470 results about "Signal transformation" patented technology

The invention relates to the technical field of data processing, in particular to a metal wire data analysis method and system. The method comprises the following steps: acquiring wire interface data of composite metal; acquiring a wire rod interface characteristic matrix based on the wire rod interface data; constructing a wire rod interface characteristic model according to the wire rod interface characteristic matrix; differential signal conversion processing is carried out on the wire rod interface characteristic model, and wire rod interface change characteristics are extracted based on a differential signal conversion result; constructing a wire rod interface defect initial indication diagram according to the wire rod interface change characteristics; performing multi-scale decomposition processing on the basis of the wire rod interface defect initial indication graph to obtain interface defect feature data; and identifying wire interface defect type features according to the interface defect feature data. According to the method, a closed-loop optimization mechanism of defect characteristics and production process parameters is established, and the interface quality and reliability of the high-performance composite metal wire are remarkably improved.

The invention requests to protect a rotating machine noise robust fault diagnosis method based on a mixed domain graph neural network. Comprising the following steps: collecting vibration data, and dividing an original vibration signal into a plurality of sections of subsamples with the length of 1024 by using non-overlapping sampling; converting the constructed subsample signal vibration signal into a frequency domain signal by using fast Fourier transform (FFT); taking the subsample frequency domain signals as nodes, the frequency spectrums as node features and the fault types as node labels, and then constructing the nodes into a graph structure; dividing the generated graph structure data set into a training set and a test set; and performing fault diagnosis on a test set by using the trained model, realizing self-adaptive weight adjustment by using a gating mechanism to fuse mechanical features, outputting a diagnosis result, calculating a performance index and performing result visualization analysis. The result shows that compared with a single-domain graph neural network, the method has higher accuracy and noise immunity, and end-to-end rotating machine fault diagnosis under the strong noise condition can be effectively achieved.

The invention provides an audio-visual voice separation method and system for a scene with a flexible number of speakers, and the method comprises the steps: extracting mixed voice signals of all speakers in audio data, converting the mixed voice signals into audio features for representation; according to the method, a multi-head self-attention branch and a multi-layer perceptron branch with convolution gating are used for coding from a mixed voice signal to obtain dependency relationships of different scales, the audio feature representation is combined to obtain final mixed voice features and block the mixed voice features, each block is separated in parallel and independently, a mask is estimated for each potential speaker, and a multi-level perceptron branch with convolution gating is obtained. The visual features are fused to obtain cross-modal fusion features of the blocks, modeling is carried out on the cross-modal fusion features on the inter-block level, and mask estimation of the blocks is obtained; and reconstructing the separated signal in the time domain according to the estimated mask, judging whether the extracted voice is effective or not, and estimating the number of speakers. The problem that most voice separation methods cannot separate speakers with the number changing dynamically can be solved at the same time.

The invention discloses a multi-mode sensing electromechanical actuator fault prediction system and a control method thereof, and relates to the field of intelligent maintenance of electromechanical integrated equipment. Comprising the following steps: S1, multi-mode sensor deployment and signal acquisition: cooperatively deploying a vibration sensor, a current sensor and a temperature sensor at a bearing, a motor winding and a driver of an electromechanical actuator; and S2, bionic pulse coding and time synchronization: converting the three-mode signal into a sparse pulse stream with time synchronization through a self-adaptive integral distribution model with dynamic threshold adjustment. Through bionic multi-mode sensing, brain-like edge calculation and dynamic fault-tolerant control, real-time health management of the high-precision electromechanical system is realized, the equipment reliability is remarkably improved, and the operation and maintenance cost is reduced. The method is suitable for industrial application with precise assembly requirements such as automobile welding lines, aerospace machining and metal 3D printing.

