Comparison of fuzzy control systems for hypothermal brain temperature regulation

The automatic control system for brain tissue temperature is studied theoretically for brain hypothermia treatment. In order to realize a human-friendly control mechanism, an automatic temperature regulation system is constructed to simulate brain hypothermia treatment by introducing a fuzzy algorithm for possible characteristic changes in patients. The brain temperature model is successfully realized to follow the desired temperature course automatically. The model reference fuzzy control of brain temperature based on water-cooling blankets is verified for clinical application to brain hypothermia treatments through various kinds of simulation experiment. Partly presented at the 16th IFAC World Congress, Prague, July 2005     

Automatic optimal-adaptive air-cooling system for brain hypothermia treatment

A new automatic air-cooling system is proposed using a cooling incubator to replace the manual water-cooling blanket which has traditionally been used to lower brain tissue temperature (BTT) in brain hypothermia treatment (BHT). This study concerns its feasibility through a simulation. First, a biothermal model is proposed for the adult incubating system based on the geometric structure and parameters of patients. Its dynamics were carefully examined by two simulation experiments testing its step response and feedback control. Then a model reference adaptive control algorithm was introduced for the automatic regulation of BTT, where the newly developed adult incubating biothermal model, represented by a state equation, was replaced for the hypothermic patient with a cooling blanket, thus introducing a first-order lag system given as its basic characteristic model. Finally, the proposed cooling incubator was controlled by the adaptive control mechanism, which gives a follow-up of BTT to a given reference temperature course, even if a possible environmental change in the therapeutic cooling system exists, including the individual differences of patients and any chronic conditional change. The automatic cooling incubating system based on the air-cooling method was confirmed to be superior to the water-cooling one. Thus, this work supports the possible development of an air-cooling adult incubating system for the automatic regulation of BTT in an intensive care unit (ICU) application.     

Development of a basic air-cooling fuzzy control system for hypothermia

Surface cooling by an automatic control system using a water blanket ensures a bloodless method of accurate brain hypothermia for cerebral protection at various stages of medical treatment. The control of temperature without any effects from the outside environment means that medical staff have less need to use their special skills for the exact management of brain temperature in clinical application, although the method is based on an easily handled surface water-cooling system. However, water is not always an appropriate medium for clinical cooling, from the viewpoint of both patients and the medical staff, because of the maintenance of the system in an intensive care unit (ICU). In order to cope with such difficulties, air-cooling is proposed based on various simulation results. Here, the aim is to realize automatic temperature control based on a fuzzy algorithm as a clinical reflection of the medical staff’s knowledge and experience. Experiments were carried out using a mannequin with human thermal characteristics in order to verify the utility of the basic automatic air-cooling control system for the treatment of brain hypothermia.     

Reduced-order modeling of human body for brain hypothermia treatment

Brain hypothermia treatment (BHT) is an active therapy for severe brain injury. It makes the temperature of the brain track a given temperature input curve so as to reduce the risk of tissue damage. BHT requires a brain-temperature control system because of environmental disturbances and changes in the human body. The thermal models of the human body devised so far are usually of a very high order and are not suitable for controlling brain temperature. This paper presents a method of finding a reducedorder thermal model of the human body for use in BHT. It combines minimal realization and balanced realization. Unlike other methods, this method yields a reduced-order model that is based on system theory and that takes the frequency characteristics of human thermal sensation into account. It features high precision in the frequency band for BHT and is suitable for the control of brain temperature.     

基于自适应联合字典学习的脑部多模态图像融合方法

针对目前全局训练字典对于脑部医学图像的自适应性不强,以及使用稀疏表示系数的L₁范数取极大的融合方式易造成图像的灰度不连续效应进而导致图像融合效果欠佳的问题,提出一种基于自适应联合字典学习的脑部多模态图像融合方法。该方法首先使用改进的K奇异值分解（K-SVD）算法自适应地从已配准的源图像中学习得到子字典并组合成自适应联合字典,在自适应联合字典的作用下由系数重用正交匹配追踪（CoefROMP）算法计算得到稀疏表示系数;然后将稀疏表示系数的"多范数"作为源图像块的活跃度测量,并提出"自适应加权平均"与"选择最大"相结合的无偏规则,根据稀疏表示系数的"多范数"的相似度选择融合规则,当"多范数"的相似度大于阈值时,使用"自适应加权平均"的规则,反之则使用"选择最大"的规则融合稀疏表示系数;最后根据融合系数与自适应联合字典重构融合图像。实验结果表明,与其他三种基于多尺度变换的方法和五种基于稀疏表示的方法相比,所提方法的融合图像能够保留更多的图像细节信息,对比度和清晰度较好,病灶边缘清晰,客观参数标准差、空间频率、互信息、基于梯度指标、基于通用图像质量指标和平均结构相似指标在三组实验条件下的均值分别为：71.0783、21.9708、3.6790、0.6603、0.7352和0.7339。该方法可以应用于临床诊断和辅助治疗。     

