Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN....Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN. The WNN has the characteristics of fast convergence and improved capability of nonlinear approximation. For the purpose of adapting the timevarying characteristics of flood routing, the WNN is coupled with an AR real-time correction model. The AR model is utilized to calculate the forecast error. The coefficients of the AR real-time correction model are dynamically updated by an adaptive fading factor recursive least square（RLS） method. The application of the flood forecasting method in the cross section of Xijiang River at Gaoyao shows its effectiveness.展开更多
A multi-loop adaptive internal model control (IMC) strategy based on a dynamic partial least squares (PLS) frame-work is proposed to account for plant model errors caused by slow aging,drift in operational conditions,...A multi-loop adaptive internal model control (IMC) strategy based on a dynamic partial least squares (PLS) frame-work is proposed to account for plant model errors caused by slow aging,drift in operational conditions,or environmental changes.Since PLS decomposition structure enables multi-loop controller design within latent spaces,a multivariable adaptive control scheme can be converted easily into several independent univariable control loops in the PLS space.In each latent subspace,once the model error exceeds a specific threshold,online adaptation rules are implemented separately to correct the plant model mismatch via a recursive least squares (RLS) algorithm.Because the IMC extracts the inverse of the minimum part of the internal model as its structure,the IMC controller is self-tuned by explicitly updating the parameters,which are parts of the internal model.Both parameter convergence and system stability are briefly analyzed,and proved to be effective.Finally,the proposed control scheme is tested and evaluated using a widely-used benchmark of a multi-input multi-output (MIMO) system with pure delay.展开更多
In situation assessment(SA)of missile versus target fighter,the traditional SA models generally have the characteristics of strong subjectivity and poor dynamic adaptability.This paper considers SA as an expectation o...In situation assessment(SA)of missile versus target fighter,the traditional SA models generally have the characteristics of strong subjectivity and poor dynamic adaptability.This paper considers SA as an expectation of future returns and establishes a missile-target simulation battle model.The actor-critic(AC)algorithm in reinforcement learning(RL)is used to train the evaluation network,and a missile-target SA model is established in simulation battle training.Simulation and comparative experiments show that the model can effectively estimate the expected effect of missile attack under the current situation,and it provides an effective basis for missile attack decision.展开更多
The model of lumped element circuit ignores the finite time of signals to propagate around a circuit. However, using modern oscilloscope, the time of nanoseconds in a circuit can be measured. Then the speed of alterna...The model of lumped element circuit ignores the finite time of signals to propagate around a circuit. However, using modern oscilloscope, the time of nanoseconds in a circuit can be measured. Then the speed of alternating electricity can be obtained in a RL circuit. A typical RL circuit is formed by a power source, wire, resistance and inductance. The basic formula is: U(t)=I(t)R+LdI(t)/dt. It can be derived from the Ohm’s law and Kirchhoff laws. Based on our experimental results, this paper has discussed the new explanation of this equation in a RL circuit. As a result, the speed of alternating electricity is greater than light in a special RL circuit. The model of lumped element circuit can be improved when considering the finite time of signals.展开更多
基金The National Natural Science Foundation of China(No.50479017).
文摘Based on analyzing the limitations of the commonly used back-propagation neural network （BPNN）, a wavelet neural network （WNN） is adopted as the nonlinear river channel flood forecasting method replacing the BPNN. The WNN has the characteristics of fast convergence and improved capability of nonlinear approximation. For the purpose of adapting the timevarying characteristics of flood routing, the WNN is coupled with an AR real-time correction model. The AR model is utilized to calculate the forecast error. The coefficients of the AR real-time correction model are dynamically updated by an adaptive fading factor recursive least square（RLS） method. The application of the flood forecasting method in the cross section of Xijiang River at Gaoyao shows its effectiveness.
基金Project supported by the National Natural Science Foundation of China (No.60574047)the National High-Tech R & D Program (863)of China (No.2007AA04Z168)the Research Fund for the Doctoral Program of Higher Education of China (No.20050335018)
文摘A multi-loop adaptive internal model control (IMC) strategy based on a dynamic partial least squares (PLS) frame-work is proposed to account for plant model errors caused by slow aging,drift in operational conditions,or environmental changes.Since PLS decomposition structure enables multi-loop controller design within latent spaces,a multivariable adaptive control scheme can be converted easily into several independent univariable control loops in the PLS space.In each latent subspace,once the model error exceeds a specific threshold,online adaptation rules are implemented separately to correct the plant model mismatch via a recursive least squares (RLS) algorithm.Because the IMC extracts the inverse of the minimum part of the internal model as its structure,the IMC controller is self-tuned by explicitly updating the parameters,which are parts of the internal model.Both parameter convergence and system stability are briefly analyzed,and proved to be effective.Finally,the proposed control scheme is tested and evaluated using a widely-used benchmark of a multi-input multi-output (MIMO) system with pure delay.
基金the National Natural Science Foundation of China(No.61627810)the Joint Fund of Advanced Aerospace Manufacturing Technology Research of China(No.USCAST2016)the National Key Research and Development Program of China(No.2018YFB1305003)。
文摘In situation assessment(SA)of missile versus target fighter,the traditional SA models generally have the characteristics of strong subjectivity and poor dynamic adaptability.This paper considers SA as an expectation of future returns and establishes a missile-target simulation battle model.The actor-critic(AC)algorithm in reinforcement learning(RL)is used to train the evaluation network,and a missile-target SA model is established in simulation battle training.Simulation and comparative experiments show that the model can effectively estimate the expected effect of missile attack under the current situation,and it provides an effective basis for missile attack decision.
文摘The model of lumped element circuit ignores the finite time of signals to propagate around a circuit. However, using modern oscilloscope, the time of nanoseconds in a circuit can be measured. Then the speed of alternating electricity can be obtained in a RL circuit. A typical RL circuit is formed by a power source, wire, resistance and inductance. The basic formula is: U(t)=I(t)R+LdI(t)/dt. It can be derived from the Ohm’s law and Kirchhoff laws. Based on our experimental results, this paper has discussed the new explanation of this equation in a RL circuit. As a result, the speed of alternating electricity is greater than light in a special RL circuit. The model of lumped element circuit can be improved when considering the finite time of signals.
