Mapping is the off-line allocation of the tasks that represent a parallelised algorithm across a multiprocessor architecture. In this paper the target architecture is heterogeneous, where a number of computationaly di...
详细信息
Mapping is the off-line allocation of the tasks that represent a parallelised algorithm across a multiprocessor architecture. In this paper the target architecture is heterogeneous, where a number of computationaly disparate processors are integrated within a single network. This paper describes three mapping approaches that attempt to minimise the cycle time of several parallelised algorithms. The mapping algorithms have been embedded into a suite of design tools that allow a rapid translation from an application algorithm to a parallel implementation. These tools are used to explore the efficacy of the mapping algorithms. A simple heuristic is appraised first, followed by an examination of a genetic algorithm (GA) approach. Initially, the GA utilises a simple parallel architecture model. However, this leads to the embedding of the target hardware within the objective function to improve performance. Finally, the effectiveness of these approaches are examined and contrasted.
Gas turbine engines are highly non-linear, multi-sample rate, multi-input/multioutput systelus with fast changing dynamics. As a consequence a representative gas turbine engine model that comprises of the engine, its ...
详细信息
Gas turbine engines are highly non-linear, multi-sample rate, multi-input/multioutput systelus with fast changing dynamics. As a consequence a representative gas turbine engine model that comprises of the engine, its accessories and the controller, is extremely complex and interactive, incorporating look-up tables derived from real engine data and a mixture of continuous and logical functions. In this paper, a SIMULINK model of the gas turbine engine and controller is described. This model is being primarily used for generating simulated fault data that is being used to train neural networks for fault diagnosis. The current work focuses on using Resource Allocating Networks (RAN) to not only identify faults but to give an indication of the duration of the faults
In this paper a soaking pit/rolling mill process model is developed based on extended coloured petri nets (ECPNs). Both the discrete events and the continuous dynamics involved in the process are modelled in a unified...
详细信息
In this paper a soaking pit/rolling mill process model is developed based on extended coloured petri nets (ECPNs). Both the discrete events and the continuous dynamics involved in the process are modelled in a unified framework, allowing better investigation of the interactions between these two parts. Corresponding ECPN diagrams are given for the critical components involved in the soaking pit/rolling mill process, including the arrival of hot ingots, the charging and discharging of soaking pits, the soaking process, the preheat furnace operation, the mill operation, etc. The ECPN diagrams are implemented in the Design/CPN environment on a Sun SPARC workstation. Computer simulations are carried out and typical results are shown, which could lead to the improvement of soaking pit/rolling mill operations
In this paper, a novel system is proposed to obtain global noise control within an enclosure. Instead of using the traditional least squares method for the controller design, the emerging technology of Genetic Algorit...
详细信息
In this paper, a novel system is proposed to obtain global noise control within an enclosure. Instead of using the traditional least squares method for the controller design, the emerging technology of Genetic Algorithms is applied for the design of an optimal multi-channels active noise control system. The advantage of this system is that it has the ability to tackle a multi-objectives error function problem so that a global noise reduction is achieved. Experimental result has illustrated that it is effective for achieving global quietness for an enclosed environment.
This paper develops a robust identification algorithm for nonlinear systems. Multi-objective criteria are defined that takes into account the diverse distributions of the unknown disturbance and noise terms. The L 2 ,...
详细信息
This paper develops a robust identification algorithm for nonlinear systems. Multi-objective criteria are defined that takes into account the diverse distributions of the unknown disturbance and noise terms. The L 2 , L ∞ -norms and a model complexity measure are the three chosen in the development of the algorithm. The algorithm combines the method of inequalities and genetic algorithms with that of least squares in optimising the multi-objective criteria. Model selection is also introduced as part of a single identification algorithm, where genetic search is carried out over models with all possible combinations from a given set of basis functions.
The Dynamical Hierarchical Optimization implements either Interaction Balance Method or Interaction Prediction Method for the coordination of local decision units; both methods are initial condition dependent and requ...
详细信息
The Dynamical Hierarchical Optimization implements either Interaction Balance Method or Interaction Prediction Method for the coordination of local decision units; both methods are initial condition dependent and require that the interconnection error reaches zero before implementing the solution to the real system which is numerically difficult; by adding a repetitive loop with a chosen control horizon the solution can be significantly improved
The Digital Wave Processor, DWP [1-10], is an example of a dependable distributed system which has been built and will be flown on the ESA/NASA Cluster satellites. Pre-flight testing of the DWP instruments is nearly c...
详细信息
The Development Framework provides a highly automatic translation from a specification to an implementation. The specification is in a popular, graphical controlengineering notation typically representing a system wi...
详细信息
The Development Framework provides a highly automatic translation from a specification to an implementation. The specification is in a popular, graphical controlengineering notation typically representing a system with stringent reliability requirements and hard real time constraints. An interface has been constructed between the Development Framework and the commercially available dependability modelling tool, SURF-2. This tool is designed to support an evaluation based design approach. Multiple design solutions can be compared to assess the implications of design decisions on the dependability of the system under development. The software demonstration will show how the interface between the Development Framework and SURF-2 is used to model the inclusion of selected fault tolerant mechanisms in the system under development.< >
This paper investigates a control strategy which uses an identified reference model to produce appropriate chaotic and nonchaotic reference trajectories. Although it is possible to dispense with the reference model in...
详细信息
This paper investigates a control strategy which uses an identified reference model to produce appropriate chaotic and nonchaotic reference trajectories. Although it is possible to dispense with the reference model in many applications, it is often desirable to obtain a model of the system in order to guarantee that the reference signal is a genuine trajectory of the system to be controlled and to reduce the effects of transients and the control effort. The influence of measurement noise and the relation between the control effort and the dynamics of the uncontrolled system are also discussed. Examples which use the Duffing-Ueda system and a discrete reference model identified from input/output records with no a priori information, are included to illustrate the main points of the paper.
暂无评论