125 results about "Hyperbolic tangent function" patented technology

An adaptive variable step-size LMS filter based on a hyperbolic tangent function and its filtering method and computer equipment belong to the field of filter technology, and solve the problem that the LMS algorithm improved based on a nonlinear function requires too many manual adjustments of parameters and that the convergence speed and convergence accuracy of the traditional LMS algorithm cannot be satisfied simultaneously. The method of the present invention includes: determining the filter tap length and the number of iterations; initializing the filter parameters; obtaining the filter input signal vector; determining the instantaneous expected signal of the filter; obtaining the instantaneous output signal; obtaining the instantaneous error signal of the filter and the accumulated amount of the error signal based on the instantaneous output signal; obtaining the instantaneous step size using a step-size update formula based on a hyperbolic tangent function; and continuously iterating the filter tap weight coefficient vector to construct an LMS filter using the filter tap weight coefficient vector obtained after the number of iterations. The present invention is suitable for improving the performance of the LMS filter, and the filter of the present invention can be widely used in various signal processing fields.

The invention relates to a mechanical arm trajectory tracking method based on improved integral terminal sliding mode control, and belongs to the technical field of robot control. According to the method, an integral sliding mode surface and a fast terminal sliding mode surface are combined to create a non-singular fast integral terminal sliding mode surface; an improved index reaching law is designed, and the buffeting problem existing in sliding mode control is weakened. The method comprises the following steps: S1, based on physical parameters of the mechanical arm, establishing a mechanical arm dynamic model by adopting a DH parameter method and a Lagrange dynamics principle; s2, constructing a non-singular fast integral terminal sliding mode surface for the dynamic system; s3, designing an improved reaching law fused with a hyperbolic tangent function; s4, performing trajectory planning by using a Cartesian space interpolation algorithm; s5, integrating the sliding mode surface, the improved reaching law and the dynamical model to derive a trajectory tracking control law; s6, constructing a closed-loop control model of the improved integral terminal sliding mode controller and the mechanical arm system; and S7, the control parameters are adjusted, and mechanical arm trajectory tracking is conducted. The control method serves as a controller to conduct trajectory tracking application on the six-degree-of-freedom mechanical arm. The result shows that on the premise of ensuring finite time convergence, the control method effectively eliminates the buffeting phenomenon of traditional sliding mode control, improves the trajectory tracking precision, has high robustness to external disturbance, and can realize high-precision tracking control of a mechanical arm end effector on a complex trajectory.

The invention discloses a single-bit image recovery method and device, a terminal and a medium, and relates to the field of image processing. The method comprises the following steps: performing single-bit quantization on a to-be-recovered image by adopting a non-zero quantization threshold, and determining an initial image quantization model; replacing a sign function in the initial image quantization model with a smooth hyperbolic tangent function, performing low-rank matrix decomposition on the to-be-recovered image, and determining a target image quantization model; determining an initial optimization model according to the truncated least square function and the target image quantization model; a semi-quadratic minimization method is adopted to convert the initial optimization model, and a target optimization model is determined; solving the target optimization model by adopting a near-end block coordinate descent algorithm to obtain each low-dimension sub-matrix and a noise matrix; and determining a reconstructed image by using each low-dimensional sub-matrix. The problem that in the prior art, quantization noise cannot be effectively restrained, and therefore image restoration precision is affected is effectively solved.

The invention relates to the technical field of intelligent control, in particular to an internet-of-things-based old-age care robot intelligent control system, which comprises a multi-mode sensing layer, a safety decision center, an execution driving module, a voice interaction optimization module and an internet-of-things communication module. According to the method, force sense data such as contact force of the mechanical arm, voice data such as voices of old people and view data such as obstacles are collected in real time through a multi-mode sensing layer, a displacement force coupling risk value is calculated through a hyperbolic tangent function coupling model of a safety decision center, and a deviation angle risk value is analyzed and calculated through an improved Lyapunov index spectrum; a total risk value is generated through fusion and is scheduled according to hierarchical intervals, mechanical operation, old people behaviors and environmental risks can be dynamically avoided, and the use safety is remarkably improved; through sub-ambiguity quantitative analysis, comprehensive ambiguity evaluation and intention correction of the voice interaction optimization module, instruction intention is optimized in combination with multi-modal data, the problem of fuzzy voice recognition of old people is effectively solved, and interaction accuracy and response efficiency are greatly improved.

