58 results about "Differential equation models" patented technology

The invention discloses a method for designing stability parameters of a self-adapting filter of a self-balancing system of a magnetic suspension rotor. The method comprises the following steps of establishing a radial rotation dynamics differential equation model of the self-balancing system of the magnetic suspension rotor with a gyroscopic effect; converting a real coefficient bivariate gyroscopic coupling system into a complex coefficient univariate equivalent system through complex transformation; calculating a closed-loop transfer function of the complex coefficient univariate equivalent system; determining a start angle range of a root locus, which meets the stability requirement of the self-balancing system, at a critical stable point of an imaginary axis; resolving a stability parameter intersection of the self-adapting filter; finally, determining the stability parameters of the self-adapting filter. While the rotor speed is estimated and tracked in real time so as to improve the self-balancing precision, the stability of the self-balancing system of the magnetic suspension rotor with the gyroscopic effect within a full speed range is ensured. The method disclosed by the invention is simpler and more convenient than other methods, and is especially suitable for practical magnetic suspension rotor systems.

The invention discloses a super-resolution image reconstruction method based on a coupled partial differential equation model. According to the reconstruction method, two partial differential models are coupled through defining a weighting function by utilizing the respective advantages of TV (Total Variation) and FPDE (Fourth Partial Differential Equation) in image restoration, a large weight is adopted for a TV model at an image edge area so as to maintain the edge and texture details of images, the large weight is adopted for an FPDE model at an image flatness area so as to inhibit a staircase effect generated by the TV model, new models serve as normalization items to reconstruct a super-resolution images, and the visual effect of image reconstruction can be enhanced.

The invention discloses a method for measuring power frequency parameters of superhigh/extrahigh-voltage alternating-current (direct-current) power transmission circuit. The method is characterized by comprising the steps of synchronously measuring zero-sequence voltage and zero-sequence current on the head end and the tail end of the power transmission circuit, obtaining a power-frequency zero-sequence voltage phasor and a power-frequency zero-sequence current phasor by utilizing a Fourier filtering method, solving the power-frequency zero-sequence parameters of the power transmission circuit on the basis of a differential equation model and particle swarm optimization of the power transmission circuit, considering the influence of electromagnetic coupling between a distributive capacitor and the circuit on the power transmission circuit on the measurement of the power-frequency zero-sequence parameters, so that the precision of the power-frequency zero-sequence parameter measurement result of the power transmission circuit can be improved. The method is not only suitable for measuring the power-frequency zero-sequence parameter of a single-circuit and a multi-circuit superhigh/ultrahigh voltage alternating-current power transmission circuit, particularly suitable for measuring the power-frequency zero-sequence parameters of a double-circuit and four-circuit superhigh/ultrahigh voltage alternating-current power transmission circuit of a same tower and also suitable for remotely measuring the power-frequency parameters of the superhigh/ultrahigh voltage direct-current power transmission circuit.

The invention discloses a hybrid multilevel converter and a variable switching frequency trajectory optimization control method thereof. The hybrid multilevel converter is connected between a direct current power grid and a three-phase alternating current power grid. The variable switching frequency trajectory optimization control method of the hybrid multilevel converter comprises the following steps that S1, differential equation models of the hybrid multilevel converter are established; S2, discretization state equations from the (K+1)th step to the (K+delta)th step are deduced; S3, a feasible sampling point set of output current predicted values is determined; S4, the optimal value f<theta> of an objective function f<theta>(k) is obtained, a switching function vector S<theta,opt>[k] corresponding to a feasible predicted position is used as a kth-step switching signal of the hybrid multilevel converter, data loading of a digital signal processor is completed, and a PWM switching signal is sent out at the moment corresponding to the optimal predicated position of a current trajectory. According to the hybrid multilevel converter and the variable switching frequency trajectory optimization control method of the hybrid multilevel converter, submodule direct current bus capacitance and voltage stable control of all full-bridge inverters and all half-bridge inverters of an upper bridge arm and a lower bridge arm can be achieved, rapid trajectory tracking of a current on the alternating current side can be achieved, and variable switching frequency optimization control can be achieved.

A method for judging magnetic suspension rotor system stability, includes through establishing magnetic suspension rotor system radial rotational motion dynamics differential equation model, adopting plurality conversion converting real coefficient two variable equation to complex system monovariant form, to obtain monovariant equivalent system, then drawing open loop transfer function dual frequency Bode chart of equifinality complex system, utilizing generalized classic frequency field stability criteria judging rotor precession and nutation stability, providing require condition for designing raising system stability correction link. The present invention converts multivariable system to equifinality monovariant system, thereby capable of using classical monovariant control theory to proceed stability analysis, relative usual multivariable method such as state space analysis method etc, not only very visual but also having good robustness, thus more fitting real system application.

