C. L. Philip Chen

Bio: C. L. Philip Chen is an academic researcher from South China University of Technology. The author has contributed to research in topics: Nonlinear system & Computer science. The author has an hindex of 16, co-authored 115 publications receiving 740 citations. Previous affiliations of C. L. Philip Chen include Dalian Maritime University & University of Macau.

Event-Trigger-Based Finite-Time Fuzzy Adaptive Control for Stochastic Nonlinear System With Unmodeled Dynamics

Abstract: This article investigates the problem of finite-time fuzzy adaptive event-triggered control design for stochastic nonlinear nonstrict feedback systems with unmodeled dynamics. The fuzzy logic systems are adopted to identify the unknown nonlinearities and a state observer is designed to estimate the unmeasured states. Using backstepping recursive design and combining it with a varying threshold event-triggered condition, a novel event-triggered-based fuzzy adaptive finite-time control algorithm is developed, where the dynamical signal function is employed to deal with the unmodeled dynamics. A power form of the errors is used to ensure a continuous stabilizer. The semi-global finite-time stability in probability of the closed-loop system is proved based on an It $\hat{\text{o}}$ differential equation and finite-time stability theory. Simulations are provided to verify the effectiveness of the developed control algorithm.

Adaptive Neural Network Control for a Class of Nonlinear Systems With Function Constraints on States.

Abstract: In this article, the problem of tracking control for a class of nonlinear time-varying full state constrained systems is investigated. By constructing the time-varying asymmetric barrier Lyapunov function (BLF) and combining it with the backstepping algorithm, the intelligent controller and adaptive law are developed. Neural networks (NNs) are utilized to approximate the uncertain function. It is well known that in the past research of nonlinear systems with state constraints, the state constraint boundary is either a constant or a time-varying function. In this article, the constraint boundaries both related to state and time are investigated, which makes the design of control algorithm more complex and difficult. Furthermore, by employing the Lyapunov stability analysis, it is proven that all signals in the closed-loop system are bounded and the time-varying full state constraints are not violated. In the end, the effectiveness of the control algorithm is verified by numerical simulation.

A Novel Adaptive NN Prescribed Performance Control for Stochastic Nonlinear Systems

Abstract: This article investigates the problem of neural network (NN)-based adaptive backstepping control design for stochastic nonlinear systems with unmodeled dynamics in finite-time prescribed performance. NNs are used to study the uncertain control plants, and the problem of unmodeled dynamics is tackled by the combination of the changing supply function and the dynamical signal function methods. The outstanding contribution of this article is that based on the finite-time performance function (FTPF), a modified finite-time adaptive NN control design strategy is proposed, which makes the controller design simpler. Eventually, by using the Ito ’s differential lemma, the backstepping recursive design technique, and the FTPFs, a novel adaptive prescribed performance tracking control scheme is presented, which can guarantee that all the variables in the control system are bounded in probability, and the tracking error can converge to a specified performance range in the finite time. Finally, both numerical simulation and applied simulation examples are provided to verify the effectiveness and applicability of the proposed method.

Novel Efficient RNN and LSTM-Like Architectures: Recurrent and Gated Broad Learning Systems and Their Applications for Text Classification

Abstract: High accuracy of text classification can be achieved through simultaneous learning of multiple information, such as sequence information and word importance . In this article, a kind of flat neural networks called the broad learning system (BLS) is employed to derive two novel learning methods for text classification, including recurrent BLS (R-BLS) and long short-term memory (LSTM)-like architecture: gated BLS (G-BLS). The proposed two methods possess three advantages: 1) higher accuracy due to the simultaneous learning of multiple information, even compared to deep LSTM that extracts deeper but single information only; 2) significantly faster training time due to the noniterative learning in BLS, compared to LSTM; and 3) easy integration with other discriminant information for further improvement. The proposed methods have been evaluated over 13 real-world datasets from various types of text classification. From the experimental results, the proposed methods achieve higher accuracies than LSTM while taking significantly less training time on most evaluated datasets, especially when the LSTM is in deep architecture. Compared to R-BLS, G-BLS has an extra forget gate to control the flow of information (similar to LSTM) to further improve the accuracy on text classification so that G-BLS is more effective while R-BLS is more efficient.

