The Set Pair Grey Clustering Evaluation Of The Human-Machine Interface Of The Master Control Room In Special Steel Factory

The dissertation is concerned about the evaluation of "human-hardware" interface. The research work on human-machine interface is mainly focused on the display, controller in machines, which communicate with human beings, and the entities associated with human's operations, in order to raise the work efficiency and reduce the misoperation.At present, there are many ways that can be used in evaluation of the human-machine interface, for example, the hierarchy analysis method, fuzzy mathematics method, grey co-relation analysis method and so on. Unfortunately, there exists the uncertainty in evaluations in all these methods, such as the fuzzy in the result of the evaluation. In this dissertation, the author proposes the grey cluster analysis based on set pair method in human-machine evaluation. The method organically fuses the set pair analysis and grey fixed-weight cluster method, effectively avoiding the defects in evaluation, for instances, the over-fuzzy evaluation result and the non-obvious disparity. Therefore the evaluation result is more scientific and richer in hierarchy.The main research works in the dissertation includes the following aspects:1. On the basis of the preliminary research work of the research group, the evaluation index system of human-machine interface is established for the display and control system. The subjective and objective evaluation index and standard in human-machine engineering is determined for the various units on the human-machine interface, such as displays, operation units and control panel. The index and standard are the foundation of the algorithm research.2. After the analysis of the various evaluation methods, the evaluation model with grey cluster model based on set pair is put forward. The relevant evaluation software is compiled with C# language running on Visual Studio 2005.3. Virtual reality scenes are created for the display and control system using CATIA and Kingview software in order to collect the data required for evaluation.4. The evaluation is made for the "human-hardware" interface of No.4 machine group of special steel Co.Ltd, Benxi Steel Group by use of the human-machine evaluation software proposed in this dissertation. The comparison is performed between the result with GCMBSP and that with neural network method. The research result shows that the GCMBSP method is well suitable for the evaluation of human- machine interface.