Prescribed-time robust ZNN models for solving equality and inequality systems

At present, there are few studies on solving time-variant linear equality and inequality systems (TVLEIS) under noise interference, and the numerical algorithm has limitations in solving the TVLEIS problems. Therefore, to determine the online solution of the TVLEIS in a complex environment, two prescribed-time robust zeroing neural network (PTRZNN) models are proposed, investigated, and verified in this paper. The PTRZNN models have a faster convergence rate and superior robustness compared with other zeroing neural network models activated by common activation functions. In addition, the detailed theoretical derivation of the prescribed-time convergence and robustness of the PTRZNN models is provided. The effectiveness and superiority of the PTRZNN models for determining the TVLEIS are further demonstrated by simulation results. It is worth mentioning that the design idea of the PTRZNN models is applied to the multi-agent system, which shows the practical value of the PTRZNN models.     

Super finite-Time variable parameter ZNN models for time-Variant linear matrix inequality

Stuck in the speed and dimensionality of settling time-variant linear matrix inequality (LMI), this paper for the first time proposes two finite-time variable parameter zeroing neural network (FTVPZNN) models to settle the time-variant LMI. The first model is called the FTVPZNN-C model activated by the conventional sign-bi-power (S-B-P) function. The second model is called the FTVPZNN-T model activated by a tunable parameter S-B-P function. Different from the finite-time fixed-value zeroing neural network (FTFZNN) model, the proposed FTVPZNN models with variable parameters have better convergence performance and smaller upper bounds of finite-time convergence. Three theorems are presented to guarantee the stability and finite-time convergence of the FTVPZNN models. Especially, through detailed theoretical analysis and calculations, the finite-time convergence upper bounds of the proposed FTVPZNN models are obtained. Finally, a numerical simulative example is given to affirm the effectiveness and excellent convergent performance of the proposed models for settling the time-variant LMI.     

Solving future equation systems using integral-type error function and using twice ZNN formula with disturbances suppressed

In this paper, for solving future equation systems, two novel discrete-time advanced zeroing neural network models are proposed, analyzed and investigated. First of all, by using integral-type error function and twice zeroing neural network (or termed, Zhang neural network) formula, as the preliminaries and bases of future problems solving, two continuous-time advanced zeroing neural network models are presented for solving continuous time-variant equation systems. Secondly, a one-step-ahead numerical differentiation rule termed 5-instant discretization formula is presented for the first-order derivative approximation with higher computational precision. By exploiting the presented 5-instant discretization formula to discretize the continuous-time advanced zeroing neural network models, two novel discrete-time advanced zeroing neural network models are proposed. Theoretical analyses on the convergence and precision of the discrete-time advanced zeroing neural network models are proposed. In addition, in the presence of disturbance, the proposed discrete-time advanced zeroing neural network models still possess excellent performance. Comparative numerical experimental results further substantiate the efficacy and superiority of the proposed discrete-time advanced zeroing neural network models for solving the future equation systems.     

Fixed-time convergence integral-enhanced ZNN for calculating complex-valued flow matrix Drazin inverse

The complex-valued flow matrix Drazin inverse has recently attracted considerable interest from researchers due to its great academic value. In this paper, a fixed-time convergence integral-enhanced zeroing neural network (FTCIEZNN) model is proposed and investigated for calculating the Drazin inverse of complex-valued flow matrix. Since the FTCIEZNN model possesses fixed-time convergence, its upper limit of convergence time is irrelevant to initial conditions and can be adjusted by specified system parameters. Meanwhile, by adopting the newly designed reformed nonlinear activation function (RNAF) and variable parameters, the FTCIEZNN model converges rapidly in a relatively fast fixed-time and its robustness is dramatically strengthened. In addition, the upper limit of the convergence time in the absence of noise and the upper limit of the steady-state error in the presence of time-varying bounded noise are given by a scrupulous mathematical logic calculation. Furthermore, the outcomes of the numerical simulations demonstrate that the FTCIEZNN model outshines existing zeroing neural network models in calculating complex-valued flow matrix Drazin inverse. Finally, an application based on the FTCIEZNN model in image encryption fully illustrates the practical value of the FCIEZNN model.     