The invention belongs to the technical field of intelligent fault diagnosis, and discloses an aero-engine bearing fault diagnosis method based on clustering enhancement domain generalization. Firstly, a time domain vibration signal is converted into frequency domain representation through fast Fourier transform, a two-stage convolutional neural network architecture comprising a domain alignment encoder and a classification encoder is constructed, and cross-working-condition feature alignment under adversarial training is realized by using a domain discriminator; introducing a multi-source domain maximum mean difference statistical alignment strategy and a clustering enhanced triple loss mechanism, and synchronously optimizing intra-class feature compactness and inter-class separability through pseudo label clustering center anchoring and dynamic interval constraint; and gradient inversion adversarial training and multi-modal feature fusion are combined. The method provided by the invention can effectively improve the accuracy and robustness of generalization judgment when the aero-engine bearing fault diagnosis lacks target domain data.

The invention discloses a temperature effect monitoring method and system for a metal roof, particularly relates to the technical field of building structure health monitoring, and is used for solving the problems of decoupling of temperature monitoring data and structural mechanical response, failure of a static early warning threshold and the like in the prior art. Acquiring three-dimensional temperature field data through a distributed temperature sensing network, and extracting a temperature gradient characteristic value from the three-dimensional temperature field data; a piezoelectric film sensor is used for collecting constraint point micro-vibration signals, and dynamic constraint reaction force spectrum characteristic values are extracted through signal transformation and manifold learning; establishing a temperature gradient and mechanical characteristic coupling model to generate a temperature stress evaluation value; dynamically adjusting an early warning threshold range according to the environmental parameters; when the evaluation value exceeds a threshold value, a structure safety early warning signal is output; multi-physical field coupling analysis of temperature field distribution and structural response is achieved, an environment-adaptive dynamic early warning mechanism is constructed, and the accuracy and reliability of metal roof temperature effect monitoring are remarkably improved.

The invention provides an intelligent rotating machine fault diagnosis method, and belongs to the technical field of mechanical fault diagnosis, and the method comprises the steps: firstly, obtaining a vibration signal to be subjected to fault diagnosis under a source domain working condition and a target domain working condition, and carrying out the noise reduction processing; secondly, converting the denoised signal into a two-dimensional time-frequency image through short-time Fourier transform, and respectively making a source domain data set and a target domain data set; thirdly, constructing an attention enhancement network (SE-CNN-BiLSTM-MSMHA), and performing pre-training on the model based on the source domain data set to obtain a source domain working condition model; then, freezing part of parameters of the source domain working condition model, and finely adjusting the parameters of the source domain working condition model by using the target domain data to obtain a target domain working condition model; and finally, inputting target domain data to realize fault diagnosis of target domain working conditions. The method can provide a scientific basis for improving the cross-working-condition fault diagnosis accuracy under the strong noise background.

The invention relates to the technical field of data collection control, in particular to a connection transmission control method and system for PCB transmission, and the method comprises the steps: carrying out the data collection of a PCB card when the PCB card is detected to reach a preset scanning region, and obtaining the identification information of the PCB card; the identification information is transmitted to the MES subsystem to obtain a first instruction, and the first instruction is an instruction which is sent to the PLC subsystem after the MES subsystem receives the identification information and carries out processing based on the identification information; performing signal conversion on the first instruction to obtain a target instruction; and controlling the PCB card to carry out transmission according to the target instruction. According to the method, through cooperative work of the PLC subsystem and the MES subsystem, automatic data acquisition and transmission control of the PCB card are realized, manual intervention is reduced, continuity and real-time performance of a production process are ensured, and the automation level of a production line is improved.