The invention discloses a hyperbolic continuous robust control method for a vertical system of a special vehicle launching tube, which is based on the characteristic of external interference suppression in traditional sliding mode control, combines the smooth characteristic of a hyperbolic tangent function, and achieves the robust control of the vertical system of the launching tube under the condition that strong uncertainty nonlinearity and strong external interference exist in the system. The system flutter phenomenon can be prevented while nonlinearity is compensated and interference is suppressed, the control precision is effectively improved, the error convergence time is shortened, the tracking performance is improved, and finally an exponential stable error convergence result is obtained. Compared with the prior art, the control method disclosed by the invention effectively solves the problems caused by strong nonlinearity and strong external interference of the vertical stabilization system of the special vehicle launching tube, so that the control precision of the whole system is improved, and good tracking performance is obtained.

The invention relates to the technical field of semiconductor device measurement, and provides a semiconductor nonlinear output junction capacitance measurement method based on fitting iteration, and the method comprises the steps: firstly building a junction capacitance extraction circuit, collecting the turn-off transient waveform of a to-be-measured device through a test platform, extracting the oscillation frequency fexp through fast Fourier transform, and carrying out the measurement of the to-be-measured device; based on the relation between the oscillation frequency and the junction capacitance and parasitic parameters, the initial junction capacitance Coss is reversely deduced; then, fitting is carried out by adopting a hyperbolic tangent function, fitting parameters are optimized through a least square method, a preliminary junction capacitance Coss-test voltage udc relation curve under different voltage grades is obtained, an oscillation frequency simulation circuit is further built, simulation frequency fm is obtained, fm is compared with fexp, if deviation exists, the fitting curve is adjusted and iterative optimization is carried out, and the test voltage udc relation curve of the junction capacitance under different voltage grades is obtained. And ending until the error of two adjacent rounds of fitting results is lower than a set error threshold value, thereby obtaining a final accurate junction capacitance Coss-test voltage udc relation curve.

The invention relates to the technical field of image processing, and discloses a three-dimensional medical image segmentation method, which comprises the following steps: acquiring two prediction segmentation probability graphs of each sample based on two parallel sub-networks of a dual-network segmentation model, and calculating a soft pseudo label of an unlabeled sample; utilizing a distance regression head and a hyperbolic tangent function to obtain two prediction symbol distance fields of each sample based on the decoding feature map; calculating segmentation loss and regression loss of each labeled sample, and adding the segmentation loss and the regression loss to obtain supervision loss; calculating the pseudo label consistency loss and the symbol distance consistency loss of each unlabeled sample, and adding the pseudo label consistency loss and the symbol distance consistency loss to obtain consistency loss; optimizing the dual-network segmentation model under the joint constraint of supervision loss and consistency loss; and obtaining an input segmentation result of the to-be-segmented three-dimensional medical image by using the trained dual-network segmentation model. According to the method, the problem of class imbalance in the three-dimensional medical image is effectively relieved, and the segmentation precision of the weak edge region is improved.

The invention provides a hot rolled steel strip surface defect detection method based on target detection, and aims to solve the problems of difficulty in multi-scale small target recognition, insufficient feature fusion and complex background interference in hot rolled steel strip surface defect detection. Extracting multi-scale defect features by using wavelet convolution through decomposition, convolution and reconstruction processes; multi-scale features are adaptively fused based on channels and space attention through a dynamic feature fusion module (DFF); a dynamic hyperbolic tangent function (DyT) is used to replace a normalization layer to dynamically adjust the input range and compress extreme values to reduce the amount of calculation. The method has the effects of high detection average precision (mAP), low missing report rate and remarkably reduced calculation complexity, effectively improves the efficiency and reliability of identifying the surface defects of the hot rolled steel strip in a complex industrial environment, and is suitable for intelligent steel manufacturing quality control.

