195 results about "Non linear functions" patented technology

The invention discloses a multi-dimensional layered current limiting method based on spatio-temporal feature fusion. The method comprises the following steps: acquiring a hardware limit speed, and generating a global quota based on a reinforcement learning model; calculating a variable coefficient, and if the variable coefficient exceeds a preset critical value, introducing a conservative coefficient and generating a weight factor through nonlinear function mapping to suppress the micro network jitter; constructing a priority factor matrix based on the service priority and the service type, and obtaining a service priority factor; correcting the average bandwidth of the node by integrating the multi-dimensional dynamic weight factor and the service priority factor to obtain a suggested speed, inputting a historical performance index into the LSTM model to predict the load of the next period, and if the historical performance index exceeds a critical value, triggering connection migration before executing current limiting to realize active avoidance; otherwise, comparing and selecting the minimum value of the global quota, the suggested speed and the hardware limit speed as the final current-limiting speed. The problems of current limiting strategy lag and wide oscillation can be solved, and the bandwidth utilization rate and the system stability are improved.

The invention provides a heterogeneous multi-agent system control method under a DoS attack based on an event triggering strategy. Robust cooperative control under a complex attack environment is realized through the following steps: S1, performing state description and topological structure modeling on a heterogeneous multi-agent system; s2, carrying out approximation on an unknown nonlinear function in the system; s3, restraining the error dynamic state through a performance function, and ensuring that the system response meets the preset precision; s4, designing a time-varying threshold function, and dynamically adjusting a trigger condition to reduce communication burden; s5, designing an anti-attack controller in combination with event triggering and a Lyapunov theory; s6, designing an adaptive law to carry out online estimation on disturbance; and S7, testing synchronization precision, communication efficiency and anti-attack performance based on a physical platform. According to the method, deep fusion of the event triggering mechanism and the preset performance constraint is realized for the first time, and the self-adaptive dynamic threshold strategy is introduced, so that malicious interference of DoS attacks on a communication link is effectively dealt with, and it is ensured that the heterogeneous nodes can still meet preset performance indexes in an attack scene.

The invention provides a level protection-oriented network security configuration checking method, which comprises the following steps of: constructing a knowledge graph associated with a security control item and a technical implementation mode by relying on a standard document and configuration data, and implementing a control item logic dependency and semantic weight reconstruction model; security configuration of assets is automatically collected and semantically mapped to map nodes, configuration deviation is recognized, a risk influence path template is matched in combination with asset importance and service context, semantic propagation path influence intensity is dynamically calculated, and an accumulated risk influence value of configuration deviation is generated in an aggregation mode; according to the method, risk level judgment is carried out by adopting a piecewise nonlinear function, a structured risk assessment record is output, risk identification and reinforcement suggestion closed loop are realized, and the hierarchical management and control and automatic compliance response capability of security configuration deviation in a network environment is improved.

The invention discloses a heat dissipation system and method for a sealed motor, and belongs to the technical field of sealed motors and transmission devices. The system comprises a motor unit, a multistage gearbox unit, a liquid cooling unit, an air cooling unit, a plurality of groups of sensors and a controller. The liquid cooling unit is integrated in the motor outer cover, and the air cooling unit is installed on the motor radiating ribs. According to the method, the total heating power and the variable target temperature of the system are dynamically calculated by collecting the current, the temperature and the rotating speed of the motor and gear signals of the gear box in real time, the required heat dissipation power is calculated in real time through a nonlinear function based on the temperature difference, and then heat dissipation resources of the air cooling unit and the liquid cooling unit are cooperatively distributed with the optimal energy efficiency ratio. According to the invention, feedforward prediction temperature control of dynamic working conditions is realized, the problems of heat dissipation lag, low energy efficiency and overheating risk of the sealed motor under complex load and multi-gear operation are solved, and the reliability, adaptability and overall energy efficiency of the system are significantly improved.

