This paper mainly focuses on a disturbance rejection control methodology for a series of mismatched and matched disturbed complex systems. Firstly, a novel finite-time arbitrary-order extended state observer (A-ESO) i...
详细信息
In this paper, we propose a novel regularization algorithm that is introduced as a penalty term to the loss function. Differing from conventional L1 and L2 regularization methods, our approach does not aim to diminish...
In this paper, we propose a novel regularization algorithm that is introduced as a penalty term to the loss function. Differing from conventional L1 and L2 regularization methods, our approach does not aim to diminish the weights of individual neurons or enforce sparsity by driving certain neurons to zero. Instead, it functions by increasing the differences between neurons and enhancing the diversity of neurons within each layer. Our method incorporates ensemble learning techniques by treating the layer weight matrix as a collective learning model, where each neuron serving as a weak learner within the layer. The proposed algorithm improves the performance of DCNN by simultaneously considering the distance between multiple filters in the same layer. This algorithm reduces the redundancy of the parameter layer filters in DCNN and enhances its robustness. The penalty term proposed by our algorithm dynamically adjusts its value in a cyclical manner, compelling the neural network to navigate away from its current gradient state. In the parameter space, different weights correspond to different locations. The proposed algorithm quantifies the distance between neurons and iteratively increases the distance between neurons during thereby encouraging greater diversity within the network. Experimental evaluations demonstrate the effectiveness of our algorithm in enhancing neural network performance without requiring adjustments to other hyper-parameters.
In order to improve the displacement control precision of continuous casting mold, the establishment of precise mathematical model of non-sinusoidal vibration displacement control system is important. The modeling met...
详细信息
ISBN:
(纸本)9781509009107
In order to improve the displacement control precision of continuous casting mold, the establishment of precise mathematical model of non-sinusoidal vibration displacement control system is important. The modeling method of BP neural network optimized by genetic algorithm is adopted, which can solve the problem of the mechanism modeling of various approximate treatments. Firstly the system model of BP neural network is constructed according to the measured data received from industrial field. Secondly, BP neural network model of genetic optimization is built. Then simulations are eventually performed, and then by comparing the error between the predicted displacement outputs of BP neural network optimized by genetic algorithm and the actual output displacement of the system, the simulation results show that the system error of the identified model is within 4%, and the identification method of BP neural network optimized by genetic algorithm is superior to the mechanism modeling.
In this paper, the technology of ZigBee based on wireless sensor network was adopted to carry out the real-time oxyhemoglobin saturation monitoring. C8051F340 was used to control UZ2400 as MCU(Micro control Unit). Ele...
详细信息
The consensus based distributed frequency control strategy makes all measurements and control units in microgrids vulnerable to false data injection (FDI) attacks. This paper presents an observer-based resilient frequ...
详细信息
The problem of adaptive robust control for uncertain nonlinear systems with time-varying delay is considered in this paper. The upper bound of the uncertainty is assumed to be nonlinear functions with unknown coeffici...
详细信息
Obstacle avoidance is a significant research content in multi-agents formation control. The obstacle avoidance of multi-agents systems is investigated in this paper, and an improved artificial potential field method (...
详细信息
Deploying reconfigurable intelligent surfaces (RIS) in vehicular networks can effectively improve the quality of wireless channel and increase network throughout, which is realized by optimizing the phase shifters. Ho...
详细信息
The quality and yield of products in the process of aluminum hot rolling is affected seriously by the load distribution. The load distribution of the hot tandem rolling is actually a problem of multi-objective optimiz...
详细信息
ISBN:
(纸本)9781467374439
The quality and yield of products in the process of aluminum hot rolling is affected seriously by the load distribution. The load distribution of the hot tandem rolling is actually a problem of multi-objective optimization(MOO. The model of the rolling schedule was established based on slippage factor, which was also set as an optimization objective. In this paper, an improved NSGA-II is applied to optimize the rolling schedule of aluminum hot tandem rolling. This paper also proposed a method to select the best solution based on the preference after process of the improved NSGA-II. This method shows its ability on the slippage preventing and load distribution.
A robust model predictive control (MPC) algorithm is presented for a class of discrete-time uncertain systems with time-varying delay, input constraints and norm-bounded nonlinear perturbations. By minimizing the uppe...
详细信息
ISBN:
(纸本)9787894631046
A robust model predictive control (MPC) algorithm is presented for a class of discrete-time uncertain systems with time-varying delay, input constraints and norm-bounded nonlinear perturbations. By minimizing the upper bound of a cost function, an MPC state feedback control law is calculated at each step. The proposed MPC method for delayed system is delay-dependent, and it is effective to deal with nonlinear perturbations. It has been shown that the robust stability of the closed-loop systems is guaranteed by the proposed algorithm. Finally, a numerical example is given to illustrate the effectiveness of the proposed method.
暂无评论