The proceedings contains three papers from the 1998 ieecolloquium entitled `Industrial Automation and control: applications in the automotive Industry'. Topics discussed include: road prediction for intelligent v...
详细信息
The proceedings contains three papers from the 1998 ieecolloquium entitled `Industrial Automation and control: applications in the automotive Industry'. Topics discussed include: road prediction for intelligent vehicles using video;advanced collision warning systems;and fuzzy modeling techniques applied to an air/fuel ratio control system.
A precursor to control is modelling, for nonlinear, uncertain, timevarying, unknown dynamical processes, neurofuzzy algorithms(1) have many useful properties including, convergence in learning, real time adaptability,...
详细信息
A precursor to control is modelling, for nonlinear, uncertain, timevarying, unknown dynamical processes, neurofuzzy algorithms(1) have many useful properties including, convergence in learning, real time adaptability, and transparency. Unfortunately, they suffer from the curse of dimensionality, and recent research(5) on parsimonious modelling schemas such as LOLIMOT, ASMOD, MASMOD, MENN, have shown how this problem can be effectively overcome. Equally, this method of databased modelling coupled with local operating point models(2) enables classical linear control and estimation algorithms to be applied directly to these processes. In this seminar the basic theory of intelligent modelling via Neurofuzzy Algorithms(1) will be developed, local models(2), local controllers(3,4), intelligent estimators(4,5) and applications in modelling, control and estimation (tracking) for advanced transportation(6,7) will be used to illustrate the basic principles. The talk will be supported by a series of practical demonstrations via videos.
Due largely to the need for vehicle manufacturers to meet requirements set out in stringent vehicle exhaust emission legislation, control of the air/fuel (A/F) ratio is an active area of research within the automotive...
详细信息
Due largely to the need for vehicle manufacturers to meet requirements set out in stringent vehicle exhaust emission legislation, control of the air/fuel (A/F) ratio is an active area of research within the automotive industry. In this context, fuzzy modelling techniques have been applied to the simulated control of the A/F ratio. The method identifies structured nonlinear models, due to the existence of the G and H matrices, that could be readily adapted on-line. The model, used within the control scheme, has been selected using the YIC, which is used to assess the appropriateness of normally linear models, for control. The performance of the model, as part of the combined feedforward/feedback controller, appears satisfactory.
Fuzzy modelling techniques have been applied to the simulated control of the air/fuel ratio. The method enables the identification of structured nonlinear models, due to the existence of the G and H matrices, that cou...
Fuzzy modelling techniques have been applied to the simulated control of the air/fuel ratio. The method enables the identification of structured nonlinear models, due to the existence of the G and H matrices, that could be readily adapted online. The model used within the control scheme has been selected using the Young information criteria, which is used to assess its appropriateness, normally linear models, for control. The performance of the model, as part of the combined feedforward/feedback controller, appears satisfactory. When the nonlinear and time varying nature of the automotive engine system is considered, the fuzzy modelling methods are considered to offer some potential for engine controlapplications.
Summary form only given. A precursor to control is modelling. For nonlinear, uncertain, time-varying, unknown dynamical processes, neurofuzzy algorithms have many useful properties including convergence in learning, r...
Summary form only given. A precursor to control is modelling. For nonlinear, uncertain, time-varying, unknown dynamical processes, neurofuzzy algorithms have many useful properties including convergence in learning, real-time adaptability, and transparency. Unfortunately, they suffer from the curse of dimensionality, and recent research on parsimonious modelling schemas such as LOLIMOT, ASMOD, MASMOD, MENN, have shown how this problem can be effectively overcome. Equally, this method of data-based modelling coupled with local operating point models enables classical linear control and estimation algorithms to be applied directly to these processes. In this seminar the basic theory of intelligent modelling via neurofuzzy algorithms will be developed, and local models, local controllers, intelligent estimators and applications in modelling, control and estimation (tracking) for advanced transportation will be used to illustrate the basic principles.
Researchers in the artificial intelligence community view system identification as a training task, while those with a control background see it as a parameter estimation problem. A third and more general perspective ...