The invention provides a printed circuit board surface defect intelligent detection method and system, and the method comprises the steps: firstly collecting a batch-level defect classification original confidence sequence and synchronous process state parameters, extracting the mean value, variance, change rate and other dynamic characteristics of confidence, and fusing the dynamic characteristics with process parameters to obtain a data fusion model; inputting a pre-training time sequence model to predict the confidence coefficient trend of the next batch; the classification threshold value of the key defect category is dynamically adjusted through the deviation between the actual confidence coefficient mean value and the predicted confidence coefficient mean value, non-linear function self-adaptive mapping is adopted for the adjustment coefficient, and self-adaptive optimization of the classification model threshold value along with quality fluctuation is achieved; when the production line is in a stable stage, threshold adjustment is automatically frozen. The method improves the accuracy of defect identification and the adaptability of the production line.

The invention provides an IGBT junction temperature prediction model construction method and system, and relates to the technical field of power electronic device thermal management, and the method comprises the steps: obtaining an initial data set representing the transient thermal response relation between an IGBT chip and an NTC temperature measurement point in an IGBT power module where the IGBT chip is located; training a neural network model based on the initial data set, and learning and representing a nonlinear function relationship between time and thermal impedance; inputting a preset new time sequence into the trained neural network model, and generating a time-continuous and noise-suppressed optimized thermal impedance data set; and based on the data set, performing parameter fitting on a preset lumped parameter thermal network model to identify thermal resistance and thermal capacity parameters of the model so as to complete construction of the junction temperature prediction model. According to the method, the neural network is used as a data optimization tool in a physical model parameter identification process, and accurate and convenient online prediction of the IGBT junction temperature is realized only by using the temperature signal of the NTC temperature measurement point of the module.

The invention relates to the technical field of network acceleration, and discloses an edge-end large language model reasoning acceleration method and accelerator, and the method comprises the following steps: reconstructing a calculation process of a decoding stage, and carrying out the deep fusion of a multi-head attention mechanism and the calculation operation of a feedforward network; the weight and key value data are stored in HBM, and the coefficient and the accumulated attention score are stored in DDR; for linear matrix calculation, a unified matrix calculation unit is used for executing multi-precision matrix operation; for nonlinear function calculation, a mathematical transformation and linear fitting method is adopted, a Softmax function is converted into operation with 2 as the bottom through a bottom conversion formula, and truncation and third-order linear fitting are conducted on a Sigmoid function; and constructing a key value screening algorithm based on the accumulated attention score, dynamically adjusting a key value storage position, maintaining a recent key value cache region and an important key value cache region in a limited cache space, and realizing key value efficient cache in long text reasoning.

A data processing method, comprising: receiving a first fragment among two fragments obtained by performing secret fragmentation on input data (S602); dividing an original calculation function for processing the input data into a non-linear function in a symmetric interval and a linear function in an asymmetric interval (S604); obtaining a first representation vector of the first fragment on the basis of a Fourier series expression of the non-linear function, sending to a second computing service provider a first ciphertext obtained by performing homomorphic encryption on the first representation vector, receiving a blinded ciphertext returned by the second computing service provider on the basis of the first ciphertext and a second fragment among the two fragments, and decrypting the blinded ciphertext to obtain a result fragment of the non-linear function corresponding to the first fragment (S606); and, on the basis of the result fragment of the non-linear function corresponding to the first fragment and a result fragment of the linear function corresponding to the first fragment, obtaining a first result fragment, which is used for, together with a second result fragment obtained by the second computing service provider on the basis of the second fragment, generating a computing result corresponding to the input data (S608).

A simulation system and method for implementing a model based on an iterative neural network, the system comprising: a simulation vector-matrix multiplication circuit that encodes a weight matrix of the model based on the iterative neural network; and an analog non-linear circuit that encodes a non-linear function arranged in a feedback loop configured to return an output signal from the non-linear circuit as input to the vector-matrix multiplication circuit, wherein the system is configured to output a solution vector of values of the model based on the iterative neural network upon convergence of the system.

