Weihua Cao

Intelligent Decoupling Control of Gas Collection Process of Multiple Asymmetric Coke Ovens

This paper presents a hierarchical intelligent decoupling control method for the gas collection process of three asymmetric coke ovens of an iron and steel company. The process features strong asymmetric coupling, large disturbances, and strong nonlinearity. These characteristics are exploited in the following steps to implement intelligent decoupling control. First, three fuzzy expert controllers and two feedforward controllers stabilize the pressure of each gas collector. Then, the three collectors are divided into two groups based on an analysis of the couplings among them to facilitate decoupling control. An intelligent decoupling control method is used to design intragroup and intergroup decoupling controllers. Finally, expert valve control handles the nonlinearity of butterfly valves, and a multirate sampling strategy further improves the control performance. The results of actual runs show that the system satisfactorily performs the intelligent decoupling control of the gas pressures of the coke ovens, sufficiently suppresses disturbances, and accurately stabilizes the pressures.

Aperiodic Disturbance Rejection in Repetitive-Control Systems

A repetitive control system (RCS) provides good control performance for periodic signals, but it may not satisfactorily reject aperiodic disturbances. Our solution is an equivalent-input-disturbance (EID)-based RCS in which aperiodic disturbances are estimated by an EID estimator and rejected by incorporation of the estimate into the repetitive control law. A stability criterion and a design algorithm have been developed, and the validity of the method has been demonstrated through simulations and experiments on a rotational speed control system. A comparison of the disturbance rejection performance of this method, proportional-integral-differential control, conventional repetitive control, and H∞ repetitive control demonstrates the superiority of this method.

Stable Control Strategy for Planar Three-Link Underactuated Mechanical System

A stable-control strategy for a planar three-link passive-active-active underactuated mechanical system not subject to gravity constraints has been devised. The control objective is to move the end effector from any initial position to any target position. First, a dynamic model of the system is built, and its properties are analyzed. Next, based on the complete integrability of an underactuated planar acrobot (UPA), the control of a planar three-link system is divided into two stages. In each stage, keeping the angle of one active link constant reduces the planar three-link system to a UPA, thereby enabling the use of quadrature to obtain the angle constraint relationships between the passive and active links. Then, the target angles associated with the target position are calculated by particle swarm optimization based on the angle constraint relationships. Finally, a controller for each stage is designed, ensuring that the control objective will be reached. Simulation results demonstrate the validity of this control method.

Disturbance rejection in nonlinear systems based on equivalent-input-disturbance approach

This paper presents a new system configuration and a design method that improves disturbance rejection performance for a nonlinear system. The equivalent-input-disturbance (EID) approach is used to construct an EID estimator that estimates the influence of exogenous disturbances and nonlinearities on the output of the system. Sufficient stability conditions for state- and output-feedback control are derived in terms of linear matrix inequalities. New EID-based control laws that combines an EID estimate with a state- or an output-feedback control laws ensure good control performance. A numerical example illustrates the design method. A comparison between the EID-based control, the conventional disturbance observer, the disturbance-observer-based-control, and the sliding mode control methods demonstrates the validity and superiority of the EID-based control method.

Integrated Intelligent Control of Gas Mixing-and-Pressurization Process

This paper presents an integrated intelligent control method for a gas mixing-and-pressurization process with the objective of controlling the calorific value and pressure of the mixed gas. The control system contains two subsystems: a calorific-value-and-pressure-decoupling (CVPD) control subsystem and a pressurization control subsystem. The CVPD control subsystem takes the calorific value as the primary control target and keeps the pressure of prepressurized mixed gas within a given range. It employs fuzzy and expert control, fuzzy decoupling control, and the concept of relative gain matrix to keep the calorific value constant at a specific value, suppress the influence of pressure, and intelligently adjust the amounts that four butterfly valves are open. The pressurization control subsystem regulates the converters of the pressurization machine through feedforward and feedback expert controllers so as to make the pressure of the mixed gas track a given value. Runs on an actual system show that this control strategy is effective and keeps the fluctuations in calorific value and pressure at a low level. This reduces production costs and improves product quality.