A modified noise-tolerant ZNN model for solving time-varying Sylvester equation with its application to robot manipulator

Recently, Xiao et al. (2021) proposed an efficient noise-tolerant zeroing neural network (NTZNN) model with fixed-time convergence for solving the time-varying Sylvester equation. In this paper, we propose a modified version of their NTZNN model, named the modified noise-tolerant zeroing neural network (MNTZNN) model. It extends the NTZNN model to a more general form and then we prove that, with appropriate parameter selection, our new MNTZNN model can significantly accelerate the convergence of the NTZNN model. Numerical experiments confirm that the MNTZNN model not only maintains fixed-time convergence and noise-tolerance but also has a faster convergence rate than the NTZNN model under certain conditions. In addition, the design strategy of the MNTZNN is also successfully applied to the path tracking of a 6-link planar robot manipulator under noise disturbance, which demonstrates its applicability and practicality.     

A novel finite-time complex-valued zeoring neural network for solving time-varying complex-valued Sylvester equation

A novel finite-time complex-valued zeroing neural network (FTCVZNN) for solving time-varying Sylvester equation is proposed and investigated. Asymptotic stability analysis of this network is examined with any general activation function satisfying a condition or with an odd monotonically increasing activation function. So far, finite-time model studies have been investigated for the upper bound time of convergence using a linear activation function with design formula for the derivative of the error or with variations of sign-bi-power activation functions to zeroing neural networks. A function adaptive coefficient for sign-bi-power activation function (FA-CSBP) is introduced and examined for faster convergence. An upper bound on convergence time is derived with the two components in the function adaptive coefficients of sign-bi-power activation function. Numerical simulation results demonstrate that the FTCVZNN with function adaptive coefficient for sign-bi-power activation function is faster than applying a sign-bi-power activation function to the zeroing neural network (ZNN) and the other finite-time complex-valued models for the selected example problems.     

Finite-time convergent zeroing neural network for solving time-varying algebraic Riccati equations

Various forms of the algebraic Riccati equation (ARE) have been widely used to investigate the stability of nonlinear systems in the control field. In this paper, the time-varying ARE (TV-ARE) and linear time-varying (LTV) systems stabilization problems are investigated by employing the zeroing neural networks (ZNNs). In order to solve the TV-ARE problem, two models are developed, the ZNNTV-ARE model which follows the principles of the original ZNN method, and the FTZNNTV-ARE model which follows the finite-time ZNN (FTZNN) dynamical evolution. In addition, two hybrid ZNN models are proposed for the LTV systems stabilization, which combines the ZNNTV-ARE and FTZNNTV-ARE design rules. Note that instead of the infinite exponential convergence specific to the ZNNTV-ARE design, the structure of the proposed FTZNNTV-ARE dynamic is based on a new evolution formula which is able to converge to a theoretical solution in finite time. Furthermore, we are only interested in real symmetric solutions of TV-ARE, so the ZNNTV-ARE and FTZNNTV-ARE models are designed to produce such solutions. Numerical findings, one of which includes an application to LTV systems stabilization, confirm the effectiveness of the introduced dynamical evolutions.     

Accelerating a recurrent neural network to finite-time convergence using a new design formula and its application to time-varying matrix square root

In this paper, a new design formula is presented to accelerate the convergence speed of a recurrent neural network, and applied to time-varying matrix square root finding in real time. Then, according to such a new design formula, a finite-time Zhang neural network (FTZNN) is proposed and investigated for finding time-varying matrix square root. In comparison with the original Zhang neural network (ZNN) model, the FTZNN model makes a breakthrough in the convergence performance (i.e., from infinite time to finite time). In addition, theoretical analyses of the design formula and the FTZNN model are provided in details. Comparative results further verify the superiority of the proposed FTZNN model to the original ZNN model for finding time-varying matrix square root.     

Gradient-based neural networks for solving periodic Sylvester matrix equations

This paper considers neural network solutions of a category of matrix equation called periodic Sylvester matrix equation (PSME), which appear in the process of periodic system analysis and design. A linear gradient-based neural network (GNN) model aimed at solving the PSME is constructed, whose state is able to converge to the unknown matrix of the equation. In order to obtain a better convergence effect, the linear GNN model is extended to a nonlinear form through the intervention of appropriate activation functions, and its convergence is proved through theoretical derivation. Furthermore, the different convergence effects presented by the model with various activation functions are also explored and analyzed, for instance, the global exponential convergence and the global finite time convergence can be realized. Finally, the numerical examples are used to confirm the validity of the proposed GNN model for solving the PSME considered in this paper as well as the superiority in terms of the convergence effect presented by the model with different activation functions.     