The invention relates to the technical field of battery management systems, and discloses a lithium battery SOC estimation method based on model predictive control and dynamic parameter calibration, and the method comprises the steps: constructing a second-order RC equivalent circuit model of a lithium battery, and obtaining a nonlinear function relation between an open-circuit voltage and a state of charge; real-time current and voltage data are collected, model parameters are identified on line through a recursive least square method with a forgetting factor, and dynamic calibration of the parameters is achieved; constructing a discrete state space prediction model based on the identified parameters; and solving an optimal control sequence through rolling optimization under a constraint condition by utilizing a model prediction control algorithm, and calculating a state-of-charge estimation value of the lithium battery in combination with feedback correction. According to the method, on-line parameter identification and model prediction control are combined, the problem of model mismatch under battery parameter time varying and dynamic working conditions is effectively solved, and the precision and robustness of lithium battery SOC estimation are improved.

The invention relates to the field of reinforcement learning, in particular to a reinforcement learning-based large language model antagonism fine-tuning enhancement system. The state prediction module is used for acquiring system state data and robot action data; based on the robot action data, generating a prediction state vector through a neural network prediction model, and comparing the prediction state vector with the system state data to obtain a residual vector; the action decoding module inputs the residual vector into a feedforward neural network, codes the residual vector to generate a natural language state report, and performs autoregression decoding through a large language model to obtain a loss value; the strategy optimization module converts a penalty function value into a reinforcement learning reward signal through a nonlinear function, and the PPO algorithm calculates the total loss according to the reinforcement learning reward signal; and calculating, optimizing and updating the gradient of the low-rank adaptive parameter in the large language model based on the total loss. According to the invention, through the feedforward neural network and the PPO algorithm, the tiny dynamic deviation is captured, and the predictability and safety of the system are improved.

The application provides a kind of IGBT junction temperature prediction model construction method and system, it is related to power electronic device thermal management technical field, method includes: obtaining the initial data set of the transient thermal response relationship between the IGBT chip and the NTC temperature measuring point in the IGBT power module;Based on the initial data set, the neural network model is trained, and the nonlinear function relationship between time and thermal impedance is learned and characterized;The preset new time series is input into the trained neural network model, and the optimized thermal impedance data set with continuous time and noise suppression is generated;Based on the data set, the lumped parameter thermal network model is parameter fitted, to identify its thermal resistance and heat capacity parameters, complete the construction of junction temperature prediction model.The application uses neural network as a data optimization tool in the parameter identification process of physical model, only uses the temperature signal of NTC temperature measuring point provided by the module, realizes the accurate, convenient online prediction of IGBT junction temperature.

The application discloses an advertisement optimization method, system, device and medium. The method comprises the following steps: obtaining an error value between a current advertisement effect and an expected effect; calculating a nonlinear PID control signal by using a nonlinear PID algorithm according to the error value; obtaining historical data of the effect of the advertisement, and predicting the error by using a long short-term memory network on the historical data to obtain error prediction data; converting the error prediction data by using a nonlinear function to obtain a feedforward compensation signal; and optimizing the advertisement according to the nonlinear PID control signal and the feedforward compensation signal. Compared with the related art, the application optimizes the PID algorithm by introducing a nonlinear function, and improves the response speed of the advertisement optimization system while improving the accuracy of the advertisement optimization system by combining the long short-term memory network (LSTM) for feedforward compensation prediction.

This invention discloses a rapid method for detecting the moisture content of fine aggregates used in concrete, belonging to the field of building materials testing technology. The method involves collecting samples of different types of fine aggregates, preparing multiple moisture content samples from oven-dry to saturated surface-dry states, and then measuring the resistivity values ​​after compaction under constant pressure to form a dataset. Based on this dataset, a specific model is established using nonlinear function fitting, and a general prediction model is established using machine learning algorithms. During on-site testing, the resistivity values ​​of the fine aggregates to be tested are measured after compaction under the same constant pressure, and the moisture content is calculated according to the type of aggregate using the appropriate model. This invention achieves rapid and non-destructive testing of the moisture content of fine aggregates, providing real-time data for adjusting water usage in concrete production.