The invention provides an information spreading model based on a cellular automaton, comprising the evolution rule, parameter and algorithm of the model and main technological indexes for measuring states. On such a basis, user states are divided into sever types, a rumor spreading process is analyzed and researched, and a rumor spreading mechanism is revealed according to differences of different parameter comparison results. A rumor spreading control strategy is further provided, and a great suppression effect on a final information spreading extent is achieved. Simulation of the model built based on the cellular automaton starts with microcosm to reflect macroscopic behaviors of groups, a simulation result is more intuitional, meanwhile, the model is flexible in structure, and can randomly change the control strategy during an evolutionary process, which is incomparable with a traditional differential equation model. Real data and a simulation experiment result have a relative high fitting degree, so that the built model has the actual application value, and can provide a support for making the rumor control strategy.

The invention discloses an uncertain multi-agent consistency tracking protocol design method for multiple agents in communication under DoS attack. The method comprises the following steps: for the multiple agents in communication under DoS attack, constructing a differential equation model of the agents; based on a switching system stability theory, transforming a multi-agent system consistency tracking problem into an asymptotic stability problem of a group of decoupling switching system; through a linear matrix inequality (LMI), constructing a decoupling switching system Lyapunov function,designing a multi-agent coordination consistency tracking protocol algorithm and communication duration proportion condition; and based on the linear matrix inequality (LMI) and the least communication duration proportion, designing the multi-agent consistency tracking protocol algorithm; and based on the linear matrix inequality (LMI) and the least communication duration proportion, designing a multi-agent rounding control algorithm.

The invention relates to an equation for analyzing the propagation law of malicious programs. As shown in Formula 1, the invention also relates to a malicious software diffusion prediction method based on a differential equation model. The method comprises the following steps: the number and information of malicious programs and the type of malicious programs are counted according to a selected area to obtain statistical data; data analysis is conducted to calculate the statistical data and to obtain the initial value, latency, infection rate and controlled rate; the initial value, incubationperiod, infection rate and controlled rate are substituted into the equations; a Kutta method is used to solve the problem, and the number of infected equipment is predicted. The invention adopts thebig data technology and the ordinary differential equation model to analyze and predict the equipment infected by the malicious program, so as to understand the spreading trend of the malicious program and formulate the related strategy.

The invention discloses an adaptive dynamic load monitoring method for power distribution network power failure events. According to the method, weights of four indexes, such as planned power failure time, power failure alarm time, power failure beginning time and power failure continuity time are calculated on the basis of an entropy weight method and by utilizing power distribution network business data, and the weights are integrated into one index; a differential equation model is established on the basis of a non-mechanism method to carry out data modeling, and parameters of the model are identified on the basis of the particle swarm optimization algorithm; an adaptive dynamic load monitoring model is calculated and verified, and power failure event diagnosis analysis and calculation result discuss are carried out according to a load cluster group. The method is advantaged in that, through model calculation and verification, the effect is substantially better than that of a dynamic monitoring model. The method is applied to actual business data analysis or other related fields, input data mining can be effectively carried out, load characteristic classification and decision are realized, and power failure event monitoring accuracy and validity are improved.

The invention relates to a radio energy transmission tuning control technology, in particular to a radio energy transmission system tuning feedback control method considering dynamic performance. Themethod includes the steps of obtaining the phase nonlinear differential equation model of the transmitting end and the receiving end of a system through analysis by utilizing a coupling mode theory and a circuit theory; utilizing a small signal linearization method to linearize a phase nonlinear differential equation model nearby a system resonance point, and analyzing the closed-loop dynamic performance of the system; taking the linear model as a controlled object and the phase value of the output voltage of an inverter circuit of the system as a reference phase, and forming a system tuning closed-loop feedback control loop by using the phase value of the transmitting end and the phase difference value of the transmitting end and the receiving end as the given values of the system transmitting end and the receiving end control loops when the system works in a resonance state. The control method can quickly and accurately control a magnetic coupling wireless energy transmission systemto work under the working states of high transmission efficiency and power, and has good robustness for detuning disturbance.

The invention discloses a directional partial differential equation filtering method used for an electronic speckle interference fringe pattern. The directional partial differential equation filtering method includes the steps of (1) inputting the electronic speckle interference fringe pattern u, (2) discretizing the electronic speckle interference fringe pattern u; (3) calculating the included angle theta<i,j> of the fringe direction and the x-axis direction; (4) obtaining a time step delta<t> in a self-adaptive mode; (5) obtaining iterations N<c>; (6) determining first-order partial derivatives u<x> and u<y> and second-order partial derivatives u<xx>, u<xy> and u<yy> of every pixel of the pattern u for every iteration; (7) determining the numerical solution of every pixel of the pattern u on the basis of the discretization format of a directional partial differential equation; (8) obtaining a filtered pattern when the numerical value of the largest set iteration N<c> is reached. The directional partial differential equation filtering method used for the electronic speckle interference fringe pattern can be widely used for filtering of high-noise electronic speckle interference fringe patterns.