An Adaptive Automated Robotic Task-Practice System for Rehabilitation of Arm Functions After Stroke

We present a novel robotic task-practice system, i.e., adaptive and automatic presentation of tasks (ADAPT), which is designed to enhance the recovery of upper extremity functions in patients with stroke. We designed ADAPT in accordance with current training guidelines for stroke rehabilitation; ADAPT engages the patient intensively, actively, and adaptively in a variety of realistic functional tasks that require reaching and manipulation. A general-purpose robot simulates the dynamics of the functional tasks and presents these functional tasks to the patient. A novel tool-changing system enables ADAPT to automatically switch between the tools corresponding to the functional tasks. The control architecture of ADAPT is composed of three main components: a high-level task scheduler, a functional task model, and a low-level admittance controller. The high-level task scheduler adaptively selects the task to practice and sets the task difficulty based on the previous performance of the patients. The functional task model generates desired trajectories based on learned models of task dynamics. Tasks dynamics are modeled with receptive field weighted regression (RFWR), such that the feel of the task tools is accurately modeled, and the task difficulty can be easily adjusted. The low-level admittance controller, which is also learned with RFWR, implements the selected task trajectory for robot–patient interaction. The results of a preliminary experiment with a healthy subject demonstrate the successful operation of ADAPT.      

Indirect adaptive nonlinear control of drug delivery systems

This paper investigates the use of adaptive neural network techniques for modeling and automatic control of mean arterial pressure through the intravenous infusion of sodium nitroprusside. An indirect model reference-based adaptive nonlinear control scheme with neural networks approximating the unknown nonlinearities, is developed. In this formulation nonlinear estimators are used to adaptively approximate the system uncertainty and augment the linear control law for improved performance. The overall design is based on self-tuning the controller to the specific response characteristics of individual patients. Computer simulations illustrate the ability of radial basis function networks to model the unknown nonlinearities and improve the closed-loop system characteristics     

ADAPTIVE DYNAMIC MATRIX CONTROL FOR NETWORK‐BASED CONTROL SYSTEMS WITH RANDOM DELAYS

Random transfer delays in network‐based control systems (NCSs) degrade the control performance and can even destabilize the control system. To address this problem, the adaptive dynamic matrix control (DMC) algorithm is proposed. The control algorithm is derived by applying the philosophy behind DMC to a discrete time‐delay model. A method to estimate the network‐induced delays is also presented to facilitate implementation of the control algorithm. Finally, an NCS platform based on the TrueTime simulator is constructed. With it, the adaptive DMC algorithm is compared with the conventional DMC algorithm under different network conditions. Simulation results show that the proposed adaptive DMC algorithm can respond to various network conditions adaptively and achieve better control performance for NCSs with random transfer delays.     

磁悬浮球系统的非线性自适应控制方法

针对磁悬浮球系统被控对象变化时控制器自适应问题,提出了一种反馈线性化和在线参数辨识相结合的非线性自适应控制方法。基于状态反馈精确线性化方法建立磁悬浮球系统的数学模型,通过状态反馈设计了一种非线性控制器,并给出了控制器参数的在线辨识方法。MATLAB平台上在线实验结果表明,与反演滑模自适应控制方法相比,提出的方法无须在平衡位置近似线性化,可以在平衡位置实现对不同对象的自适应控制,且具有理想的稳态调节性能。     

A stochastic rudder control law for ship path-following autopilots

A heuristic stochastic rudder control law for ship automatic path-following in restricted waters is presented. The objective is to make feasible an automatic pilot with an adaptive capacity of operating in situations where there is a great lack of knowledge of ship dynamics. State estimation and feedback control determination are separately treated. The extended Kalman filter is combined with a dynamical model compensation technique and a state noise adjustment procedure, in a situation where the observations have a high local level of information. An adaptive scheme results where unmodeled effects are estimated together with the system state. A feedback control law that uses state estimation information and steering conditions is proposed. In each discretization interval it takes a stepwise approximation for the control, as the solution of a parameter estimation problem. The performance of the resulting state estimator and controller is illustrated by results obtained through digital simulation. The test case selected is that of ship automatically steering in a channel under severe geometric and environmental conditions.     