Disclosed in the present invention are a power grid impedance measurement method and apparatus. The method comprises: generating a characteristic harmonic current instruction on the basis of a harmonic-signal generation algorithm, and inputting the characteristic harmonic current instruction into a direct-current voltage feedback control loop; receiving the characteristic harmonic current instruction to generate a control signal, and performing current tracking control to obtain a harmonic current to be injected; converting an information electronic signal of a main circuit into a driving signal in a device control loop, and selecting a preset harmonic injection mode in real time on the basis of a harmonic injection target frequency; when active harmonic injection is selected, controlling the turning-on and turning-off of a switch tube, and injecting said harmonic current into a power grid; when passive harmonic injection is selected, controlling the turning-on and turning-off of an IGBT, controlling a time and a current flow direction of when a capacitor is connected to the power grid for operation, and injecting said harmonic current into the power grid; and selecting voltage and current signals to perform signal decomposition, processing and analysis, so as to obtain a power grid impedance. By means of the present embodiment, a power grid impedance is effectively measured, the disturbance to a power grid during impedance measurement is reduced, and the measurement precision is improved.

A secure system for sending and receiving remote signals that control semiautonomous critical systems and infrastructure. The secure system includes a plurality of independent operator stations to provide command signals to the remote critical systems and infrastructure by means of wired and / or wireless communications. Decoupling of the secure system from raw signals is performed by a data broker that serves to check for time-dependent and signal-dependent corroboration between the raw signals and then provide limited-scope instructions to the secure system controller. Further security is afforded to the secure system through encryption of the incoming command signals, convoluting the encrypted command signals within data structures, and padding with erroneous information with the implementation of a synthesized command signal that renders successfully decrypted and deconvoluted command signals neither readable by humans or machines. The final method of data security is appropriated through the synthesis of transmitter / receiver unique “languages” that serve to translate decrypted and unconvoluted signals into their final human / machine readable instructions.

To provide a dimension measurement device which can efficiently perform dimension measurement of a work material in one direction with high accuracy, and to provide a cutting tool system and a dimension measurement method.SOLUTION: A dimension measurement device includes a sensor unit and a measurement device body. The sensor unit includes: a first distance sensor using an eddy-current sensor; and a signal conversion unit which outputs output signals of the distance sensor as detection values at a preset time interval. The measurement device body includes: a detection control unit which detects a distance to a work material with the sensor unit while rotating the work material; an acquisition unit which acquires the multiple detection values output from the sensor unit at the time interval in a preset specified time period; a calculation unit which calculates an average value of the detection values acquired by the acquisition unit; and a result output unit which outputs a measurement result of the work material based on the average value of the detection values calculated by the calculation unit.SELECTED DRAWING: Figure 1

The invention relates to the technical field of microelectronics, semiconductors and communication, and discloses a radio frequency switch chip, and the circuit structure of the radio frequency switch chip comprises a transformer type differential inductor T-L1, a transformer type differential inductor T-L2, a capacitor C1, an NMOS transistor M1, an NMOS transistor M2, a gate resistor RG1, and a gate resistor RG2. The radio frequency switch chip provided by the invention adopts a differential structure, so that the parasitic inductance effect of a grounding key and a wire during packaging and testing can be eliminated, and the insertion loss and the isolation performance are improved; compared with a single-end port, the lower impedance of the differential port is more beneficial to improving the power capacity of the switch; the differential structure allows the transformer type differential inductor to use differential signal characteristics, the equivalent quality factor of the inductor is improved, and the circuit performance is improved; the differential structure can improve the linearity of the circuit, especially the second-order intermodulation effect; the differential structure can be compatible with a modern silicon-based CMOS radio frequency system, and an additional balun is not needed for signal conversion.

The invention discloses a multi-sensor cooperative partial discharge waveform identification method and system, and the method comprises the steps: extracting an envelope line of an oscillation waveform through Hilbert transformation for a collected partial discharge signal, carrying out the Logistic function fitting and exponential decay function fitting of each sensor signal, and carrying out the recognition of the partial discharge waveform through the Logistic function fitting and exponential decay function fitting. Taking the characteristic parameters extracted by each sensor as node characteristics in a graph structure, establishing an edge connection relation in a graph according to the maximum value starting time difference and the similarity of signals between the sensors, constructing a graph neural network model, performing training and reasoning by utilizing the graph structure, and finally outputting a recognition result of whether a partial discharge event occurs or not; the system comprises a signal conversion module, a feature extraction module, a data fusion module and a partial discharge diagnosis module. According to the invention, the accuracy and adaptability of partial discharge signal identification are improved, and the on-line monitoring requirement of the transformer in a complex electromagnetic environment is met.

