10 results about "Field transformation" patented technology

The application discloses a kind of deep neural network gas identification method and system based on internet of things electronic nose, gas sample response signal obtained by gas sensor array collection is sent to cloud data center by terminal equipment with sensor array and internet of things module.In the pre-processing of data to microcontroller MCU, gram angle field transformation is carried out, so that the two-dimensional sensor response data received by host computer becomes three-dimensional data that can be input into convolutional neural network after dimensionality increasing, and the gas label is output through classifier, which can represent gas category and concentration grade, realizes the real-time, anti-interference and high detection accuracy of remote monitoring.Gives play to the advantages of strong feature extraction ability, fast model convergence and high recognition accuracy of convolutional neural network.It can be widely applied in environmental detection, industrial production, medical treatment and safety fields.

This invention discloses a method and device for fault type diagnosis and fault segment location in distribution networks. The method includes selecting eight-channel waveforms from a single measurement point on a faulty line: three-phase zero-sequence voltage, zero-sequence voltage, three-phase current, and zero-sequence current as the judgment waveforms. Based on fault waveform datasets collected from actual lines and batch simulation waveform datasets, a high-quality waveform dataset with broad coverage and balanced type distribution is generated through a conditional adversarial generative network. Each channel of the fault waveform undergoes classification and normalization, feature regions are extracted, and then the dimensions are expanded into a two-dimensional field matrix through field transformation. The eight sets of two-dimensional matrices are preprocessed into eight-channel two-dimensional matrix tensors. These tensors, combined with fault type and fault segment labels, form a model training set. This invention enables the classification and identification of fault types and fault segments in fault waveforms. This invention can accelerate fault investigation, improve the power supply reliability of distribution networks, and provide technical support for ground fault diagnosis that relies entirely on data-driven methods.

The application discloses a long-text-oriented episodic screenplay generation method and system, and belongs to the technical field of artificial intelligence content generation and digital video production. The method forms multiple episode objects through voice-over role initialization, text semantic compression and episode segmentation, and extracts role candidate objects and location time sequence candidate objects; based on normalization rules, an identity register of roles and a location time sequence register are constructed respectively, and unified identification reuse or new registration is performed on the candidate objects; under the constraint of the register, scene planning results and structured screenplay constraint patterns are generated, and multiple generation agents are called to output screenplay objects; and then, through structure verification, field completion and unified field transformation, the standardized results of episodes, roles, locations and screenplays are output. The application can improve the consistency of cross-episode roles and locations, the stability of the screenplay structure and the reuse efficiency of downstream production.

The application discloses a kind of double-channel rolling bearing fault diagnosis methods of image time sequence fusion, belong to rotating machinery intelligent fault diagnosis technical field.It includes: collecting rolling bearing vibration signal, constructs fixed length one-dimensional time sequence sample by segmented pretreatment, and generates corresponding image by gram angle difference field transformation;Then respectively establish image channel and time sequence channel, image channel extracts overall spatial feature and local detail feature, time sequence channel uses LSTM to extract time sequence dependent feature, and sets fusion node in multilevel;Further, the corresponding level features of two channels are complementarily fused based on joint context channel-by-channel gating, obtain joint fault feature layer by layer, finally input classification module to complete fault identification.The application solves the problem of insufficient single feature expression and insufficient cross-modal information fusion in rolling bearing fault diagnosis, and improves the diagnosis accuracy and robustness.

This invention discloses a noise separation method and system based on spatial sound suppression and equivalent field transformation. The method includes: constructing mutually independent target space, observation space, and virtual sound insulation space; constructing the sound pressure transmission relationship between the target space, observation space, and virtual sound insulation space based on analytical Green's function; establishing a mathematical model using the virtual source point location in the target space, and introducing weighting coefficients to construct a weighted calculation model for the sound pressure at the point to be measured in the observation space; constructing a regularized optimization problem with physical constraints, using the zero sound field amplitude in the virtual sound insulation space as a constraint, and solving for the weighting coefficients of the observation array; using the weighting coefficients to perform spatial domain weighting processing on the signals acquired by the observation array, and extracting the target sound field signal. This invention solves the technical problems of low separation accuracy, numerical instability, and high sampling requirements of existing equivalent source methods under strong noise interference environments, and has many advantages such as strong noise suppression capability, low engineering cost, and high flexibility.