The application discloses a Helicobacter pylori breath test signal denoising fitting method and system, and belongs to the technical field of medical detection and signal processing. The method processes the continuous monitoring signal of the 13CO2 / 12CO2 isotope ratio in exhalation: firstly, the original signal is subjected to band-pass digital filtering to suppress high-frequency noise; then, low-frequency drift is eliminated through adaptive baseline correction to generate a drift-removed signal; subsequently, a robust peak value recognition algorithm is used to extract physiological response characteristics; finally, the identified peak value is taken as the center, a nonlinear function with physiological constraints is used for local fitting, and the Delta Over Baseline value is calculated based on the generated fitting curve with high precision. The application effectively solves the signal distortion problem caused by breathing fluctuations, instrument noise and the like in in vivo detection, realizes high-fidelity restoration of the Helicobacter pylori metabolic response, and significantly improves the detection accuracy and reliability of the breath test.

The invention discloses a nonlinear multi-agent system double-clock asynchronous hybrid neural adaptive control method, and relates to the technical field of nonlinear multi-agent system cooperative control. According to the method, a double-clock asynchronous framework is provided, time decoupling is carried out on updating of the leader observer and updating of the local observer, the limitation of synchronous updating of all assemblies is broken through, and the leader observer and the local assembly are made to operate independently; an event trigger pulse mechanism is designed, sampling is carried out only when a specific event occurs, the communication and calculation cost is remarkably reduced, the event trigger mechanism allows each agent to autonomously determine a trigger moment, and self-adaptive resource allocation between heterogeneous dynamic agents is achieved; and the numerical value of the nonlinear function is simulated and estimated by using the neural network, so that approximate state information can still be obtained under different conditions. And the weight of the neural network is only updated at the triggering moment, so that the resource consumption in the learning process is reduced.

The application discloses a correction method for saturated spectral line intensity in laser spectroscopy, and relates to the technical field of spectral data analysis, wherein saturated characteristics in a spectrum are identified, a saturated point is determined based on an intensity threshold, and the original intensity of the saturated point is recorded; the wavelength range of a target spectral peak containing the saturated point is determined; the saturated point is removed from the wavelength range of the target spectral peak, and the unsaturated point is reserved; the unsaturated point is fitted through a nonlinear function, and the maximum intensity of the fitting is obtained; if the maximum intensity of the fitting is lower than the original intensity, the intensity is corrected in a linear extension mode based on the unsaturated point, otherwise, the intensity is directly corrected based on the fitting model; the corrected spectral intensity data are output, and the maximum intensity of the spectral peak and the integral area of the spectral peak are calculated. The application improves the accuracy of saturated determination and the reliability of correction results, and provides direct intensity data basis and core derivative parameters for quantitative analysis.

The invention discloses a field weakening trajectory tracking control method for permanent magnet synchronous motors based on enhanced model predictive current control, comprising: S1, establishing a mathematical model of the permanent magnet synchronous motor in a synchronous rotating coordinate system and determining the optimal operating trajectory in the field weakening region; S2, constructing an extended virtual voltage vector set synthesized from the basic voltage vector, and constructing a virtual voltage vector containing continuous components, excess components, and disturbance components as feedback quantities, which are adjusted by the field weakening current controller. d The shaft reference current limits the virtual voltage amplitude within the inverter output voltage limit; S3, a nonlinear extended state observer based on a nonlinear function is designed to estimate the current state and total disturbance in real time, and correct the current prediction model. The optimal voltage vector is selected through a cost function, and a switching pulse sequence is generated to drive the inverter. This invention can effectively reduce current ripple when motor parameters are mismatched and overcome the deviation of the field weakening trajectory when model parameters are mismatched.

The application discloses a nonlinear Fourier reweighting image resolution enhancement method and system without a point spread function. The method comprises the following steps: acquiring an original digital image; performing Fourier transform on the original digital image to separate an amplitude spectrum and a phase spectrum; constructing a nonlinear mapping function with two adjustable parameters, and performing reweighting processing on the amplitude spectrum by using the function to balance high and low frequency energy of the amplitude spectrum and generate a new amplitude spectrum; recombining the new amplitude spectrum and the original phase spectrum; and performing inverse Fourier transform on the combined spectrum to output a resolution enhanced image. The application does not depend on PSF measurement or estimation, only needs one Fourier transform process, and has high calculation efficiency. The core construction is to directly reweight the amplitude in the frequency domain by using a specific nonlinear function, so that the ill-conditioned problem and artifacts of a traditional deconvolution algorithm are avoided. The application can be widely applied to optical fluorescence microscopy, astronomical imaging and clear processing of general digital images.