The invention discloses a mechanical arm self-adaptive sliding mode control method based on a disturbance observer, and relates to the technical field of mechanical arm control, and the mechanical arm self-adaptive sliding mode control method comprises the following steps: constructing an initial dynamic model based on a Lagrange equation; constructing a simplified two-degree-of-freedom rigid mechanical arm dynamic model based on the initial dynamic model; designing a novel reaching law fusing the advantages of an exponential reaching law and a power reaching law, and combining a hyperbolic tangent function to replace a sign function to construct a sliding mode surface; designing a nonlinear disturbance observer, estimating the total disturbance of the system in real time, carrying out feed-forward compensation, and bringing the composite disturbance influence into a control law; designing a control law and an adaptive law of the mechanical arm through a Lyapun function; the stability of the system is analyzed through a Lyapunov function, tracking errors are proved to be asymptotically converged, and the method is used for ensuring the reliability and stability of the control system. According to the method, the disturbance observer and the self-adaptive sliding mode control strategy are combined, and the control precision and robustness of the mechanical arm when external disturbance and model uncertainty exist are effectively improved.

The invention provides a multi-unmanned aerial vehicle optimal game limited accelerated reinforcement learning control method and device, and relates to the technical field of accelerated reinforcement learning. The method comprises the following steps: constructing an evaluation neural network, and approaching a performance index function of a hyperbolic tangent function and a game neighbor item, the optimal limited control input of the unmanned aerial vehicle and the limited control input of the unmanned aerial vehicle under the worst condition; based on the input parameters, constructing a Hamilton-Jacobian error equation; through an error equation, constructing a summation square error which contains current information and past information and is provided with adjusting parameters; designing a weight updating law for evaluating the neural network; calculating the weight of the next iteration according to a weight updating law; and solving an error between two adjacent iteration weights, comparing the error with a preset threshold value, if the error is smaller than the preset threshold value, stopping iteration, and outputting a solution approaching the optimal game consistency control problem of the distributed unmanned aerial vehicle system. According to the invention, the cooperation efficiency between the unmanned aerial vehicles can be improved.

The invention provides a cooperative anti-swing control method and system for double marine cranes based on preset performance, and relates to the technical field of crane control. Comprising the following steps: establishing a marine crane dynamical model considering double-ship rolling disturbance by using a Lagrange method; determining a drivable variable, and establishing a tracking error of the drivable variable; introducing an exponential preset performance function, constraining the convergence rate, overshoot and steady-state value of the tracking error, and converting the constrained tracking error into an unconstrained variable by using a hyperbolic tangent function; a performance index function considering tracking errors and energy consumption is constructed, a Hamiltonian equation minimization performance index function is constructed, the Hamiltonian equation is solved through a reinforcement learning algorithm, and the optimal control rate is obtained. According to the invention, the convergence speed of the error of the system variable is limited in a preset interval, the steady-state and transient-state performance of error convergence is improved, and accurate control and anti-swing of the load are realized.

The invention discloses a non-autoregressive sign language generation method based on KAN enhancement, a medium and equipment, and belongs to the technical field of computer vision technologies. The invention aims to solve the problems of insufficient action details and poor action time sequence coherence of the generated sign language video caused by continuous propagation of local errors along with the increase of the sequence length when the sign language long sequence is generated in the existing method for generating the sign language video based on the autoregression architecture. According to the method, a Transform model of a non-autoregression mechanism is used, a space-time self-attention mechanism is used for replacing a multi-head attention mechanism, a KAN network is used for replacing an MLP layer, a dynamic hyperbolic tangent function is used for replacing layer normalization of a traditional encoder and a traditional decoder, and sign language skeleton sequences are generated in parallel; and performing kinematics optimization and frame-by-frame rendering on the generated sign language skeleton sequence to obtain a sign language video with sufficient detail display and coherent actions. The method is mainly used for generating sign language videos from text data.

