps: processing patient data to generate treatment failure time data; constructing a time-dependent Cox model, describing the influence of each variable on a risk function through the model, calculating the risk ratio among the variables based on the risk function model, and generating risk ratio data; analyzing each covariable based on a univariate time-dependent Cox model, screening variables with statistical significance or clinical correlation with patient survival in the univariate time-dependent Cox model, and adding the screened variables into a multivariate time-dependent Cox model; processing the multivariable time dependence Cox model based on parameter accuracy, proportional risk hypothesis test, akaike information criterion, consistency index and clinical correlation to generate a model evaluation result; and processing the target patient information to generate target scoring information.

The invention relates to an equipment life prediction method and device based on multi-modal fusion. The method comprises the following steps: acquiring equipment health degree time sequence data; performing quality evaluation based on the time sequence data, and calculating a comprehensive quality score; a Savitzky-Golay filtering parameter is dynamically adjusted on the basis of the comprehensive quality score; on the basis of the adjusted Savitzky-Golay filtering parameters, judging degradation and akaike information criteria through CUSUM, and calculating a first residual life prediction value of the equipment; calculating a second remaining life prediction value of the device based on a physical mechanism of the device; a multi-sensor Wiener random model based on the equipment is constructed; calculating a third residual life prediction value of the equipment through the Wiener random model; and fusing the first residual life prediction value, the second residual life prediction value and the third residual life prediction value through confidence to obtain a final life prediction value of the equipment. Through fusion of multiple life prediction models, the accuracy of equipment life prediction is improved, and the risk of accidental shutdown is reduced.

A method and a device for structural damage detection in a nonlinear system are provided. The method includes: applying an excitation force to a specimen to be detected to induce a vibration phenomenon in the specimen to be detected, and performing signal collection on the specimen to be detected by a laser vibrometer to obtain an excitation signal and a response signal; constructing, according to the excitation signal, the response signal and imported order combinations, initial NARX models corresponding to different orders; selecting, based on an Akaike information criterion, a target NARX model corresponding to an optimal order combination; optimizing the target NARX model to obtain a NARX model to be detected; comparing order information of the NARX model to be detected with order information of a benchmark NARX model, and determining, based on an order comparison result, whether there is a defect in the specimen to be detected.

The application relates to a device life prediction method and device based on multi-modal fusion, which comprises the following steps: acquiring device health time series data; performing quality evaluation based on the time series data and calculating a comprehensive quality score; dynamically adjusting Savitzky-Golay filtering parameters based on the comprehensive quality score; calculating a first residual life prediction value of the device by CUSUM discrimination and Akaike information criterion based on the adjusted Savitzky-Golay filtering parameters; calculating a second residual life prediction value of the device based on the physical mechanism of the device; constructing a Wiener random model based on the multi-sensor of the device; calculating a third residual life prediction value of the device through the Wiener random model; and obtaining a final life prediction value of the device by fusing the first residual life prediction value, the second residual life prediction value and the third residual life prediction value through confidence. The application improves the accuracy of device life prediction and reduces the risk of accidental shutdown by fusing multi-life prediction models.

The application discloses a water quality prediction method based on ARIMA and an improved chicken swarm algorithm, and comprises the following steps: acquiring water quality data from an initial period to a t period; performing stationarity test on the water quality data by using a unit root test method to determine a difference order d; determining an autoregressive order p and a moving regression order q by using an Akaike information criterion AIC or a minimum Bayesian information criterion BIC; establishing an ARIMA water quality prediction model according to the difference order d, the autoregressive order p and the moving regression order q, obtaining water quality prediction data of the t period through the ARIMA water quality prediction model; introducing an optimization operator, re-encoding the chicken swarm algorithm in combination with the water quality prediction data of the t period; initializing a population, defining a fitness function and stipulating a chicken swarm swimming strategy; performing chicken swarm identity attribution on the water quality prediction data according to the size of the fitness, and performing position updating through different identities to obtain optimal water quality prediction data of the t period. The application can realize more accurate prediction of water quality data.

The invention discloses a power system operation state evolution analysis method, device and equipment and a medium, and the method comprises the steps: obtaining the output data of various generator sets in a plurality of geographic regions and the power load data corresponding to each geographic region, so as to generate a power system operation time sequence; outputting a state transition probability matrix through a pre-trained hidden Markov model according to the operation time sequence of the power system; wherein the hidden state number of the hidden Markov model is determined according to an akaike information criterion and a Bayesian information criterion so as to ensure the optimality of the model; the state transition probability matrix is used for identifying an implicit operation state in the operation process of the power system; and acquiring real-time operation data of a target power system, and predicting an operation state corresponding to a preset future time sequence according to the real-time operation data and the state transition probability matrix. Compared with the prior art, the dynamic property and the accuracy of the evolution analysis of the operating state of the power system can be improved.

This invention discloses a method for constructing a multilayer ionospheric model based on prior constraints, comprising the following steps: obtaining the ionospheric electron density matrix output by the NeQuick-G model; processing the ionospheric electron density matrix using principal component analysis, and determining the number of ionospheric layers based on the cumulative contribution rate of the principal components; clustering the ionospheric electron density and corresponding height using the K-means clustering algorithm based on the number of layers to obtain the height and boundary of each layer; obtaining simulated total electron content data based on the height and boundary of each layer; selecting the optimal polynomial order for each layer and each modeling time period from multiple candidate polynomial orders using the Akaike information content criterion, and constructing a polynomial function model for each layer and each time period; constructing an ionospheric model using the original GNSS observation data based on the number of layers, the height and boundary of each layer, and the polynomial function model, and solving the ionospheric model parameters.