15 results about "Time domain electromagnetics" patented technology

The application discloses a semi-airborne full-time electromagnetic forward method based on an Octree grid, which performs spatial discretization on time-domain Maxwell equations, thereby effectively reducing unknowns of equations to be solved. For full-time electromagnetic field numerical simulation, the application combines the high calculation precision of the second-order backward Euler method with the high OFF-Time calculation efficiency of the model reduction algorithm, adopts the second-order BDF and the model reduction algorithm to respectively solve ON-Time and OFF-Time time-domain control equations, forms a joint algorithm to realize simulation of semi-airborne time-domain electromagnetic full-time response, and finally completes the forward calculation process. Through numerical simulation verification, in the full-time transient electromagnetic forward process, the application can keep small occupied memory and also has higher calculation efficiency, and the calculation precision is not lower than that of other methods.

The invention relates to the technical field of time-domain electromagnetic geological exploration, in particular to a time-domain superconducting electromagnetic multi-effect multi-parameter extraction method and system, and the method comprises the steps: calculating time-domain superconducting electromagnetic induction-polarization-magnetization multi-effect response, and constructing a data error fitting term based on the multi-effect response and the residual error of measured data; constructing a constraint matrix of multi-parameter correlation coefficients including resistivity, polarizability and magnetic susceptibility, and constructing a multi-parameter statistical constraint term for describing statistical correlation among multiple parameters based on the constraint matrix; constructing a regularization objective function; an improved Gaussian-Newton optimization algorithm is adopted, the regularization objective function is minimized, the optimal value of the regularization objective function is iteratively solved, and the resistivity, the polarizability and the magnetic susceptibility are obtained through synchronous inversion. According to the invention, multi-parameter efficient and accurate extraction imaging in the deep mineral exploration and geological structure identification field is realized.

The application discloses a borehole time domain electromagnetic far detection device and method, which comprises a ground system, an armored cable, a remote transmission system and a time domain electromagnetic logging instrument; the remote transmission system is connected with the time domain electromagnetic logging instrument and is arranged in a borehole; the time domain electromagnetic logging instrument comprises an electronic circuit, an acquisition module and a transmission module which are sequentially arranged from top to bottom; the electronic circuit is connected with the acquisition module and the transmission module for detection; the acquisition module comprises a first array type three-component magnetic receiving sensor, a second array type three-component magnetic receiving sensor, an electric receiving sensor, a third array type three-component magnetic receiving sensor and a fourth array type three-component magnetic receiving sensor which are sequentially arranged from top to bottom; and the transmission module comprises a magnetic source and an electric source which are sequentially arranged from top to bottom. The time domain magnetic and electric hybrid transmission source is adopted for transmission, the array type magnetic sensor is adopted for reception and the electric sensor is adopted for reception, so that the source distance of the detector is shortened, and the borehole far detection is realized.

The present application relates to a kind of system response correction methods of time domain electromagnetic detection system.The present application method is used to correct the influence of system response on effective signal, removes the influence of inherent characteristics of receiving sensor on effective signal by training adaptive output compensation of the theoretical and actual response of correction device;And by the real measured current waveform and adaptive current sensor compensation, the real transmitting current waveform is obtained, and the response of earth under unit step excitation is obtained according to its inverse function.The present application can remove the influence of system response on time domain electromagnetic detection system signal, improve signal quality, reduce the blind area time of system detection, improve the detection range of system.

The present application relates to the technical field of geophysical electromagnetic data exploration, in particular to a three-dimensional time-frequency domain electromagnetic joint inversion method based on finite element algorithm. The method obtains time domain and frequency domain electromagnetic observation data of the region to be measured respectively; the region to be measured is profiled to obtain a non-structural tetrahedral mesh to construct a time-frequency domain inversion initial model; the time-frequency domain inversion initial model is forward calculated to obtain time domain and frequency domain electromagnetic response data; time domain electromagnetic data residuals and frequency domain electromagnetic data residuals are obtained, and their corresponding adaptive weights are obtained respectively to establish a joint inversion objective function; the joint inversion objective function is iteratively solved to calculate the update vector of the inversion initial model; the inversion initial model is iteratively updated until the set condition is reached, and the time-frequency domain electromagnetic joint inversion is completed. The present application effectively improves the accuracy of three-dimensional inversion through joint inversion, thereby enhancing the practical effect of three-dimensional inversion.