The invention discloses a video coding method based on multi-domain perception feature fusion, and belongs to the technical field of image communication. The method comprises the following steps: extracting perception features of a space domain, a time domain and a frequency domain from an input video frame sequence; normalizing the multi-domain sensing features, and generating a sensing importance factor for each coding unit through a dynamic fusion weight mechanism; constructing perceptual distortion evaluation based on perceptual importance factors, replacing a traditional rate distortion objective function, and guiding code rate distribution of coding units; and according to the perception importance factor, performing nonlinear mapping by using a hyperbolic tangent function, calculating to obtain a coding unit level quantization parameter adjustment amount, and obtaining a final quantization parameter in combination with the basic quantization parameter, thereby realizing adaptive rate distortion optimization. According to the method, the visual sensitive area can be accurately identified, and the code rate is remarkably reduced while the subjective visual quality is guaranteed.

The invention provides a design and application method of a sliding-mode deadbeat reaching law, and relates to the technical field of electrical engineering, and the method comprises the steps: designing a sliding-mode deadbeat reaching law SMDRL, designing a control strategy through a deadbeat control principle, defining a state change law on a sliding-mode surface through a sign function, and achieving the design of the sliding-mode deadbeat reaching law SMDRL. It is ensured that the system state converges to the sliding mode surface at the fastest speed in a control period; an improved sliding mode deadbeat reaching law ISMDRL is designed, a traditional sign function is replaced with a hyperbolic tangent function, and parameters of the hyperbolic tangent function are adjusted according to the system state; the SMDRL and the ISMDRL are applied to a speed driving system of a permanent magnet synchronous motor (PMSM), and the speed driving system comprises a PMSM mathematical model, a sliding mode speed controller with a speed disturbance observer and model prediction current control. The SMDRL is introduced from the traditional sliding mode reaching law for the first time through theoretical derivation, the convergence time of the SMDRL is the shortest, and the SMDRL is applied to a permanent magnet synchronous motor speed driving system.

The invention discloses a mathematical modeling AI prediction system in financial risk assessment, which relates to the field of financial data processing and comprises a feature principal component extraction module, a compensation alignment execution module, a momentum regulation and control decision module, a gradient direction optimization module and a risk assessment output module. According to the method, principal components are extracted through covariance matrix decomposition of the qualification-debt ratio and the cash flow fluctuation ratio, feature redundancy is eliminated, the core risk characterization efficiency is improved, a cross-market compensation matrix of transaction frequency and price fluctuation is constructed in combination with SVD and cosine similarity, and the model environment adaptability is enhanced. And analyzing momentum attenuation by adopting a dynamic sliding window of a second derivative matrix eigenvalue ratio, and establishing a regulation and control mechanism to suppress gradient oscillation. Orthogonal interference of historical momentum vectors is eliminated through Schmidt orthogonalization projection, and the accuracy of the gradient optimization direction is guaranteed. And converting abstract numerical values into interpretable risk level classification by using a hyperbolic tangent function in combination with credit rating interval mapping.

The application discloses an infrared image segmentation method based on a local region self-organizing mapping algorithm, which comprises the following steps: firstly, a local region self-organizing mapping algorithm is designed, a plurality of feature bias fields are extracted by moving a local sliding window before iterative evolution of a level set, and a plurality of feature local data driving terms are constructed under a multiplicative bias field model framework; secondly, a plurality of feature global data driving terms are calculated, and a plurality of feature hybrid data driving terms are constructed through an adaptive weight function; then, an adaptive regularization function is used to regularize an energy value range of the plurality of feature hybrid data driving terms, and an initial level set is driven to perform iterative evolution; then, in the iterative process of the level set, a hyperbolic tangent function is used to keep the level set in the iterative evolution, and a symbol rule feature of positive outside and negative inside and a distance rule feature with a modulus value of 1 are kept; meanwhile, a mean filter template is used to continuously smooth the level set and eliminate redundant non-boundary contour lines; finally, a gradient descent method is used to continuously iteratively solve a minimum value of an energy function until a convergence criterion of the level set is reached or a maximum iteration number is reached, and then the method is stopped, at this time, a position of a zero level set is output, and image segmentation is completed. The local region self-organizing mapping algorithm can accurately mine a foreground contour from an infrared image, and has good segmentation speed and precision.