Reduced-Order Observer-Based Dynamic Event-Triggered Adaptive NN Control for Stochastic Nonlinear Systems Subject to Unknown Input Saturation

Abstract: In this article, a dynamic event-triggered control scheme for a class of stochastic nonlinear systems with unknown input saturation and partially unmeasured states is presented. First, a dynamic event-triggered mechanism (DEM) is designed to reduce some unnecessary transmissions from controller to actuator so as to achieve better resource efficiency. Unlike most existing event-triggered mechanisms, in which the threshold parameters are always fixed, the threshold parameter in the developed event-triggered condition is dynamically adjusted according to a dynamic rule. Second, an improved neural network that considers the reconstructed error is introduced to approximate the unknown nonlinear terms existed in the considered systems. Third, an auxiliary system with the same order as the considered system is constructed to deal with the influence of asymmetric input saturation, which is distinct from most existing methods for nonlinear systems with input saturation. Assuming that the partial state is unavailable in the system, a reduced-order observer is presented to estimate them. Furthermore, it is theoretically proven that the obtained control scheme can achieve the desired objects. Finally, a one-link manipulator system and a three-degree-of-freedom ship maneuvering system are presented to illustrate the effectiveness of the proposed control method.

The sector bound approach to quantized feedback control | NOVA. The University of Newcastle's Digital Repository

Abstract: This paper studies a number of quantized feedback design problems for linear systems. We consider the case where quantizers are static (memoryless). The common aim of these design problems is to stabilize the given system or to achieve certain performance with the coarsest quantization density. Our main discovery is that the classical sector bound approach is nonconservative for studying these design problems. Consequently, we are able to convert many quantized feedback design problems to well-known robust control problems with sector bound uncertainties. In particular, we derive the coarsest quantization densities for stabilization for multiple-input-multiple-output systems in both state feedback and output feedback cases; and we also derive conditions for quantized feedback control for quadratic cost and H/sub /spl infin// performances.

Adaptive Event-Triggered SMC for Stochastic Switching Systems With Semi-Markov Process and Application to Boost Converter Circuit Model

Abstract: In this article, the sliding mode control (SMC) design is studied for a class of stochastic switching systems subject to semi-Markov process via an adaptive event-triggered mechanism. Network-induced communication constraints, semi-Markov switching parameters, and uncertain parameters are considered in a unified framework for the SMC design. Due to the constraint of measuring transducers, the system states always appear with unmeasurable characteristic. Compared with the traditional event-triggered mechanism, the adaptive event-triggered mechanism can effectively reduce the number of triggering than the static event-triggered mechanism. During the data transmission of network communication systems, network-induced delays are characterized from the event trigger to the zero-order holder. The aim of this work is to design an appropriate SMC law based on an adaptive event-triggered communication scheme such that the resulting closed-loop system could realize stochastic stability and reduce communication burden. By introducing the stochastic semi-Markov Lyapunov functional, sojourn-time-dependent sufficient conditions are established for stochastic stability. Then, a suitable SMC law is designed such that the system state can be driven onto the specified sliding surface in a finite-time region. Finally, the simulation study on boost converter circuit model (BCCM) illustrates the effectiveness of the theoretical findings.

Predicting COVID-19 in China Using Hybrid AI Model

Abstract: The coronavirus disease 2019 (COVID-19) breaking out in late December 2019 is gradually being controlled in China, but it is still spreading rapidly in many other countries and regions worldwide. It is urgent to conduct prediction research on the development and spread of the epidemic. In this article, a hybrid artificial-intelligence (AI) model is proposed for COVID-19 prediction. First, as traditional epidemic models treat all individuals with coronavirus as having the same infection rate, an improved susceptible–infected (ISI) model is proposed to estimate the variety of the infection rates for analyzing the transmission laws and development trend. Second, considering the effects of prevention and control measures and the increase of the public’s prevention awareness, the natural language processing (NLP) module and the long short-term memory (LSTM) network are embedded into the ISI model to build the hybrid AI model for COVID-19 prediction. The experimental results on the epidemic data of several typical provinces and cities in China show that individuals with coronavirus have a higher infection rate within the third to eighth days after they were infected, which is more in line with the actual transmission laws of the epidemic. Moreover, compared with the traditional epidemic models, the proposed hybrid AI model can significantly reduce the errors of the prediction results and obtain the mean absolute percentage errors (MAPEs) with 0.52%, 0.38%, 0.05%, and 0.86% for the next six days in Wuhan, Beijing, Shanghai, and countrywide, respectively.