The application relates to a time-domain electromagnetic space-time double-gradient resistivity imaging method, device, computer equipment, computer readable storage medium and computer program product. The method comprises the following steps: acquiring transient electromagnetic method data to be processed, wherein the data to be processed comprises electric field component attenuation data of m*n measuring points, and a far zone condition is met between the m*n measuring points and a long wire source; acquiring a spatial gradient expression of apparent resistivity and a time gradient expression of apparent resistivity; performing finite difference fusion based on the spatial gradient expression of resistivity and the time gradient expression of apparent resistivity to obtain a gradient resistivity expression; and generating a resistivity distribution map of a subsurface medium according to the data to be processed and the gradient resistivity expression. The abnormal body boundary description accuracy is improved.

The present disclosure belongs to the technical field of geophysical exploration signal processing, and particularly relates to a time-domain electromagnetic polarization parameter extraction method based on an improved physical guided neural network, comprising generating a sample set; constructing a loss function constrained by physical information; using the sample set, minimizing the loss function, and using an improved whale optimization algorithm to optimize the hyperparameters of the neural network to obtain a time-domain electromagnetic polarization parameter neural network inversion model; wherein the improved whale optimization algorithm introduces chaotic mapping to generate an initial population in the initialization stage, introduces a dynamic opposite learning strategy in the bubble net attack stage, and introduces an adaptive t-distribution disturbance variation in the search prey stage; obtaining measured data and preprocessing the same, using the time-domain electromagnetic polarization parameter neural network inversion model to process the preprocessed measured data, and extracting polarization medium model parameters. The method improves the accuracy and robustness of polarization parameter interpretation.

This invention discloses a rapid detection method for shallow ore bodies based on the fusion of time-domain electromagnetic method and AI, comprising: acquiring time-domain electromagnetic induction data of each unit within the detection area and extracting voltage decay curves to obtain preliminary ore body electromagnetic response characteristics; analyzing electromagnetic signal abrupt change points between adjacent units based on the preliminary ore body electromagnetic response characteristics to determine ore body boundary location information; processing the ore body boundary location information using a clustering algorithm to generate a three-dimensional geological structure model; obtaining clear electromagnetic response signals based on the three-dimensional geological structure model; identifying aliased electromagnetic response components in overlapping areas based on the clear electromagnetic response signals to determine existing ore body structures; when multiple ore body structures are determined to exist, acquiring response components and decomposing them into independent electromagnetic signal components using a signal separation algorithm to obtain time constant difference characteristics of each ore body; constructing location modeling relationships based on the time constant difference characteristics and generating a spatial distribution map of overlapping ore bodies.

The invention discloses a machine learning-based time domain electromagnetic transmitter performance evaluation method, which comprises the following steps of: extracting key parameter data from an operation log of an electromagnetic transmitter, and preprocessing the key parameter data to obtain a standardized data set; constructing a multi-dimensional feature set according to the standardized data set, and dividing the multi-dimensional feature set into an input feature vector and a target output variable; training a regression prediction model according to the multi-dimensional feature set to obtain a trained performance prediction model; and predicting the to-be-evaluated data according to the performance prediction model to obtain a prediction result, and performing feature importance analysis on the prediction result to obtain an analysis result. According to the method, intelligent evaluation of transmitter performance under the conditions of multiple devices and multiple working conditions can be achieved, the limitation of a traditional experience or single-index method is overcome, the precision and interpretability of performance analysis are improved, and a new data driving scheme is provided for instrument operation optimization and abnormality diagnosis.

The application discloses a GPU acceleration-based large-loop-source transient electromagnetic fast forward method, which comprises the following steps: firstly, parameter setting and pretreatment are performed; then, the large-loop-source is divided into four edges of north, south, east and west, and each edge is discretized into N electric dipoles by using a Gauss-Legendre quadrature method; a four-level parallel computing architecture of loop edge-frequency sampling point-wave number-time channel is constructed to calculate each edge; wherein, the parallel computing architecture of frequency-wave number-time is used in each edge in the GPU to perform parallel calculation on each electric dipole, so that the frequency-time domain response of each electric dipole is obtained; finally, after the calculation of all electric dipoles is completed, the vectorization calculation and summation of the transient responses of all electric dipoles are performed on the GPU to obtain the total field response of the large-loop-source at the current observation point, and the final time domain electromagnetic field value is output. The above method can efficiently and fully exert the large-scale parallel computing advantage of the GPU, so that the calculation efficiency is effectively improved under the premise of ensuring the calculation precision.