基于L1自适应着舰纵向控制与特性分析

考虑到舰载机着舰在最后阶段受到舰尾气流的不利影响。因此,为了提高舰载机着舰控制系统的鲁棒性,提出一种基于L1自适应控制的纵向着舰控制律设计方法。首先,对舰尾流进行分类并将其转化为控制系统的干扰模型,通过分析舰尾气流特性来构建与之相对应的L1自适应控制律结构;然后,基于L1自适应控制方法设计纵向自动着舰控制律,其中包括设计状态观测器、L1自适应控制律、低通滤波器和自适应律,并且对设计好的系统进行了稳定性分析;最后,讨论不同尺度风速的舰尾流对自动着舰系统的影响规律,并且采用“单因素方差分析法”对不同舰尾流环境下的着舰点进行分析。仿真结果表明不同风速下的舰尾流对着舰精度存在一定影响规律,并且所设计基于L1自适应纵向着舰控制律具有较强的自适应性和抗干扰能力。     

热电偶检定炉炉温模型参考自适应控制的研究

热电偶校检定炉在对热偶进行高温点的校验时,炉温难于稳定,主要原因在于控高温时加热时间较长环境温度产生较大变化,从而造成检定炉加热模型产生时变,导致控制器效果明显变差,本文利用模型参考自适应控制方法在线调整PID控制器参数以适应环境温度变化而造成的加热炉模型时变,试验结果表明该方法适应性良好,从而为热电偶检定炉的温度控制提供了新的方法。     

Adaptive control of a class of nonlinear systems with convex/concave parameterization

This paper deals with adaptive control of a class of second-order nonlinear systems with a triangular structure and convex/concave parameterization. In Annaswamy et al. (Automatica 33(11) (1998) 1975–1995) it was shown that nonlinearly parameterized systems that satisfy certain matching conditions can be adaptively controlled in a stable manner. In this paper, we relax these matching conditions and include additional dynamics between the nonlinearities and the control input. Global boundedness and convergence to within a desired precision is established. No overparameterization of the adaptive controller is required.     

水泥篦冷机出口熟料温度自适应辨识模型

为了优化控制系统,建立篦冷机温度熟料出口的识别模型,利用篦冷机内熟料换热机理,找出熟料冷却过程的关键影响因素;利用回声状态网络辨识篦冷机运行数据,基于递归最小二乘法推导网络的在线学习算法,实现权值自适应调整,从而建立了篦冷机出口熟料温度的自适应辨识模型.仿真实验可知,在系统发生变化时构建的模型能够自适应调整网络输出权值矩阵,使模型快速收敛.与其他离线方法相比,提出的熟料出口温度的自适应模型更加持久有效,可以作为辨识模型指导篦冷机的控制.     

Automated adaptive cardiovascular flow simulations

We present an automatic adaptive procedure to perform blood flow simulations in the cardiovascular system. The procedure allows the user to start with subject-specific data collected through clinical measurements, like magnetic resonance imaging (MRI) data, and evaluate physiological parameters of interest, like flow distribution, pressure variations, wall shear stress, in an automatic and efficient manner. The process involves construction of geometric models of blood vessels, specification of flow conditions and application of an adaptive flow solver. The latter is based on incompressible Navier–Stokes equations using adaptive spatial discretization (meshing) techniques. In this article, we demonstrate the method on a model of a human abdominal aorta of a normal subject with geometry and flow rates assimilated from MRI data. The results obtained show that boundary layer mesh adaptivity offers a better alternative leading to more accurate predictions, especially for key physiological quantities like wall shear stress.     