The invention relates to a fingerprint texture image denoising enhancing method. According to the method, a non-local diffusion tensor diffusion-fluctuation partial differential equation model is provided. The method comprises the following steps that an image is preprocessed; the non-local structure tensor is calculated; the non-local diffusion tensor is calculated through the non-local structure tensor; iteration updating is carried out, and a result image is obtained. The texture characteristic can be enhanced while the noise is eliminated, a certain repairing function of the fracture texture is achieved, and the fingerprint recognition can be carried out more accurately in the later process.

The invention relates to a method for establishing a one-tube multi-machine differential equation model through a multi-machine form of a tunnel and a surge shaft, and belongs to the technical field of stability analysis and control of water turbines and hydraulic power units. N paths of steel pipes are arranged behind the surge shaft, m branch pipes are arranged at the tail end of each path of steel pipe, and the total number of the units is n * m; the hydraulic system dynamically comprises a tunnel, a surge shaft, a shared pipeline and a bifurcated pipe. The rated flow Qr and the rated waterhead Hr of the water turbine are taken as base values, the unit of Qr is 3 meters per second, and the unit of Hr is meter, the specific steps are as follows: 1. Dynamically decomposing the hydraulicpower of a tunnel and a surge shaft into a multi-machine form; and 2, constructing a differential algebraic multi-machine model of the one-pipe multi-machine belt surge shaft. The method provides a convenient calculation method and means for researching the stability of the complex hydraulic system and the hydropower station under the condition of electromechanical multi-field coupling and the multi-machine cooperative nonlinear control design of the hydropower station.

The invention discloses an inverter modeling method including an amplitude limiter and a related device. The method comprises the steps of building a differential equation model of a voltage and current control loop in an inverter, wherein the differential equation model comprises a piecewise function of a saturation amplitude limiter; determining the upper and lower boundaries and a critical saturation point of the saturation limiter based on the piecewise function of the saturation limiter; performing continuous function conversion on the piecewise function based on a Sigmoid function, and configuring parameters in the continuous function obtained by conversion through the upper and lower bounds of the saturation limiter and a critical saturation point to obtain a continuous function ofthe saturation limiter; and replacing the piecewise function of the saturation limiter with the continuous function of the saturation limiter to obtain a continuous differential equation model, thereby solving the technical problem that the inverter modeling through the differential equation requires that the inverter model has continuity and cannot be applied to the inverter comprising the limiter in the prior art.

The invention belongs to the field of big data and machine learning, particularly relates to a data-driven complex system mechanism automatic learning method, a system and equipment, and aims to solve the problems that an existing system modeling technology is difficult to predict a behavior trend from field observation data, a reconstructed mechanism model is not matched with physical observation data, the robustness is poor and the like. The method comprises the steps of obtaining historical multi-modal data and real-time multi-modal data, constructing a time sequence long-range correlation hypergraph model through airspace circulation memory coding, performing normalized combination on the hypergraph model through a neural differential equation model, and performing automatic iterative search on a continuous game network structure to obtain a system mechanism continuous dynamic model, and performing biological evolution simulation to obtain a causal model, and recalculating an association weight to obtain an active early warning system. According to the method, non-linearity, emergence, balance step, adaptability and special property description of a feedback loop of a complex system are realized, and the prediction accuracy of the model is improved.

The invention discloses a random drift particle swarm optimization method having a von Neumann structure. For biological network modeling, such as synthetic gene loop modeling, a differential equation model of a biological system needs to be established, and model parameters are inferred according to observed values; and in the parameter calculating process, an appropriate objective function and restraint conditions need to be defined first, and meanwhile, support of an optimization algorithm is needed. Parameter estimation of a differential equation is equivalent to a problem of nonlinear programming having differential-algebraic constraints, and an appropriate parameter set is found to fit measurement data; for the boundedness that optimization calculation in the parameter estimation process is easy to run into partial optimization, the von Neumann structure is introduced to the random drift particle swarm optimization algorithm to enhance global search capability; and the improved algorithm is improved for the boundedness of a full-mesh global topological structure, and adopts a local topological structure to serve as an inter-particle information sharing mode, so that global search capability of the algorithm is enhanced, and high-reliability model parameters can be given.