The invention discloses a shallow ore body rapid detection method based on fusion of a time domain electromagnetic method and AI, and the method comprises the steps: obtaining time domain electromagnetic induction data of each unit in a detection region, and extracting a voltage attenuation curve to obtain initial ore body electromagnetic response features; analyzing electromagnetic signal mutation points between adjacent units based on the preliminary ore body electromagnetic response characteristics, and determining ore body boundary position information; processing the ore body boundary position information by adopting a clustering algorithm to generate a three-dimensional geologic structure model; obtaining a clear electromagnetic response signal based on the three-dimensional geological structure model; on the basis of the clear electromagnetic response signals, aliasing electromagnetic response components in the overlapping area are identified, and an existing ore body structure is determined; when it is judged that multiple ore body structures exist, response components are obtained and decomposed into independent electromagnetic signal components through a signal separation algorithm, and time constant difference characteristics of all ore bodies are obtained; and constructing a position modeling relation based on the time constant difference characteristics, and generating an overlapped ore body spatial distribution diagram.

The application belongs to the technical field of petroleum physical exploration, and particularly discloses a wellbore electric source time-domain electromagnetic measurement device and method. The device comprises a ground collection system, a cable, a remote transmission short section, an array electric source receiving sensor and a transmitting device. The ground collection system is connected with the remote transmission short section through the cable. The remote transmission short section is connected with the array electric source receiving sensor. The array electric source receiving sensor is connected with the transmitting device. The measurement mode of electric source excitation and array electric source receiving is adopted, the manufacturing process difficulty of three-dimensional magnetic source in the limited space of the wellbore is reduced, the problem of difficult quick shutdown of the multi-turn coil is solved, and the ultra-long detection of the radial geological information beside the well is realized.

This invention discloses a time-domain electromagnetic data removal method based on continuous wavelet transform, belonging to the field of time-domain electromagnetic noise processing technology. The method includes: acquiring accelerated transient electromagnetic signals; processing and superimposing averaging methods for accelerated transient electromagnetic signals; continuously performing time-frequency transform on the superimposed "total field" signal to obtain a wavelet coefficient matrix; performing dual filtering on the wavelet coefficients, including frequency filtering and amplitude filtering; performing continuous inverse time-frequency transform on the filtered wavelet coefficients to reconstruct the time-domain signal; applying amplitude enhancement processing to the reconstructed signal; and performing Gaussian smoothing processing on the enhanced time-domain signal to facilitate feature recognition. This invention is based on GPU-accelerated continuous wavelet transform + fast convolution. Wavelet transform has multi-scale analysis capabilities, realizing multi-scale decomposition of the signal in the time-frequency domain. Combined with band filtering and amplitude filtering, it performs targeted noise suppression, accurately suppressing specific frequency noise such as power frequency interference and harmonics.

The invention relates to the technical field of geological exploration, and particularly discloses a three-dimensional exploration method based on a composite field source, which comprises the following steps: S1, composite field source arrangement: arranging a plurality of different types of field sources in an observation system; s2, multi-field-source synchronous data acquisition: on the basis of acquisition by adopting a time domain electromagnetic method, performing supplementary measurement by using a frequency domain electromagnetic method; s3, data processing and joint inversion are carried out, data of different field sources are inversed respectively, a plurality of geophysical models which are possibly contradictory to each other can be generated, and a unified underground electrical structure model is searched through joint inversion; s4, performing result interpretation and verification, including comprehensive section compilation, three-dimensional geological modeling and drilling verification and correction; and S5, a final output result is obtained. The problems that a traditional electrical exploration excitation source is poor in adaptability and insufficient in detection performance, and complex requirements are difficult to meet are solved.

The invention discloses a transformer fault prediction method based on frequency domain and time domain combined feature extraction, and belongs to the technical field of power equipment fault diagnosis. The method comprises the following steps: firstly, synchronously acquiring a sweep frequency response curve and transient voltage and current recording data of a transformer; extracting frequency domain structure sensitivity features and time domain electromagnetic sensitivity features in parallel; fusing the different-source features to construct a high-dimensional joint feature vector, and performing dimensionality reduction by adopting principal component analysis; and finally, inputting the fusion feature vector into a pre-trained random forest machine learning model to realize joint recognition, severity evaluation and location positioning of the fault. Through complementation of frequency domain and time domain features, the problems that an existing method depends on a single signal type, is not sensitive to early faults and is difficult to identify multiple fault types at the same time are solved, and the accuracy rate and the early warning capacity of transformer fault diagnosis are remarkably improved.