An intelligent control system based on prediction of the burn-through point for the sintering process of an iron and steel plant

This paper concerns an intelligent control system based on prediction of the burn-through point (BTP) of the sintering process of an iron and steel plant. The system has a two-level hierarchical configuration: intelligent-control level and basic-automation level. At the intelligent-control level, first, a BTP prediction model is derived using an intelligent, integrated modeling method based on grey theory and back-propagation neural networks. Next, a hybrid fuzzy-predictive controller for the BTP is established using fuzzy control, predictive control, and a flexible switching control strategy. Finally, an intelligent coordinating control algorithm based on the satisfactory solution principle is employed to coordinate BTP control and bunker-level control. Then, a satisfactory sinter strand velocity is obtained and used as the target value. The basic-automation level regulates the speed of the motor driving the strand so as to make the strand velocity track the target value. The results of actual runs show that the system adequately suppresses the variation in BTP, increases the quantity and quality of sintering agglomerate, and ensures process safety.

Design and implementation of a new third order chaotic system

This paper designs a new three-dimensional chaotic system with a simple structure and only a nonlinear term, it contains three design parameters. We demonstrate the system is chaotic theoretically by the stability analysis of the equilibrium point, Lyapunov index calculation, and bifurcation phenomenon of the system when design parameter changes. Finally, numerical simulation shows that a typical chaotic attractor happens in this system by using MATLAB software, and the period-doubling bifurcation phenomenon is observed according to the bifurcation diagram. The simulation results prove that the system possesses the characteristics of chaos.

Speech emotion recognition based on feature selection and extreme learning machine decision tree

Abstract Feature selection is a crucial step in the development of a system for identifying emotions in speech. Recently, the interaction between features generated from the same audio source was rarely considered, which may produce redundant features and increase the computational costs. To solve this problem, feature selection method based on correlation analysis and Fisher is proposed, which can remove the redundant features that have close correlations with each other. To improve the recognition performance of the feature subset after proposal feature selection further, an emotion recognition method based on extreme learning machine (ELM) decision tree is proposed according to the confusion degree among different basic emotions. A framework of speech emotion recognition is proposed and the classification experiments based on proposed classification method by using Chinese speech database from institute of automation of Chinese academy of sciences (CASIA) are performed. And the experimental results show that the proposal achieved 89.6% recognition rate on average. By proposal, it would be fast and efficient to discriminate emotional states of different speakers from speech, and it would make it possible to realize the interaction between speaker-independent and computer/robot in the future.

Model-based learning with Bayesian and MAXQ value function decomposition for hierarchical task

How to improve efficiency of learning is always the key issue for implementation of reinforcement learning. This paper makes use of advantages of both hierarchical learning and model-based learning, so that an improved algorithm, named Bayesian-MAXQ learning, is introduced, in which several modifications are developed to solve the value update of hierarchy, while the possible performance damages brought by prioritized sweeping is reduced to trivial. The simulation results show that, Bayesian-MAXQ learning performs with high efficiency, and it can serve as a good framework for further study on hierarchical model-based learning.

A new bat algorithm based on iterative local search and stochastic inertia weight

Abstract Bat algorithm (BA) is a heuristic optimization algorithm based on swarm intelligence and the inspiration from the nature behavior of bats. It has some advantages including fast solving speed, high precision and only few parameters need to be adjusted. However, BA is easy to fall into local optima and has unstable optimization results due to low global exploration ability. In order to overcome these weakness, a new bat algorithm based on iterative local search and stochastic inertia weight (ILSSIWBA) is proposed in this paper. A kind of local search algorithm, called iterative local search (ILS) is introduced into the proposed algorithm. The ILS algorithm disturbs the local optimum and do some local re-search, so that the ILSSIWBA has strong ability to jump out of the local optima. In addition, a weight updating method, called stochastic inertia weight (SIW) is also introduced into the proposed algorithm. Considering the SIW in the velocity updating equation can enhance the diversity and flexibility of bat population, so that the ILSSIWBA has stable optimization results. Meanwhile, the pulse rate and loudness are modified to enhance the balance performance between global and local search. Moreover, the global convergence of ILSSIWBA is proved by the convergence criteria of stochastic algorithm. In the end, the ILSSIWBA is compared with directional bat algorithm (DBA) and other algorithms on 10 classic benchmark functions, CEC 2005 benchmark suite, and two real-world problems. The results show that ILSSIWBA has remarkable advantages in optimization accuracy, solving speed and convergence stability. This algorithm lays a solid foundation for solving modeling, optimization and control problems of complex systems.