The present invention discloses a single-bit image restoration method, device, terminal, and medium, relating to the field of image processing. The method employs a non-zero quantization threshold to single-bit quantize an image to be restored, determining an initial image quantization model; employs a smooth hyperbolic tangent function to replace the sign function in the initial image quantization model, and performs low-rank matrix decomposition on the image to be restored, determining a target image quantization model; determines an initial optimization model based on a truncated least squares function and the target image quantization model; employs a semi-quadratic minimization method to transform the initial optimization model and determine a target optimization model; employs a proximal block coordinate descent algorithm to solve the target optimization model to obtain low-dimensional submatrices and a noise matrix; and utilizes the low-dimensional submatrices to determine a reconstructed image. This method effectively addresses the problem that prior art cannot effectively suppress quantization noise, thereby affecting image restoration accuracy.

The application provides a training method and device of a current density distribution identification model, the method comprising: inputting a plurality of sampling source points into the current density distribution identification model, and obtaining a loss function in the current density distribution identification model, which is constructed based on a current density distribution function and an electric field integral equation; wherein the loss function comprises a hyperbolic tangent function, a derivative of the hyperbolic tangent function and a product of the Green function; performing series expansion on the hyperbolic tangent function in the loss function, and extracting singular terms from the electric field integral equation in the series expanded loss function to obtain a plurality of singular integrals; performing inner integral calculation and outer integral calculation on each high-order singular integral in the singular integrals to determine the current density distribution function corresponding to the minimum value of the loss function on all sampling source points. The application can quickly solve the singular integrals comprising the hyperbolic tangent function, the derivative of the hyperbolic tangent function and the product of the Green function to obtain the current density distribution function.

The invention discloses an industrial manipulator anti-saturation tracking control method based on an extended state observer, and belongs to the field of industrial manipulator automatic moving trajectory control, and the method comprises the following steps: based on a dynamic model of an industrial manipulator, carrying out modeling processing on the output saturation characteristic of an execution mechanism by adopting a hyperbolic tangent function, and obtaining an output saturation characteristic of the execution mechanism; obtaining a kinetic model subjected to anti-saturation processing; designing an extended state observer with a semi-global consistent final stability characteristic on the basis of the dynamic model subjected to anti-saturation processing, and outputting an unknown function estimation value on the basis of the extended state observer; and based on an unknown function estimation value, a recursive anti-saturation control signal is designed by adopting a dynamic surface control architecture, and accurate tracking control of the industrial manipulator is realized. The design of the extended state observer not only can estimate the unknown dynamic state in the controlled system in real time, but also saves the calculation overhead caused by training the weight of the neural network, so that the execution and deployment efficiency is relatively high.

This invention provides an improved sliding mode active disturbance rejection control (ADRC) method and system for an automotive shimmy system based on an electromagnetic linear actuator. Addressing the system's strong nonlinear characteristics and unknown external disturbances, a mathematical model of the automotive shimmy system based on the electromagnetic linear actuator is established. A nonlinear tracking differentiator is constructed to process the desired input signal of the wheel shimmy angle, rationally arrange the transient process, and synchronously extract the differential signal of the desired signal. A continuous and smooth Ifal function is constructed based on trigonometric sine and inverse proportional functions, and a novel extended state observer is designed to estimate the system's internal and external disturbances in real time. A super-helical sliding mode control law based on the hyperbolic tangent function is designed to replace the error feedback control law in the traditional ADRC, outputting a control voltage. This invention effectively estimates and compensates for the impact of vehicle state parameter changes, model uncertainties, and external disturbances on system stability, improving dynamic performance and control accuracy, and exhibiting good robustness and anti-interference capabilities.

The invention discloses an actuator capability and control rate explicitly coupled satellite attitude adjustment method, and belongs to the technical field of satellite attitude adjustment and control. The method comprises the following steps: firstly, establishing a satellite attitude kinematics and dynamics system, modeling an actuator dynamics system by adopting a hyperbolic tangent function, and widening an attitude control system with limited actuator capability into a three-order attitude control system; then designing an attitude anti-saturation regulation controller, taking the maximum control capability of an actuator as a parameter selection reference, and ensuring the asymptotic stability of the system through the Lyapunov theory; and finally, solving an inequality differential equation of the energy function, and establishing an explicit mathematical relationship between the actuator capability and the control rate. According to the method, the relevance between the actuator capacity and the control rate is revealed through an accurate mathematical equation, a theoretical basis is provided for satellite actuator type selection, and meanwhile, the attitude adjustment stability under the condition that the actuator capacity is limited is ensured.