This invention provides a random field modeling method that accurately quantizes the intrinsic correlation of initial defects in a spherical shell and further improves the accuracy of spherical shell performance analysis. [Solution] A random field modeling method for quantizing initial defects in a spherical shell structure includes: determining simulation spherical shell data corresponding to a target spherical shell, including geometric information and material information of the spherical shell; determining a target correlation function and representative discrete points corresponding to the target spherical shell based on the geometric information and material information of the spherical shell; determining a correlation coefficient matrix based on the target correlation function and representative discrete points; decomposing the feature values ​​of the correlation coefficient matrix to obtain a standard random field; and performing a random field transformation on the standard random field based on a random field transformation formula to obtain a target random field corresponding to the target spherical shell.

This application provides a method, electronic device, and storage medium for evaluating and detecting the near-field planar properties of an antenna. The method includes using an electromagnetic field probe to move in a grid pattern within a two-dimensional plane at λ / 2 sampling step intervals, simultaneously measuring the amplitude and phase values ​​of the electric field vector at each sampling point; analyzing and identifying sub-regions of feed phase faults based on phase distribution characteristics; analyzing and outputting the boundaries of adjacent regions of continuous anomalies based on the sub-regions of feed phase faults and near-field amplitude anomalies, performing far-field transformation on these adjacent regions, and outputting far-field radiation characteristic data for verifying the anomaly results of these regions. This processing method, through anomaly region identification and association, secondary detection verification, combined with near-field fine scanning positioning and far-field back-calculation verification, forms a closed-loop diagnostic logic. This approach improves the reliability of the final output anomaly results and reduces quality misjudgments.

The application discloses a kind of based on multichannel GADF-CNN-LSTM gearbox fault diagnosis method and system.Method includes: using sensor synchronous acquisition gear box X axis, Y axis, Z axis three directions vibration signal, and sample cutting is carried out by sliding window technique;Utilize segmented aggregate approximation algorithm to reduce dimension to signal, and construct two-dimensional feature matrix of each direction by gram angle difference field transformation respectively;Three-way feature matrix is mapped to R, G, B channel of color image respectively, and three-channel RGB gram angle field difference image is generated by fusion;CNN-LSTM hybrid deep learning model is constructed, spatial texture feature of image is extracted using CNN, and deep time sequence dependence relationship is mined using LSTM, to realize fault classification.The application can effectively fuse multidimensional space vibration information, combined with the sensitivity of GADF to transient impact and the spatiotemporal feature extraction capability of hybrid model, significantly improve the accuracy and robustness of gearbox fault diagnosis.

This invention discloses a plane wave synthesizer, testing system, and method based on a dual / multi-band co-aperture ultra-wideband Vivaldi dual-polarized array. The plane wave synthesizer includes an amplitude and phase controller, and an ultra-wideband Vivaldi three-band co-aperture antenna array or an ultra-wideband Vivaldi dual-band co-aperture antenna array. The testing system includes the plane wave synthesizer, a microwave anechoic chamber, a turntable within the anechoic chamber, and the device under test (DUT). The microwave anechoic chamber also includes an electromagnetic shield and radio frequency absorbing materials around its perimeter. During testing, the amplitude and phase controller generates multi-band radio frequency excitation signals to drive the ultra-wideband Vivaldi three-band co-aperture antenna array or the ultra-wideband Vivaldi dual-band co-aperture antenna array, forming a uniform plane wave in the near-field radiation. This invention eliminates the need for near-field and far-field transformation of electromagnetic waves, reducing the required space by three-quarters compared to systems based on compact fields, making it particularly suitable for generating large-size ultra-wideband quiet zones.

Disclosed are a jacket underwater crack detection system based on acoustic emission signals and a method thereof, which adopts an artificial intelligence detection algorithm based on deep learning, takes the acoustic emission echo signals of a jacket to be detected by an acoustic emission sensor as input data, and after field transformation of the acoustic emission echo signals in a time sequence under a Cartesian coordinate system to a polar coordinate system, extracts feature distribution information about the jacket focusing on spatial positions and channel contents in the acoustic emission echo signals, and detects and judges whether the jacket has underwater cracks by fusing the feature information of the two. In this way, the jacket can be accurately and intelligently detected for underwater cracks to ensure the safety of offshore wind power jackets.