Researchers in the artificial intelligence community view system identification as a training task, while those with a control background see it as a parameter estimation problem. A third and more general perspective is to view it as an optimization problem in which a performance index is minimised with respect to the parameters being identified. While these diverse interpretations result in differing terminologies and representations, the algorithms involved are essentially equivalent. Here the optimization perspective will be adopted. From this perspective neural modelling structures (NARX or NARMAX) can be classified as linear, nonlinear or mixed linear-nonlinear (hybrid) in the parameters. Linear, nonlinear or hybrid optimization techniques are then used for identification.
While of undoubted value for nonlinear identification and control of dynamic systems, neural networks have a number of limitations for practical applications. Thus, in online training, due consideration must be given ...
While of undoubted value for nonlinear identification and control of dynamic systems, neural networks have a number of limitations for practical applications. Thus, in online training, due consideration must be given to the necessity for regularisation with noisy data and to the choice of network architecture. More fundamentally, the nontransparent black-box nature of neural models make it difficult to include a priori system information, and to interpret the final structure meaningfully in terms of physical process characteristics. Neural approaches also fail to exploit the significant body of theoretical results available for conventional model-based control, making it difficult to analyse the closed-loop behaviour in terms of stability and robustness. The aim of this paper is to describe a nonlinear modelling architecture, called the local model network (LMN), which introduces transparency while offering distinct advantages for nonlinear model-based control. Simulation results for a pH neutralisation process are used to illustrate the performance benefits of LMNs for nonlinear dynamic matrix control (DMC) and for nonlinear internal model control (IMC).
The application of a rapid prototyping environment for strategy development work carried out by the advanced In-line Gasoline and Diesel Engineering (AIGDE) group within the Ford Motor Company is presented. The use of...
详细信息
The application of a rapid prototyping environment for strategy development work carried out by the advanced In-line Gasoline and Diesel Engineering (AIGDE) group within the Ford Motor Company is presented. The use of rapid prototyping tools allows the abstraction of strategy development to the appropriate level for control system design. Automatic code generation binds this control system development environment to dSPACE and AIGDE custom hardware. Manifold air-charge and fuel flow modeling are example applications developed using rapid prototyping environment. Accurate estimation of both air and fuel flows is necessary to maintain the desired air-fuel ratio during transients.
This paper describes the application of a rapid prototyping environment for strategy development work carried out by the AIGDE group within Ford Motor Company. The system is comprised of dSPACE real time hardware, cus...
This paper describes the application of a rapid prototyping environment for strategy development work carried out by the AIGDE group within Ford Motor Company. The system is comprised of dSPACE real time hardware, custom interfacing and Mathworks' MatlabTM software suite. The use of rapid prototyping tools allows the abstraction of strategy development to the appropriate level for control system design. Automatic code generation binds this control system development environment to dSPACE and AIGDE custom hardware. The latter allows flexibility in the choice of sensors and actuators used for development. Manifold air-charge and fuel-flow modelling are example applications developed using the rapid prototyping environment.
作者：
K.M. BossleyM. BrownS.R. GunnC.J. HarrisImage
Speech and Intelligent Systems Research Group Department of Electronics and Computer Science University of Southampton Southampton UK
Often the quality of the available numerical and linguistic knowledge conventionally used to identify neurofuzzy systems is poor. This problem is overcome by the use of advanced model identification algorithms present...
详细信息
Often the quality of the available numerical and linguistic knowledge conventionally used to identify neurofuzzy systems is poor. This problem is overcome by the use of advanced model identification algorithms presented in this paper. Parsimonious models are identified via data-driven construction algorithms which match the structure to the data and allow the application of neurofuzzy modelling to high-dimensional real world problems. However, the inherent structure of neurofuzzy models can produce redundant degrees of freedom which are poorly identified by the data. As a solution to this problem Bayesian regularisation is applied to these models, smoothing out any irregularities in the structure, hence controlling unidentified rules. This is important in control and system identification scenarios where data may only be gathered around a collection of operating points.
暂无评论