12 results about "Nonlinear saturation" patented technology

The invention relates to the technical field of unmanned vehicle control, and discloses an unmanned vehicle control algorithm with an emergency obstacle avoidance function, which comprises the following steps: acquiring a transverse deviation between a current attitude of a vehicle and an obstacle avoidance target trajectory, and determining an error flux according to a product of a first derivative of the transverse deviation and a curvature change rate of the obstacle avoidance target trajectory; converting the error flux into a gain adjustment factor based on a preset nonlinear mapping function, and performing real-time weighted weakening on a proportional gain and a differential gain in a steering closed-loop control loop, so that a steering actuator is in a controlled under-gain state in an obstacle avoidance cut-in stage; after the error flux falls back to the physical linear threshold value, feedback gain is recovered according to the energy recovery curve, physical matching of the control rigidity and input sudden change is achieved, the nonlinear saturation risk of an actuator is avoided, and the loop stability of a control system under the extreme working condition is guaranteed.

The application belongs to the field of power equipment, and discloses a transformer inter-turn short-circuit fault modeling method, system, device and medium, which comprises the following steps: a two-dimensional geometric model of a transformer core and winding is established, the two-dimensional geometric model is discretized and divided into a plurality of magnetic network units to obtain a discrete magnetic network model; a three-dimensional electrostatic field entity model of the winding is constructed, and a distribution capacitance matrix between turns of the winding and to the ground is calculated; a short-circuit grid point is set in the discrete magnetic network model, the short-circuit grid point is connected in parallel to a metallic short-circuit branch model to obtain a short-circuit closed branch; a transformer circuit is obtained, a coupling relationship between the transformer circuit and the discrete magnetic network model is constructed based on the distribution capacitance matrix, and a transformer inter-turn short-circuit fault model is constructed. The application can accurately describe the internal space electromagnetic distribution characteristics and the local nonlinear saturation characteristics of the core of the transformer.

This invention relates to the fields of digital marketing technology and artificial intelligence optimization technology, and particularly to a method and system for joint budget allocation and ROI optimization of cross-platform advertising expenditures. It addresses the channel contribution assessment bias caused by traditional "last click" attribution through the Shapley value attribution model; it achieves quantitative prediction of inter-channel synergy and crowding-out effects by constructing a joint prediction model using nonlinear saturated response functions and cross-influence coefficients; it overcomes budget inefficiencies caused by independent optimization on a single platform through global optimization solutions using nonlinear programming and sequential quadratic programming algorithms; and it ensures the safe implementation of automated decision-making through safety executors and anomaly circuit breakers. This method elevates advertising budget allocation from a "local, static, and biased" level to a "global, dynamic, and scientific" level, significantly improving the overall advertising ROI by an estimated 15%-30%.

The invention relates to the technical field of process safety and explosion prevention, in particular to a graded explosion suppression-explosion venting control method and system based on dynamic pressure characteristics. The method comprises the steps of information collection and preprocessing, dynamic feature extraction, feature parameter normalization, dynamic weight distribution, comprehensive risk index calculation, execution of a corresponding grading response strategy and degradation and revocation control. Through multi-feature fusion, a dynamic weight, a nonlinear saturation function and a hierarchical trigger mechanism with hysteresis, the explosive risk assessment is fundamentally converted from static and single states to dynamic and comprehensive states, and the prevention and control response spans from isolation and extensive states to collaborative and fine states. Finally, an intelligent explosion-proof system which is high in sensitivity, low in false alarm rate, self-adaptive and reversible is constructed, and the safety protection level of the combustible gas container under various complex working conditions is remarkably improved.

The application discloses a kind of constant temperature controllable load regulation and control method and system based on multi-agent control, comprising: the performance parameter and refrigeration performance index of each constant temperature controllable load in large-scale constant temperature controllable load cluster are collected, and the thermodynamic model corresponding to the room where each constant temperature controllable load is located is constructed, the large-scale constant temperature controllable load cluster is defined as multi-agent system, and control framework and communication model for controlling multi-agent system through aggregator are established, nonlinear consistency controller is constructed according to preset constant temperature controllable load state variable calculation method, nonlinear saturation consistency algorithm, relay node state update model and fallback mechanism, the initial state of the large-scale constant temperature controllable load cluster is obtained, according to the initial state and the thermodynamic model, the nonlinear consistency controller is regulated and controlled, for realizing the unified regulation of large-scale constant temperature controllable load.

This invention discloses a method for identifying the linear growth rate of combustion instability, comprising: establishing a stochastic dynamics model of the combustion system; constructing an identification model based on the stochastic dynamics model of the combustion system; and identifying the linear growth rate, nonlinear saturation coefficient, and noise intensity of the combustion system based on the identification model. This invention aims to solve the problem in traditional methods where the changes in the linear growth rate, nonlinear saturation coefficient, and noise intensity cannot be identified from the data when the units of the experimental data are transformed.