Trajectory tracking double two-loop adaptive neural network control for a Quadrotor

In this paper, the development and experimental validation of a novel double two-loop nonlinear controller based on adaptive neural networks for a quadrotor are presented. The proposed controller has a two-loop structure: an outer loop for position control and an inner loop for attitude control. Similarly, both position and orientation controllers also have a two-loop design with an adaptive neural network in each inner loop. The output weight matrices of the neural networks are updated online through adaptation laws obtained from a rigorous error convergence analysis. Thus, a training stage is unnecessary prior to the neural network implementation. Additionally, an integral action is included in the controller to cope with constant disturbances. The error convergence analysis guarantees the achievement of the trajectory tracking task and the boundedness of the output weight matrix estimation errors. The proposed scheme is designed such that an accurate knowledge of the quadrotor parameters is not needed. A comparison against the proposed controller and two other well-known schemes is presented. The obtained results showed the functionality of the proposed controller and demonstrated robustness to parametric uncertainty.     

Passivity analysis of coupled neural networks with reaction–diffusion terms and mixed delays

In this paper, we intend to discuss the passivity of coupled neural networks (NNs) with reaction–diffusion terms and mixed delays. By constructing appropriate Lyapunov functional, and with the help of liner matrix inequalities, some inequality techniques, several sufficient conditions are derived to guarantee the output strictly passive, input strictly passive, passive of the proposed neural network model. Then, a stability criterion is presented according to the obtained passivity results. Moreover, the proposed neural network model herein is more general than some recent studies, which can improve and enrich the previous research results. Finally, a numerical example is presented to show the effectiveness of the theoretical criteria.     

Cooperative control with designated convergence rate for high-order integrators under heterogeneous couplings

In this paper, we investigate the cooperative control problem of high-order integrators under heterogeneous couplings. A new class of distributed control algorithms are developed for the designated convergence rate (DCR) problem of high-order integrators, which could explicitly show the convergence margin of the closed-loop system, and has better robustness than conventional consensus algorithms. We first propose state consensus control algorithms for high-order integrators, where necessary and sufficient convergence conditions are proposed by theoretical analysis. Then we extend the results to the case of output leaderless consensus of heterogeneous high-order integrators with heterogeneous couplings. Finally, simulation examples are given to validate the effectiveness of the proposed algorithms.     

Terminal iterative learning control for discrete-time nonlinear systems based on neural networks

The terminal iterative learning control is designed for nonlinear systems based on neural networks. A terminal output tracking error model is obtained by using a system input and output algebraic function as well as the differential mean value theorem. The radial basis function neural network is utilized to construct the input for the system. The weights are updated by optimizing an objective function and an auxiliary error is introduced to compensate the approximation error from the neural network. Both time-invariant input case and time-varying input case are discussed in the note. Strict convergence analysis of proposed algorithm is proved by the Lyapunov like method. Simulations based on train station control problem and batch reactor are provided to demonstrate the effectiveness of the proposed algorithms.     

基于粒子群优化神经网络的煤层顶板导水裂隙带高度预测研究

阐述了运用粒子群优化人工神经网络建立煤层顶板导水裂隙带高度预测模型的思路与方法。利用粒子群优化神经网络模型的权值和阈值,克服了神经网络容易收敛到局部最小值,以及收敛速度慢的缺点。煤层导水裂隙带高度预测实例表明,该方法不仅能更快地收敛于最优解,且预测精度有明显的提高。     

Fixed-time neuroadaptive practical tracking control based on extended state/disturbance observer for a QUAV with external disturbances and time-varying parameters

This paper investigates the fixed-time neural network adaptive (FNNA) tracking control of a quadrotor unmanned aerial vehicle (QUAV) to achieve flight safety and high efficiency. By combining radial basis function neural network (RBFNN) with fixed time adaptive sliding mode algorithm, a novel radial basis function neural network adaptive law is proposed. In addition, an extended state/disturbance observer (ESDO) is proposed to solve the problem of unmeasurable state and external interference, which can obtain reliable state feedback and interference input. Unlike most other ESO applications, this paper does not set the uncertainty model and external disturbances as total disturbances. Instead, the external disturbances are observed by extending the states and the observed states are fed back to the controller to cancel the disturbances. In view of the time-varying resistance coefficient and inertia torque in the QUAV model, the neural network is introduced so that the controller does not need to consider these nonlinear uncertainties. Finally, a numerical example is given to verify the effectiveness of the coupled non-simplified QUAV model.     