The invention relates to the technical field of optical current transformer detection, and provides an optical current transformer real-time calibration and health diagnosis method and system, and the method comprises the steps: carrying out the harmonic analysis of an initial digital signal sequence obtained through the conversion of a continuous modulation optical signal of a target optical current transformer, and obtaining harmonic component data; physical state evaluation is carried out according to the higher harmonic component data and preset health state baseline data to generate a physical state evaluation result, nonlinear disturbance compensation is carried out according to an error control signal generated based on the higher harmonic component data, and after a linear working point is reached, nonlinear disturbance compensation is carried out; and calculating an electrical measurement value according to the output target fundamental component data, and performing comprehensive analysis according to the physical state evaluation result and the electrical measurement value to generate a diagnosis report. According to the invention, through a measurement calibration and state diagnosis mechanism based on harmonic analysis, while the electrical performance measurement precision is effectively improved, reliable quantitative diagnosis and accurate traceability of different physical faults in the equipment are realized.

The invention discloses a power electronic device fault early warning method based on multi-physics field coupling, and the method comprises the steps: (1) collecting aging data of a power electronic device under the condition of electric field and thermal field coupling, and dividing the aging data into a source domain data set and a target domain data set; (2) establishing a power electronic device aging parameter screening method, selecting a parameter which can best represent an aging rule as an aging precursor, and converting discrete voltage and current signals into time sequence signals which can be accepted by a neural network; (3) compressing the time sequence signal set into sliding window graph nodes, calculating a distance relationship between the graph nodes, and obtaining a graph matrix containing data structure information; (4) building a space-time fusion fault predictor, wherein the space-time fusion fault predictor comprises a feature extractor, a time sequence predictor and a domain discriminator; and (5) a weak supervision adversarial learning strategy is adopted, a small amount of target domain label data participates in the model training process, and the source domain prediction loss, the target domain prediction loss and the discrimination loss are combined for training.

The invention relates to the field of rotating machine fault diagnosis, in particular to a rotating machine data enhancement method based on an SCF-CVAE-GAN network, and the method comprises the steps: obtaining rotating machine vibration signals in different fault modes; the rotating machinery vibration signals are converted into a two-dimensional time-frequency image, and the SCF-CVAE-GAN network is trained by adopting the two-dimensional time-frequency image; generating a two-dimensional time-frequency image sample by adopting the trained SCF-CVAE-GAN, calculating the structural similarity and the distance score of the generated sample, and evaluating the quality of the generated sample; mixing the generated sample meeting the quality requirement and the two-dimensional time-frequency image to obtain an enhanced data set; according to the method, a double-self-correction network is provided, a frequency domain regularization strategy is integrated, global information and local textures of data are deeply mined, and the diversity and fidelity of generated samples are ensured by constraining distribution of the samples in a frequency domain.

The invention discloses an intelligent start-stop control method and system for a magnetic suspension refrigeration compressor, and the method comprises the steps: obtaining a position perception enhancement source through collecting and analyzing the current response data of a magnetic bearing coil, and extracting the current response change characteristics; rotor position estimation is carried out based on current response change characteristics, a positioning precision driving source is extracted from estimation deviation, and a non-impact starting excitation current instruction is generated; a multi-sensor fusion data network is established, configuration is reconstructed through an abnormal response activation signal, and a degradation operation fault-tolerant mode is formed; converting a response preparation gain according to the shutdown instruction signal, performing shutdown state evaluation in combination with signal reconstruction configuration, and generating an emergency shutdown trigger condition; and the power supply state of a bus capacitor is detected, gradient control potential is released, the active damping control opportunity is determined, a damping force control instruction is generated, finally intelligent start-stop control over the magnetic suspension refrigeration compressor is achieved, and safe and reliable operation of the system under various working conditions is ensured.