The application aims to provide a kind of micro-ring resonator-based activation function device manufacturing method, comprising the following steps: determining the standard nonlinear function to be fitted by optical device;Preparation of micro-ring resonator;Coupling ring PN junction and micro-ring;Set up auxiliary light source and output signal measurement equipment;Set up TIA transimpedance amplifier and bias unit.The application can realize the function of activation function by optical device, and it is relatively easy to combine with other optical elements that can realize the summation of neural unit weight, realize the complete function of single neuron, thereby improve the operation efficiency and operation accuracy of optical neural network.The device is small in size and easy to integrate, and the operation efficiency and operation accuracy of the optical neural network integrated by the neural unit integrated by the device can be significantly improved.

This invention discloses a method and system for transformer partial discharge pattern recognition based on an optimized probabilistic neural network. The method first acquires the transformer partial discharge signal; then, it establishes a two-dimensional spectrum of the partial discharge phase distribution pattern based on the partial discharge signal and extracts discharge statistical features from this spectrum; finally, it inputs the discharge statistical features into the optimized probabilistic neural network for pattern recognition to obtain the partial discharge pattern. The optimized probabilistic neural network uses a pollination algorithm to optimize the smoothing factor, and the switching probability in the pollination algorithm is a nonlinear function that decreases with the number of iterations. This invention can more accurately and efficiently classify and identify different partial discharge patterns of transformers, providing a data foundation for transformer fault diagnosis and resolution.

The invention relates to a fault-tolerant control method for preset performance of a nonlinear multi-agent system, and the method comprises the following steps: S1, building a high-order nonlinear multi-agent system model in consideration of a multiplicative fault and an additive time-varying fault of an actuator; s2, aiming at an unknown nonlinear function and a fault item in the high-order nonlinear multi-agent system model, constructing an online identifier by utilizing an RBF neural network, reconstructing a system state and estimating a fault value; s3, designing a sliding mode surface to reduce the system order based on the state information provided by the online identifier; s4, on the basis of the converted error system, defining a performance index function, deducing an HJB equation and a theoretical optimal control law, and converting fault-tolerant control into an optimal adjustment problem; and S5, solving an optimal adjustment problem by adopting Actor-Critic reinforcement learning, obtaining optimal control, and outputting a final control quantity to act on the system. According to the invention, multi-agent adaptive optimal fault-tolerant control is realized.

The application discloses a method for adaptive secure consensus control of a multi-agent spatio-temporal dynamic system with unknown boundary nonlinearity under hybrid attacks, comprising the following steps: constructing a multi-agent spatio-temporal dynamic system model with unknown boundary nonlinearity and a virtual leader model; establishing a hybrid network attack model containing deception attacks and denial of service (Dos) attacks, and using an adaptive radial basis neural network to approximate the unknown boundary nonlinearity function; defining a consensus error signal to obtain a consensus state error system; designing a composite adaptive neural network secure boundary consensus control scheme, constructing a Lyapunov function for the error system, and obtaining a sufficient condition for the error system to realize mean square secure consensus control. The application effectively solves the problem that the traditional method is difficult to simultaneously cope with complex network attacks and unknown boundary nonlinear disturbances, and significantly improves the security and robustness of the system.

The application discloses an adaptive tracking control method for a random nonlinear system based on output feedback, comprising the following steps of S1, constructing a mathematical model of a random nonlinear system with hysteresis input, unknown control direction, arbitrary switching and output constraint; S2, designing a linear state observer, and reconstructing unmeasurable states of the random nonlinear system by using output and control input of the random nonlinear system; S3, introducing a Nussbaum type function to process the unknown control direction, and constructing a barrier Lyapunov function to ensure that the output constraint is not violated; and S4, designing a framework based on a backstepping method, combining a radial basis neural network to approximate unknown nonlinear functions, designing a virtual control law, an adaptive law and an actual control law, and realizing adaptive tracking control of the output of the random nonlinear system on a reference signal.