General value iteration based single network approach for constrained optimal controller design of partially-unknown continuous-time nonlinear systems

In this paper, a novel iterative approximate dynamic programming scheme is proposed by introducing the learning mechanism of value iteration (VI) to solve the constrained optimal control problem for CT affine nonlinear systems with utilizing only one neural network. The idea is to show the feasibility of introducing the VI learning mechanism to solve for the constrained optimal control problem from a theoretical point of view, and thus the initial admissible control can be avoided compared with most existing works based on policy iteration (PI). Meanwhile, the initial condition of the proposed VI based method can be more general than the traditional VI method which requires the initial value function to be a zero function. A general analytical method is proposed to demonstrate the convergence property. To simplify the architecture, only one critic neural network is adopted to approximate the iterative value function while implementing the proposed method. At last, two simulation examples are proposed to validate the theoretical results.     

Knowledge-guided multi-granularity GCN for ABSA

Aspect-based sentiment analysis aims to determine sentiment polarities toward specific aspect terms within the same sentence or document. Most recent studies adopted attention-based neural network models to implicitly connect aspect terms with context words. However, these studies were limited by insufficient interaction between aspect terms and opinion words, leading to poor performance on robustness test sets. In addition, we have found that robustness test sets create new sentences that interfere with the original information of a sentence, which often makes the text too long and leads to the problem of long-distance dependence. Simultaneously, these new sentences produce more non-target aspect terms, misleading the model because of the lack of relevant knowledge guidance. This study proposes a knowledge guided multi-granularity graph convolutional neural network (KMGCN) to solve these problems. The multi-granularity attention mechanism is designed to enhance the interaction between aspect terms and opinion words. To address the long-distance dependence, KMGCN uses a graph convolutional network that relies on a semantic map based on fine-tuning pre-trained models. In particular, KMGCN uses a mask mechanism guided by conceptual knowledge to encounter more aspect terms (including target and non-target aspect terms). Experiments are conducted on 12 SemEval-2014 variant benchmarking datasets, and the results demonstrated the effectiveness of the proposed framework.     

pH值的神经网络多步预测控制算法

ｐＨ值控制过程是一个具有较强非线性、纯滞后的过程,针对ｐＨ值控制系统提出了一种基于神经网络的多步预测控制算法（ＮＭＰＣ）。神经网络用于辨识对象的预测模型,控制算法利用了神经网络的梯度信息。控制效果表明该控制系统具有较好的动态性能和较强的鲁棒性。     

基于GA-BP神经网络的扬声器音盆性能预测技术

在电声零器件制造行业中,扬声器音盆的性能好坏直接关系到扬声器产品的质量.本文提出用遗传算法(GA)优化BP神经网络模型对扬声器音盆性能进行预测.在分析了BP神经网络原理的基础上,主要阐述了如何应用遗传算法优化BP神经网络,以改进BP神经网络收敛速度慢、易陷入局部极小值的缺点.针对音盆产品特点,建立音盆性能的GA-BP神...     

基于组合神经网络的聚合物质量预测

介绍了一种将组合神经网络用于聚合物质量预测的方法．由定量数据建立的单一神经网络模型往往缺乏泛化能力，而使用组合神经网络模型则可以显著改善模型的泛化能力．由于在建立组合神经网络模型过程中，合适的组合权重对模型是否具有良好预测性能是非常重要的，因此采用了岭回归方法来选择合适的组合权重．所提出的方法已成功应用于PVC颗粒特性的预测研究中。研究结果表明，与单一神经网络模型相比，组合神经网络模型具有更佳的模型预测精度和鲁棒性．