A data storage system for managing an operation of a solid-state device (SSD) includes the SSD, an adaptor coupled to the SSD via a first bus protocol and including a non-volatile memory storing a power configuration table. The protocol converter is configured to: couple to the adaptor using the first bus protocol, convert first signals of the first bus protocol to second signals for a second bus protocol, and couple to a host using the second bus protocol. The power configuration table includes power profiles for the SSD each corresponding to an interface type for the second bus protocol. The protocol converter is further configured to: determine an interface type of the second bus protocol, read the power configuration table from the non-volatile memory to determine a power profile corresponding to the determined interface type, and cause the SSD to operate using the determined power profile.

The invention discloses a water level early-warning device for a water conservancy project. The water level early-warning device comprises a floater, a limiting mechanism, a mounting frame, a signal conversion and processing device and display equipment, wherein a pressure sensor is arranged at the bottom of the floater; the floater is sleeved with a flow blocking pipe, the flow blocking pipe is rotationally connected with the mounting frame, a plurality of leaf plates are evenly arranged on the outer side of the lower portion of the flow blocking pipe along the rotating shaft, the leaf plates are vertically arranged, and the floater is located in the flow blocking pipe; the processor is electrically connected with an alarm and wirelessly connected with a mobile terminal. The floater is limited through the limiting mechanism, so that the floater can only move up and down to drive the sliding piece to move up and down, water level information is directly obtained through the signal conversion and processing device and displayed on the display device, and the water level sensor is simple and reasonable in structure, low in manufacturing cost, compact in structure, small in resistance in the working process, smooth and sensitive in action and high in measurement result accuracy; the measurement result is directly displayed; whether the floater works normally or not is judged through pressure information fed back by the pressure sensor; and a flow blocking pipe is arranged, so that the device is also suitable for large water waves.

In one embodiment, a method includes accessing symbol sequences with each including multiple symbols, generating a respective frequency-domain signal for each of the symbol sequences based on a respective signal transformation algorithm that is based at least in part on a Fourier transform, mapping each of the frequency-domain signals to a respective subset of subcarriers out of all subcarriers in an allocated bandwidth, generating a waveform based on the frequency-domain signals and their respective subset of subcarriers based at least in part on an inverse Fourier transform, and transmitting the waveform to a second electronic device.

The invention discloses a rolling bearing fault diagnosis method based on a time-frequency image and transfer learning. The method comprises the following steps: S1, collecting an original rolling bearing vibration signal; s2, converting the vibration signal of the rolling bearing into a time-frequency image; s3, dividing the time-frequency image into a training set, a verification set and a test set, and constructing a convolutional neural network fault diagnosis model; s4, normalizing the sample data of the training set and the verification set, inputting the normalized sample data into the network model, and carrying out iterative training to obtain an optimal model; inputting the sample data of the test set into the optimal model for classification to obtain a model diagnosis result; s5, converting the attention superposition thermodynamic diagram into a weight matrix; and S6, calculating the weight value of each fault sample signal, evaluating the closeness degree of the weight value, and determining the sample type. The invention belongs to the technical field of mechanical state monitoring and fault diagnosis, and solves the problems of insufficient interpretability and non-transparent process of an existing deep learning model in rolling bearing fault diagnosis.

The invention relates to the technical field of precision shell contour metering, and discloses an online metering method for a precision shell contour, and the method comprises the steps: obtaining an original sampling point cloud and a feature point group of a to-be-measured workpiece in a conveying state; resolving a real-time pose matrix of the to-be-measured workpiece relative to a preset measurement reference by using a spatial topology constraint relationship of the feature point groups; carrying out differential processing on a real-time pose matrix in the sampling sequence, and synthesizing an instantaneous motion vector corresponding to a sampling moment; in combination with a signal response time delay constant of the sampling system, phase compensation is carried out on a motion displacement deviator induced by signal conversion delay, and a corrected sampling point cloud is generated; according to the method, the geometric distortion of the point cloud generated by response delay of the sensor is solved, the problem of instantaneous drift of the measurement reference in an unsteady state conveying state is solved, and the contour reduction precision in a complex working condition is improved.

