50 results about "Nonlinear dynamic systems" patented technology

The invention discloses a strong tracking Kalman filter method for target tracking, belongs to the field of target tracking, and relates to a maneuvering target method based on a strong tracking Kalman filter. The method comprises the steps of building a disperse nonlinear dynamic system model; carrying out system initialization; carrying out time updating, and introducing a time varying and fading factor [lambda k]; measuring and updating; and finally carrying out filtering updating. According to the method, the time varying and fading factor is introduced into a Kalman filter, therefore, the method has the advantages that the Kalman filter is simple to realize and high in filter precision; meanwhile, the strong tracking Kalman filter has a real-time tracking capacity on the sudden change of the system.

The invention provides a state estimation method for a nonlinear dynamic system under non-Gaussian noise. According to the method, a nonlinear dynamic system model comprising a process noise signal and a non-Gaussian measurement noise signal is adopted; the real-time information of the system state is estimated based on the nonlinear dynamic process model and the measurement feedback signal to obtain a state estimation signal, and output feedback is performed; and the state estimation signal is used as the controller input at the next moment, so that the influence of the non-Gaussian measurement noise signal on the system performance is weakened, and the accuracy of state estimation is improved. According to the method, the model prediction information and the non-Gaussian signal measuredby the sensor are combined, and the current optimal state of the system is estimated by adopting the newly proposed extended Kalman filtering algorithm based on the dynamic data correction technology,so that the limitation on the measurement noise distribution in the existing Kalman filtering technology is broken through; and a selectable scheme is provided for solving the state estimation problem of the nonlinear dynamic system under the condition of non-Gaussian measurement noise.

The invention proposes a data tracking based recommendation system security detection method to overcome the shortcomings of long time for user profile injection, poor attack effect, incapability of adapting to big data processing and the like in conventional collaborative filtering recommendation system detection. According to the method, a score state of a project is tracked and predicted by using a characteristic that extended Kalman filtering (EKF) can be applied to a time nonlinear dynamic system, and then users with abnormal scores in the project are subjected to clustering analysis by utilizing linear discriminant analysis (LDA), so that attack users in the project and the profiles of the users can be determined. With the adoption of the EKF method, the detection of a large amount of unrelated data is reduced, so that the detection efficiency is improved and the system robustness is enhanced. A tracking algorithm is used for recommendation system security detection and can realize online continuous system detection, so that the error detection rate is reduced. The LDA method can perform dimension reduction on multi-characteristic users, so that the profile injection attack of malicious users is effectively detected and the detection rate is increased.

The invention provides an improved Gaussian particle filter algorithm (KLD-GPF) based on KLD, belongs to the technical field of signal processing, and relates to nonlinear filtering, and the method provided by the invention is suitable for state estimation of a nonlinear dynamic system. The algorithm can adaptively adjust the number of particles, and has a remarkable effect under the conditions that noise obeys Gaussian distribution and the statistical property of the noise is suddenly changed. According to the filtering algorithm, the Kullback-Leibler (KL) distance between a discrete probability density function (PDF) of particles and a real posterior probability density function is calculated online in the sampling process, and the size of a particle set is adjusted online according to the KL distance, so that the algorithm has relatively good robustness. The KLD-GPF can maintain a good estimation effect under the condition of sudden change of noise statistical characteristics. Compared with a KLD improved particle filtering algorithm (KLD-PF), although some filtering precision is lost, the filtering speed is greatly improved.

The invention discloses a chaotic circuit with coexistent asymmetric multi-attractors. The circuit contains two nonlinear terms described by an absolute value function and is a third-order chaotic circuit with a degraded Jerk form. The circuit is composed of a reversal, addition and integration operational module based on an operational amplifier and a resistor or a capacitor and a nonlinear operational module composed of an analog multiplier and an operational amplifier. Under a certain initial condition, the circuit can display a coexistent phenomenon, namely a multi-stability phenomenon of the asymmetric multi-attractors. As a result, a great degree of freedom is provided for engineering application of a nonlinear dynamic system, and meanwhile the circuit has important theoretical physical significance and engineering application value for research about realization of the coexistent multi-attractors and a hardware circuit comprising the coexistent multi-attractors.

The invention discloses a nonlinear analysis method for steering stability of an electric automobile. A nonlinear dynamics model capable of describing steering of the electric automobile is established, a nonlinear differential equation is constructed, then on the basis of an adiabatic elimination principle, order reduction processing is performed on the constructed nonlinear differential equation for steering of the electric automobile, in this way, fast relaxation parameters in a system are eliminated, and a simplified nonlinear differential equation is obtained; finally, on the basis of an isoclinic line principle, a multi-step recurrence method is adopted for drawing the phase trajectory of the nonlinear dynamics system for steering of the electric automobile, the balance state, stability precision and stability of the system are visually reflected, and the stability of the electric automobile in the steering process at different speeds and turning angles is analyzed.

The invention discloses a high-order extended Kalman filter design method of a strong nonlinear dynamic system. The method includes, firstly, modeling a state model and a measurement model of a strong nonlinear dynamic system based on a multi-dimensional high-order polynomial; secondly, defining a high-order polynomial in the state model as a hidden variable of the system, and equivalently rewriting the state model into a pseudo-linear model based on combination of an original variable and the hidden variable; thirdly, taking the high-order hidden variables as additive parameters of each order of the system, and establishing a dimension expansion linear state model combining the state and the parameters by performing random dynamic modeling on the high-order hidden variables; then, processing the measurement model correspondingly, and modeling an original system into a dimension expansion linear measurement model based on combination of states and parameters; and finally, designing a novel high-order extended Kalman filter aiming at original state estimation based on a dimension-extended linear system. According to the invention, the validity of the new filter is verified through digital simulation.