The invention discloses a permanent magnet synchronous fan current loop novel exponential sliding mode control method based on a high-gain disturbance observer, and the method comprises the steps: firstly constructing a terminal sliding mode surface based on an exponential convergence characteristic, designing a corresponding control law according to the sliding mode surface, carrying out the integral continuation processing of a switching function through the control law, so as to weaken a high-frequency switching component, and carrying out the control of a current loop of a permanent magnet synchronous fan; meanwhile, a hyperbolic tangent function is introduced to construct a nonlinear saturation function, the function can effectively suppress the high-frequency buffeting phenomenon of traditional sliding mode control, and meanwhile, due to the fact that the control law avoids differential operation on an exponential function, the singularity problem in traditional terminal sliding mode control is effectively eliminated; a high-gain disturbance observer (HGDO) is introduced, and real-time observation and dynamic feed-forward compensation are performed on system parameter uncertainty and external wind speed disturbance by constructing a disturbance estimation model, so that the anti-interference capability of the system is improved.

This application proposes a method and system for predicting the nonlinear trend of thermoacoustic oscillations based on the flame transfer function. The method includes: real-time acquisition of inlet excitation fluctuation signals and heat release rate fluctuation signals of the combustion chamber; identification of the flame transfer function (FTF) based on the inlet excitation fluctuation signals and the heat release rate fluctuation signals; phase correction of the FTF to obtain the corrected real part of the FTF reflecting the internal dynamics of the flame; determination of the linear growth rate of the combustion system; processing of the heat release rate fluctuation signals to extract the steady-state amplitude information of the heat release rate fluctuation signals and determine the nonlinear saturation intensity of the combustion system; and determining whether the current state of the combustion system tends to develop into a stable limiting-loop thermoacoustic oscillation based on the linear growth rate and the nonlinear saturation intensity, and outputting a prediction conclusion. The technical solution proposed in this application can quickly and accurately predict whether the system will develop into destructive thermoacoustic oscillations based solely on online measurable combustion data.

This invention discloses a deep learning inversion method combining generative adversarial networks (GANs) for high resistivity formations, belonging to the field of logging-while-drilling (LWD) technology in oil and gas exploration. Addressing the problems of insufficient accuracy and poor generalization performance in existing deep learning inversion methods under high resistivity formation conditions, this invention constructs a layered formation model and sample set, utilizes the XGBoost algorithm for feature selection of logging responses, builds a cGAN-MLP joint data augmentation framework, directionally generates and physically verifies high-resistivity formation samples, and expands them into the training set, and employs a BiLSTM-Transformer hybrid deep learning model for inversion, outputting parameters such as formation resistivity and boundary distance. This invention systematically solves the inversion bottleneck caused by the sparsity of high-resistivity formation samples and the nonlinear saturation of electromagnetic responses through feature optimization and data augmentation, significantly improving the inversion accuracy and generalization ability of high-resistivity formations, while meeting the real-time requirements of LWD geological steering, and is suitable for high-precision geological steering in deep and unconventional oil and gas reservoirs.

The embodiment of the application provides an estimation method and system of a wind turbine variable pitch nonlinear saturation system, which introduces a saturation compensation network after a standard EKF prediction step, perceives the saturation depth of an instruction and the initial prediction deviation of a physical model, and further corrects the model prediction error caused by the saturation effect. In this way, the prior state prediction can be actively and accurately compensated before the update step of the EKF, so as to effectively inhibit the impact of large new information caused by saturation on the filtering process, avoid the estimation divergence problem that may occur in the traditional method, and provide reliable state information support for accurate control and safe operation of the wind turbine under complex working conditions.

The invention relates to the technical field of data processing, in particular to an ungrounded neutral point resonance monitoring method and system for a smart power grid, and solves the technical problem of insufficient monitoring accuracy in the prior art. The method comprises the following steps: acquiring multi-source electric quantity time sequence data of a neutral ungrounded system of a smart power grid; determining resonance excitation combined potential energy according to the multi-source electric quantity time sequence data; the resonance excitation combined potential energy is used for representing the ferromagnetic resonance excitation capability of the system after transient impact; determining a resonance risk dynamic evolution coefficient according to the resonance excitation combined potential energy and the nonlinear saturation index of the iron core; determining a system-level resonance dynamic risk coefficient according to the resonance excitation combined potential energy and the resonance risk dynamic evolution coefficient; the system-level resonance dynamic risk coefficient is used for representing the current resonance instability risk level of the system.

The invention provides a robot inverse kinematics adaptive smooth constraint method and system, and relates to the technical field of robot motion control. The method aims at solving the problem that high-speed motion smoothness and low-speed positioning responsiveness cannot be considered at the same time through fixed parameter constraints. The method is executed in an iterative control loop and comprises the steps that the current state and the target state of the robot are obtained, and the pose error is calculated; calculating a theoretical joint update amount based on the pose error; according to real-time motion state parameters such as the die length of the pose error and / or the die length of the current joint angular velocity, a self-adaptive gain factor is dynamically calculated, and the value of the gain factor and the value of the parameter are in a monotone decreasing relation; a preset and continuously differentiable nonlinear saturation function is adopted, and the adaptive gain factor is used to constrain the theoretical joint update quantity to generate a final joint update quantity; and updating the state of the robot by using the final joint updating amount.