The present application relates to the field of remote sensing monitoring, and discloses a kind of soil humidity multi-source remote sensing monitoring method, equipment and storage medium, method includes steps: obtaining multi-source satellite remote sensing data;Multi-source satellite remote sensing data is preprocessed, and the data after preprocessing is obtained;According to the data after preprocessing, data set is constructed;Three-dimensional joint convolutional neural network is constructed, and three-dimensional joint convolutional neural network is trained based on data set, and the network after training is completed is obtained;Soil humidity is predicted using the network after training is completed.The beneficial effects of the present application are: effectively solve the problem of multi-source remote sensing data fusion and information complementary deficiency in the prior art, realize the information complementary and feature fusion of multi-source remote sensing data, enhance the robustness of the model, improve the prediction accuracy and stability;It has the characteristics of simple and effective, without the need for research and analysis of complex remote sensing mechanism, more accurately approximates the complex nonlinear function between remote sensing information and ground object parameters.

The invention relates to the field of big data processing, and particularly discloses a fault tree model-based immersed transformer fault analysis method, which solves the problems of static property and model simplification in a traditional analysis method by constructing a dynamic probability updating engine connected with real-time monitoring data and a fault tree model. Specifically, the method comprises the following steps: standardizing multi-source heterogeneous field data into a unified degradation degree vector; then, through a three-channel parallel computing model, degradation influence is intelligently decoupled and quantified into three effects of linear accumulation, nonlinear cooperative enhancement and mutability dominance; according to the model, the three paths of influence factors are fused through a nonlinear function, and the reference probability of a fault tree basic event is dynamically updated with high fidelity. And finally, assigning the dynamic probability to a fault tree, thereby realizing accurate identification and sorting of the most dangerous fault path in the current state, and fundamentally improving predictability and reliability of fault analysis in complex scenes such as water immersion and the like.

The present application relates to the field of artificial intelligence and control technology, in particular to a nonlinear multi-agent system fixed-time consensus control method, device and medium. The control method takes the backstepping recursive method as the control design framework, proposes an adaptive fixed-time consensus controller based on an auxiliary compensation system, and solves the nonlinear multi-agent system consensus control problem under time-varying input delay. By providing and changing the control input signal by a person, the multi-agent system motion trajectory can be modified according to the demand. The time-varying input delay and unknown nonlinear function are considered in the system model, making the system more general. Among them, the unknown time-varying delay function is processed by constructing an auxiliary compensation system, and the unknown nonlinear function is processed by using a radial basis function neural network approximation. According to the fixed-time control related lemma, a practical fixed-time adaptive consensus control method is proposed to ensure that the consensus error of the multi-agent system converges to the neighborhood of the origin within a fixed time.

This invention provides a control method for a lower limb exoskeleton rehabilitation robot based on the sliding mode optimization method, belonging to the field of medical rehabilitation robot control technology. The method includes: establishing a two-degree-of-freedom dynamic model of the lower limb exoskeleton rehabilitation robot, and transforming it into a linear state-space model using feedback linearization technology; designing a linear quadratic regulator (LQR) for the linearized model to obtain the optimal feedback gain matrix; constructing an integral sliding surface containing the optimal feedback gain matrix, so that the equivalent control law of the system in the sliding mode possesses the globally optimal dynamic performance of the LQR; and based on adaptive preset time theory, constructing a sliding mode reaching law and an adaptive gain update law containing novel nonlinear functions to generate the final control torque.

The present application relates to a kind of discrete multi-motor servo system fuzzy wavelet neural network control method with event triggering mechanism, belong to multi-motor servo system control field, comprising the following steps: S1: the discrete time system model of multi-motor servo system is established;S2: design two type fuzzy wavelet neural network to estimate unknown nonlinear function caused by external interference and internal parameter perturbation;S3: based on the backstepping control framework, introduce event triggering mechanism with dead zone operator, design discrete time fuzzy wavelet neural network controller;S4: using the discrete time fuzzy wavelet neural network controller, control multi-motor servo system is carried out.