This invention relates to the field of semiconductor device measurement technology, and proposes a method for measuring the nonlinear output junction capacitance of semiconductors based on fitting iteration. This method first constructs a junction capacitance extraction circuit, and then acquires the turn-off transient waveform of the device under test through a test platform. Finally, it uses Fast Fourier Transform to extract the oscillation frequency f. exp Based on the relationship between the oscillation frequency, junction capacitance, and parasitic parameters, the initial junction capacitance C is deduced. oss Subsequently, a hyperbolic tangent function was used for fitting, and the fitting parameters were optimized using the least squares method to obtain the preliminary junction capacitance C under different voltage levels. oss - Test voltage u dc Based on the relationship curve, a simulation circuit for the oscillation frequency was further constructed to obtain the simulation frequency f. m and f m with f exp The fitting curve is compared, and if a deviation exists, the fitting curve is adjusted and iteratively optimized until the error between two adjacent fitting results is lower than the set error threshold, thus obtaining the final accurate junction capacitance C. oss - Test voltage u dc Relationship curve.

The invention belongs to the technical field of attitude monitoring, and particularly relates to a power transmission line tower attitude monitoring method and system based on multi-dimensional sensing, and the method comprises the steps: collecting original inclination angle time sequence data and synchronous environment state data of a power transmission tower, and constructing a time-varying environment turbulence energy index through a sliding window mechanism; based on the index, utilizing a hyperbolic tangent function to construct nonlinear adaptive observation impedance; constructing a quasi-static equilibrium evolution model containing adaptive observation impedance, and iteratively updating a truth value settlement vector of the tower in combination with a direction consistency factor; and calculating the comprehensive gradient based on the truth value settlement vector and triggering a multi-stage early warning mechanism. According to the method, elastic deformation interference caused by wind vibration and thermal drift is effectively eliminated through noise quantification of a physical level and a nonlinear intelligent gating mechanism, and accurate extraction and zero false alarm monitoring of the micro plastic settlement trend of the tower under the complex meteorological condition of the goaf are achieved.

The invention discloses an intelligent teaching quality evaluation system based on multi-modal information fusion, and belongs to the technical field of education evaluation. The system comprises a data acquisition module for acquiring environment state, teacher behavior, student behavior and interactive behavior data; the environment analysis module outputs an environment state coefficient through a Sigmoid-index coupling model; the teacher behavior analysis module adopts a Lorentz function to quantify the optimal interval of the speech speed, and generates a teacher coefficient in combination with the questioning gain; the student behavior analysis module utilizes a hyperbolic tangent function to strengthen behavior cooperation and outputs a student coefficient; the interaction behavior analysis module generates an interaction coefficient based on a logarithm-S type function; the teaching quality evaluation module obtains a teacher-student behavior matching degree; the teaching example optimization module dynamically adjusts the number of examples according to the teacher-student behavior matching degree. According to the method, the problems of high subjectivity, single dimension and poor real-time performance of a traditional evaluation method are solved, and objective quantification and dynamic optimization of teaching quality are realized.

The present application relates to the technical field of industrial edge scheduling optimization, and more particularly to an industrial edge collaborative scheduling optimization method and system based on federated digital twinning. The method comprises extracting and compressing a feature vector by using local convolution and principal component analysis; mapping the feature vector into a digital twinning model of an edge node; calculating a scheduling weight score of each edge node based on the digital twinning model; constructing a global model based on a neural network, wherein the global model parameters and the learning rate are initialized based on the scheduling weight score; combining the hyperbolic tangent function and the collaborative trust score to aggregate the updated parameter gradient of each edge node and calculate a task migration score; and performing migration scheduling according to the task migration score.