The invention discloses a variable-working-condition well cementation pump valve leakage fault detecting and positioning method, and belongs to the technical field of well cementation equipment safety monitoring and fault diagnosis. According to the method, the time domain vibration signal is converted into the angle domain signal, so that the influence of rotating speed fluctuation on fault feature extraction is effectively eliminated, and the accuracy and stability of fault detection are improved. According to fault detection and positioning based on the Bayesian inference model, the prior probability, the conditional probability and the posterior probability are comprehensively considered, prior process knowledge can be fully utilized, the requirement for a large amount of fault mode data is reduced, the fault positioning precision is remarkably improved, an accurate basis is provided for timely maintenance of the well cementation pump, and the well cementation efficiency is improved. And the production loss caused by overlong troubleshooting time is reduced.

The invention discloses a natural gas pipeline leakage detection method based on novel wavelet basis transformation and singular value decomposition in a two-dimensional convolutional neural network. Firstly, a sound signal collected by a sound wave sensor is converted into a digital signal; secondly, in the data preprocessing stage, singular value decomposition is carried out on the digital signals to effectively eliminate background noise interference, and then batch normalization is carried out on the processed data; then, converting the one-dimensional time sequence signal into a two-dimensional time-frequency image by adopting a self-defined Morlet wavelet basis function; and finally, based on the time-frequency images, constructing and training a 2D-CNN model for fault classification, and presenting a diagnosis result through a confusion matrix and a comparison graph. According to the method, 97.55% of fault recognition accuracy is obtained in a public data set, and compared with other competitive methods, the method shows more excellent noise robustness and classification performance, and has higher accuracy and wide application prospects in pipeline leakage diagnosis in a complex noise environment.

The invention relates to the technical field of biomedical signal processing, in particular to an electrocardiogram reconstruction method and system based on machine learning, and the method comprises the following steps: carrying out iterative initialization; respectively carrying out data preprocessing on the PPG signal and the ECG signal; based on the convolutional neural network and a Transform encoder, constructing a signal conversion model; inputting the preprocessed data into a signal conversion model, carrying out model training, and outputting a predicted electrocardiogram signal; calculating a loss function between the predicted electrocardiogram signal and the target real signal, and optimizing the model parameters based on the loss function; when the number of iterations is smaller than a threshold value, continuously updating model parameters according to a gradient descent method and circularly training, and evaluating model performance and calculating performance indexes in a verification stage; after iteration is completed, the trained network parameters are stored, and reconstructed electrocardiogram signals are output. The electrocardiogram reconstruction method provided by the invention aims to solve the technical problem of accurately reconstructing the electrocardiogram through the photoplethysmogram.

The gas turbine fault prediction and diagnosis method comprises the steps of collecting measurement parameters of a gas turbine; converting the non-stationary signal into a frequency domain feature, and converting the slowly varying signal into a time domain feature; inputting the time domain features and the frequency domain features into a feature screening model, and outputting screened time-frequency domain features; calculating a dynamic correlation coefficient matrix among the screened time-frequency domain characteristics, and calculating a characteristic relation vector based on the dynamic correlation coefficient matrix; inputting the screened time-frequency domain features and the feature relationship vector into a feature coupling model, and outputting a coupling feature vector; performing time sequence feature extraction on the measurement parameters to obtain a time sequence feature vector; and inputting the time sequence feature vector and the coupling feature vector into a fault prediction model to obtain a predicted gas turbine fault type. A time-frequency characteristic coupling relation is quantified through a dynamic correlation coefficient matrix, TCN-FNN dual-channel fusion time sequence and coupling characteristics are realized, and accurate positioning and early prediction of a fault are realized.