The invention provides a strip steel hot continuous rolling production billet reheating fuzzy PID temperature adjusting method and belongs to the field of strip steel hot continuous rolling production. The method comprises the following steps of performing T-S fuzzy partial differential equation modeling on a nonlinear billet internal temperature evolution dynamic model with spatial and temporal distribution characteristics to obtain a T-S fuzzy partial differential equation model of nonlinear billet internal temperature evolution dynamic; constructing a T-S fuzzy observer according to the T-S fuzzy partial differential equation model by adopting an observer-based feedback control technology; constructing a fuzzy PID temperature measurement tracking regulator based on a T-S fuzzy observer by applying a PID control strategy; and determining control parameters of the fuzzy PID temperature measurement tracking regulator, applying the fuzzy PID temperature measurement tracking regulator to the nonlinear billet internal temperature evolution dynamic model, and driving the billet lower surface temperature to be regulated to an expected temperature value. According to the method, non-intrusive steel billet temperature surface measurement and adjustment can be realized, so the lower surface temperature of the steel billet reaches an expected temperature value.

The invention relates to a differential equation calculation method for a one-pipe multi-machine hydroelectric generating set adjusting system and belongs to the technical field of stability analysis and control of water turbines and hydroelectric generating sets. The method comprises steps of hydraulic coupling of a one-pipe multi-machine system under rigid water attack being represented by using dynamic parameters of a hydraulic system and state variables of multiple units, dynamically converting hydraulic power sharing a pipeline into a differential equation model of flow of each unit, constructing a water turbine torque differential equation model and a hydraulic speed regulation system model on the basis of the hydraulic power, and establishing a hydraulic torque differential equation model and a hydraulic speed regulation system model; wherein the generator system models jointly form a one-tube multi-machine hydroelectric generating set adjusting system differential equation model; and under the condition that multiple units act at the same time, transient changes of water heads, flow and output of all the water turbines being calculated. The method provides a convenient calculation method and means for researching the regulation performance of each water-turbine generator set and the cooperative control design among a plurality of water-turbine generator sets under the water-turbine-electricity multi-factor coupling condition.

The invention discloses a heat-electricity-hydrogen multi-energy-flow comprehensive energy system and an optimal scheduling method thereof, the multi-energy-flow comprehensive energy system adopts a heat-electricity-hydrogen multi-energy-flow comprehensive energy system, and the optimal scheduling method comprises the following steps: initializing the system; performing time sequence prediction based on deep learning to obtain intra-day wind-solar power generation and electric heating load power; in the first-stage scheduling, an objective function and a constraint condition are set based on economy, a scheduling step length is set, day-ahead scheduling is carried out on the multi-energy-flow integrated energy system, and an optimal solution is obtained; performing ultra-short-term prediction on the wind-light and electric heating load by establishing a multivariate time sequence differential equation model; in the second-stage scheduling, the optimal solution of the day-ahead scheduling is used as a reference value, the output of each micro source in the day is corrected with the minimum error as a target based on a rolling scheduling period, the precision of optimization control is improved, the influence of the error on the operation cost is reduced, and the reliability of a comprehensive energy system and the stability of an energy storage system are enhanced.

The invention provides a time-domain analysis method for a dynamic temperature difference power generation system segmented on the basis of a temperature difference thermocouple material. The method comprises steps as follows: firstly, a thermoelectric partial differential equation model and boundary conditions of basic physical properties of the dynamic temperature difference power generation system segmented on the basis of the temperature difference thermocouple material are established; secondly, a thermal circuit and an electric circuit of the system are more specifically analyzed and subjected to time and space discretization processing, an algebraic equation of physical quantities of internal nodes and boundary nodes of an area is established in terms of energy conservation, iteration solution is executed by starting from a time initial value, and finally, numerical solutions of temperature and electric field intensity of all positions in the dynamic temperature difference power generation system segmented on the basis of the temperature difference thermocouple material can be obtained. By means of analysis for the numerical solutions of temperature and electric field intensity in the dynamic temperature difference power generation system, the temperature difference power generation system can be further analyzed and designed specifically, and a result is more accurate.

The invention discloses a multi-factor model optimization method for a gene regulatory network, which comprises the following steps: inferring the gene regulatory network by utilizing a nonlinear differential equation model and a differential operation method of a gene and combining a machine learning algorithm, and optimizing a key parameter attenuation rate based on a multi-objective optimization idea and a genetic algorithm. According to the invention, the effects of noise, attenuation rate and time delay are introduced on the basis of a traditional differential equation model, a nonlinear regulation and control function is trained based on a machine learning algorithm, and the nonlinear dynamic process of gene expression can be better simulated; the algorithm is used for optimizing the attenuation rate and time delay of important parameters influencing gene expression, and then a more accurate and efficient mathematical model is constructed to deduce the gene regulation network from gene expression data.