The embodiment of the invention provides a creep life extrapolation method, equipment and a medium, and the method comprises the steps: obtaining the fracture time of a to-be-tested material at a preset temperature under different stresses, the to-be-tested material being 9-12% Cr martensitic steel; fitting the fracture time of the to-be-tested material at a preset temperature and under different stresses to obtain a fitting relational expression between the fracture time and the stress of the to-be-tested material at the preset temperature, the fitting relational expression being a hyperbolic tangent function; and predicting the fracture time of the to-be-tested material under the preset temperature and the target stress according to the fitting relational expression. The fitting relational expression adopts a hyperbolic tangent function to describe the characteristic that the degradation rate of the strengthening mechanism of the to-be-tested material in the service process is accelerated first and then slowed down, the inherent defect of linear extrapolation in a long-life interval is overcome, and the life prediction deviation is reduced.

This invention provides a method and related apparatus for establishing a nonlinear mathematical model of a hybrid DC transmission system considering discontinuous characteristics. The method includes: representing the dynamic processes of discontinuous components in the main and control loops of the hybrid DC transmission system using a step function; approximating the step function as a continuous function using a linear combination of hyperbolic tangent functions to obtain a continuous expression; obtaining modular nonlinear differential equations for the rectifier, inverter, and AC / DC networks, and replacing the discontinuous dynamic parts involved in the equations with the continuous expression; and interconnecting the modules according to their input-output relationships to obtain a complete nonlinear mathematical model of the hybrid DC transmission system. This invention establishes a nonlinear mathematical model of a hybrid DC transmission system that accurately reflects discontinuous dynamic characteristics by employing the continuous approximation method using step functions and hyperbolic tangent functions, thereby improving the accuracy and practicality of system modeling.

The present invention relates to the field of image processing technology and discloses a three-dimensional medical image segmentation method, comprising obtaining two predicted segmentation probability maps for each sample based on two parallel sub-networks of a dual-network segmentation model, and calculating soft pseudo labels for unlabeled samples; obtaining two predicted signed distance fields for each sample based on a decoded feature map using a distance regression head and a hyperbolic tangent function; calculating the segmentation loss and regression loss for each labeled sample, and adding them to obtain a supervised loss; calculating the pseudo-label consistency loss and signed distance consistency loss for each unlabeled sample, and adding them to obtain a consistent loss; optimizing the dual-network segmentation model under the joint constraints of the supervised loss and the consistent loss; and obtaining the segmentation result of an input three-dimensional medical image to be segmented using the trained dual-network segmentation model. The present invention effectively alleviates the class imbalance problem in three-dimensional medical images and improves the segmentation accuracy of weak edge areas.

The invention relates to the technical field of sea surface modeling, and provides a sea surface surge digital twinning modeling method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a surge parameter based on a surge factor and a hyperbolic tangent function, and determining an independent variable of the hyperbolic tangent function by a real-time wave value and a peak wave value of a target sea spectrum; determining a surge factor function based on the surge parameter and the wave number direction angle of the target sea spectrum; based on the basic direction spread function, the surge factor function and a first regularization function, a target direction spread function is determined, and the first regularization function is used for ensuring that the integral of the target direction spread function based on the wave number direction angle is 1; and constructing a sea surface surge model by taking the target sea spectrum as a bottom spectrum and combining the target direction spread function. According to the method, the sea surface surge model with relatively high trueness can be constructed under the condition of relatively low calculation amount.

The invention discloses an oil and gas pipeline multiphase flow electromagnetic heating temperature field cooperative control method and system, and relates to the technical field of electromagnetic heating and temperature field control. Comprising the following steps: dividing a pipeline system into a master control heating section and a slave heating section, and dynamically calculating a basic regulation interval, an approaching minimum value at a high risk and an approaching maximum value at a low risk on the basis of a slave section temperature detection result. Calculating a first correction coefficient through a main control section heat conduction influence model, optimizing a heat conduction intensity factor and a distance attenuation index in combination with a genetic algorithm, and balancing precision, energy consumption and stability; and calculating a second correction coefficient based on the temperature gradient of the adjacent high-temperature section, the normalized distance and the fluid phase state. And mapping the correction coefficient to an interval through a hyperbolic tangent function, and generating a dynamic regulation interval in combination with a basic time interval. And through multi-objective optimization and nonlinear response design, the temperature field control precision is improved, the adaptive capacity to complex working conditions such as multiphase flow medium phase change and flow velocity fluctuation is enhanced, and safe and stable operation of the pipeline is guaranteed.