Adaptive learning control for thermal error compensation of 5-axis machine tools

The research presented in this paper shows an adaptive approach for long-term thermal error compensation of 5-axis machine tools (MT). A system of differential equations is used to compute the model based compensation values. The model can predict thermal displacements of the tool center point (TCP) based on changes in the environmental temperature, load-dependent changes and boundary condition changes and states, like machining with or without cutting fluid. The model based compensation of the rotary axis of a 5-axis MT is then extended by on-machine measurements. The information gained by the process-intermittent probing is used to adaptively update the model parameters, so that the model learns how to predict thermal position and orientation errors and to maintain a small residual error of the thermally induced errors of the rotary axis over a long time. This approach not only increases the MT accuracy but also reduces the amount of time spent on preproduction model parameter identification. Additionally an algorithm has been developed to dynamically adjust the length of the on-machine measurement intervals to maintain a high productivity and a constant deviation of the machined parts.Experimental results confirm that the adaptive learning control (ALC) for thermal errors shows a desirable long-term prediction accuracy.     

Modeling and control of an LDPE autoclave reactor

A two-compartment four-cell model is developed for the adiabetic slim type autoclave reactor for free radical polymerization of low density polyethylene (LDPE). It is possible to determine not only the reactor performance represented by the monomer conversion and the reaction temperature but also the properties of the polymer product characterized by the average molecular weight and the polydispersity. It turns out that the reactor performance predicted is in good agreement with the plant data and the properties of the polymer product are estimated within reasonable ranges of actual values. The steady state multiplicity is found to exist and is examined by constructing the bifurcation diagram. The effects of various operation parameters on the reactor performance and the polymer properties are investigated systematically to show that the temperature distribution plays the central role for the properties of the polymer products. Therefore, it is essential to establish a good control strategy for the temperature in each compartment. The adaptive pole-placement control algorithm is applied to the temperature control of the adiabatic slim type autoclave reactor. The recursive least square method is used for the model identification. To accomplish a satisfactory control, the estimator and controller are initialized during the period of start-up. It is shown that the reactor system can be adaptively controlled by the pole-placement control algorithm, especially when the reactor temperature distribution is changed.     

Continuous process control using neural networks

We present some adaptive control strategies based on neural networks that can be used for designing controllers for continuous process control problems. Specifically, a learning algorithm has been formulated based on reinforcement learning, a weakly supervised learning technique, to solve set-point control and control scheduling for continuous processes where the process cannot be modeled easily. It is shown how reinforcement learning can be used to learn the control strategy adaptively based on exploration of the control space without making assumptions about the process model. A new learning scheme, handicapped learning, was developed to learn a control schedule that specifies a schedule of set points. Applications studied include the control of a nonisothermal continuously stirred tank reactor at its unstable state and the learning of the daily time-temperature schedule for an environment controller. Experimental results demonstrate good learning performance, indicating that the learning algorithm can be used for solving transient startup and boundary value control problems.     

含齿隙弹载舵机的全局反步模糊自适应控制

针对制导弹药电动舵机伺服系统中存在的齿隙、不确定参数及外部干扰,提出一种基于反步法的全局模糊自适应控制方法.首先,综合考虑上述非线性因素,分析描述齿隙的近似死区模型,建立含齿隙弹载舵机的双惯量机电模型,并将其引入采用"三闭环"结构的伺服系统中,构建非线性系统的状态空间;然后,采用模糊逻辑系统对齿隙等非线性因素进行自适应逼近与补偿控制,通过反步递推构造全系统Lyapunov函数,并运用Lyapunov第二法证明了整个闭环系统最终一致有界.仿真实验表明:较经典PID控制,该方法能更有效地削弱齿隙引起的传动力矩抖振与振荡冲击,既保证了系统的跟踪速度与控制精度,对不确定参数与外部干扰也具有较强的鲁棒性.     

T-S model based indirect adaptive fuzzy control using online parameter estimation

A parameter estimation scheme with an appropriate adaptive law for updating the parameters is designed and analyzed based on the Lyapunov theory for the general MIMO Takagi-Sugeno (T-S) fuzzy models. The parameters of the Takagi-Sugeno fuzzy models can be estimated by observing the behavior of the system and with the online parameter estimator, any type of fuzzy controllers works adaptively to the parameter perturbation. In order to show the applicability of the proposed estimator, an existing fuzzy state feedback controller is adopted and indirect adaptive fuzzy control design with the proposed estimator is shown. From the numerical simulations and experiments, it is shown that the derived adaptive law works for the estimation model to follows the parameterized plant model and the overall control system has robustness to the parameter perturbation.