The invention discloses a non-contact electrocardiogram generation method and system based on a millimeter wave radar, and relates to the field of electrocardiogram, and the method comprises the following steps: pairing a heartbeat signal RCG collected by the millimeter wave radar with a standard electrocardiogram signal ECG to construct a training data set; and carrying out time sequence feature extraction on the RCG, and learning forward and backward time sequence features through a bidirectional long short-term memory network Bi-LSTM. Signal importance weights at different time points are calculated by adopting an attention mechanism, and weighting processing is carried out to obtain electrocardio dynamic characteristics. And inputting the features into a generator, and performing signal conversion through a deep convolutional network. A double-discriminator structure is utilized to optimize and generate signals, the main discriminator verifies authenticity, and the auxiliary discriminator verifies consistency. And finally, carrying out processing after reverse normalization and smooth filtering to obtain a target electrocardiogram signal. By implementing the method, the accuracy of non-contact electrocardiogram generation can be improved.

The invention discloses a motor fault detection method and system based on time-frequency dual channels. Fault judgment is carried out based on power grid side currents in different states and corresponding fast Fourier transform analysis; after a fault is judged, power grid side current signals under two power frequencies are collected, a fault signal is converted into a time domain image through a Grubby angle field, and the fault signal is converted into a frequency domain image through short-time Fourier transform; a residual convolutional neural network is constructed in a two-channel mode, features are extracted from time domain and frequency domain images respectively, and finally the extracted features are fused for open-circuit fault classification of different types of inverters in a motor system. According to the invention, the existing product does not need to be disassembled, so that a user can deploy conveniently; normal and fault states can be quickly distinguished, and the consumed resources are small; the fault type can be quickly identified, and a user can conveniently process the fault in time.

The invention discloses an unmanned aerial vehicle control method integrating satellite navigation and inertial navigation and related equipment, and relates to the technical field of integrated navigation. According to the method, the control system collects the satellite quality intensity of the unmanned aerial vehicle at the current navigation point and predicts the satellite quality intensity of the unmanned aerial vehicle at the next navigation point, the reliability degree of satellite navigation is evaluated in advance, and the signal conversion type from the current navigation point to the next navigation point and the corresponding navigation strategy are determined accordingly. According to the self-adaptive navigation method based on prediction, preparation can be made before the satellite signal quality changes, and the problems of navigation precision abrupt change and positioning drift easily occurring when the signal abrupt change occurs in a traditional fixed weight fusion scheme are solved. Meanwhile, the control system carries out fusion calculation on satellite navigation data and inertial navigation data and flight control parameter adjustment based on deviation, it is ensured that the unmanned aerial vehicle can stably fly according to a preset route, and the unmanned aerial vehicle is particularly suitable for being used in complex urban environments and mountainous terrains with unstable satellite signals.

The invention discloses a video intelligent monitoring system and method based on cross-modal time-frequency fusion, and relates to the technical field of new energy power monitoring and safety management, and the method comprises the following steps: obtaining multi-modal physical data; detecting a waveform characteristic of the electrical signal, and when the waveform characteristic of the electrical signal meets a preset electrical abnormal condition, performing high frame rate sampling and frame rate compensation on the visible light video to obtain video data with enhanced time sequence continuity; converting the video data, the thermal infrared image, the acoustic signal and the electrical signal to a time-frequency domain to obtain a multi-modal video feature, and inputting the multi-modal video feature to a cross-modal fusion network to obtain a fusion feature vector; and performing anomaly type prediction on the fusion feature vector based on a long-short term memory network in combination with a semantic map model to obtain an anomaly type recognition result and a risk level judgment result. According to the method, the problem that the multi-modal data lacks a unified time reference and fusion mechanism can be solved.