The invention discloses a Gaussian process model-based predictive control method for a multi-variable nonlinear dynamic system model, belongs to the technical field of predictive control of the multi-variable nonlinear dynamic system model and aims at solving the technical problem of providing improvement of the Gaussian process model-based predictive control method for the multi-variable nonlinear dynamic system model. The technical scheme adopted for solving the technical problem is as follows: the predictive control method comprises the following steps of (1) building an external dynamic PLS framework; (2) predicting output data and obtaining a plurality of single-input and single-output systems in hidden space through decoupling of a dynamic GP-PLS model; (3) carrying out control by using the dynamic GP-PLS model and designing a model predictive controller in each single-input and single-output system; (4) obtaining an optimum control action through minimizing an objective function; and (5) reconstructing the model predictive control result in the hidden space back to original space and controlling the original space. The Gaussian process model-based predictive control method is applied to the multi-variable nonlinear dynamic system model.

The embodiments of the invention provide an autonomous underwater robot and path following control method and device and relate to the robot technology field. The method comprises the following stepsof obtaining a built autonomous underwater robot motion system; generating a nonlinear dynamic system of the autonomous underwater robot motion system according to a physics principle; receiving a real-time position acquired by the positioning module of the autonomous underwater robot, and rewriting the real-time position according to a dynamic model; and generating a nonlinear robust inverse stepcontroller based on a disturbance term of the autonomous underwater robot, the real-time position and a preset input-state stability theory, and controlling the autonomous underwater robot motion system so that the autonomous underwater robot autonomously and independently completes precise path following operation and achieves stable operation.

The invention discloses a high speed light-load mechanism nonlinear dynamic system structure evolution optimization method. The method aims at a nonlinear kinetic analysis to directly calculate sensitivity information of element strain energy and directly obtain the element strain energy at each time point, the strain energy at all the time points is superposed after normalization processing, the comprehensive sensitivity for measuring all discrete points is obtained, structure modification is performed by adopting an evolutionary structural method, then the structure is updated, the nonlinear analysis is performed again, and the process is repeated until an optimal object is realized. According to the method disclosed by the invention, the equivalent load calculation is omitted, and the change on inertial load by structural modification is directly reflected into a model to obtain a more optimal optimization result. According to the method disclosed by the invention, the equivalent static load is considered as a function of a design variable, and a corresponding calculation method is proposed, compared with a standard equivalent static load, the amplitude in the same movement condition can be lowered by half, and the performance of the high speed light-load mechanism is greatly improved.

The invention provides an industrial process operation index intelligent forecasting method, device and equipment and a storage medium. The industrial process operation index intelligent forecasting method comprises the steps of establishing an operation index dynamic model according to the characteristic that the change of an operation index depends on the dynamic characteristics of an industrial process control system, wherein the operation index dynamic model comprises an identifiable model and an unmodeled dynamic model; estimating parameters of the identifiable model in the operation index dynamic model; combining the identification error of the parameter of the identifiable model in the operation index dynamic model and the unmodeled dynamic state in the operation index dynamic model into an unknown nonlinear dynamic system; establishing an online intelligent forecasting model of the unknown nonlinear dynamic system; and obtaining a forecast value of the operation index according to output of the identifiable model in the operation index dynamic model and output of the online intelligent forecast model of the unknown nonlinear dynamic system. Aiming at the problem that industrial process operation indexes are difficult to forecast, a system identification method based on a mechanism model is combined with a deep learning method based on big data, an intelligent forecasting method of the industrial process operation indexes is provided, and the problem of forecasting of the industrial process operation indexes is solved.

The invention provides an RTK Kalman filtering optimization method for unmanned aerial vehicles, which comprises the following steps: an acquisition terminal located on an unmanned aerial vehicle acquires a plurality of geodetic coordinates of the unmanned aerial vehicle under a WGS-84 coordinate system, a state equation and an observation equation of Kalman filtering under a nonlinear dynamic system are established, and a Kalman filtering model is established according to the state equation and the observation equation of Kalman filtering; an initial state of the nonlinear dynamic system is determined, namely, a state parameter initial value, a variance matrix initial value and a dynamic noise initial variance matrix of an initial epoch are confirmed; and based on the state equation and the observation equation of the initial epoch, the state parameter initial value, the variance matrix initial value and the dynamic noise initial variance matrix are filtered by a Kalman filtering recursion equation to obtain a filtering value. The positioning precision of unmanned aerial vehicles can be improved.

The invention provides a fast Gaussian particle filter data fusion method based on artificial fish swarm optimization, belongs to the technical field of signal processing, and is mainly used for solving the problems of huge calculation workload and low precision of a particle filter in a multi-particle state. According to the method, Gaussian particle filtering is used as a framework, an artificial fish swarm algorithm is fused, and foraging behaviors and clustering behaviors are used for optimizing weights. According to the method, traditional sampling is replaced by linear transformation, the weight is optimized according to the measurement value and the weight calculation formula, the calculation speed is guaranteed while the calculation precision is improved, and the method is suitablefor application occasions such as state estimation of a nonlinear dynamic system.