Research Of Visual Human Computer Interaction Technique For Complex Control Experimental System

The experiments, as an important mean of teaching and research in Automation Subject, have important meaning on validation and application of control theory. In the experimental system, experimental software is an important front-end for human-computer interaction between researchers and mechanical devices, which plays a critical role in the operation and running of experiments. At present, although the complex control experiment software domestic and foreign all have their own characteristics, can meet users’ specific needs. There are still some deficiencies mainly in:the scalability of software architecture is not strong, not conducive to the extension; lacks of support for complex control algorithms, needs being enhanced and improved on human-computer interaction features. Therefore, to solve these problems, this paper put the visual human computer interaction technique in reference to the complex control experiments, and the test result shows that the technique has improved the efficiency of the experiment and enhanced the application and analysis of control theory. The main contents of this paper as follows:1. This paper designs the features, which are visual human machine interface for complex control experiments. The features called complex control experiment management creating and editing the experiments with the project approach, while achieving the creation and editing of the experiment file. Algorithm configuration editing realizes the creation of control diagram, associated with the control algorithm. Variable management provides variable trees, variable lists and variable rapid links. Variable operation can meet displaying the signals and adjusting the parameters. The virtual animation of control plant makes the control results more intuitive.2. Using a variety of techniques to achieve the features designed. This paper uses XML serialization, deserialization and dynamic script programming to achieve the preservation of the experiment interfaces. In addition, uses Drag and Drop to achieve the creation of control plants and association with control algorithms, Java reflection mechanism and interactive technique with Matlab, and simplifies variable management more effectively by JFace viewer and the use of DnD. Moreover, allows signals to be displayed more diversity and parameters to be adjusted more effective and efficiently, through the adapter pattern, Matlab API and DnD technique, while simplifies the configuration of animation signal and real-time displaying based on Flash animation and the creation of properties of control plants.3. The features of the software, which developed, are tested and verified. Based on functional testing and performance testing, putting the2nd tank of the double-tank as control plant, we design and validate a single-capacity water tank level control experiment. In addition, the experimental results show that the software can not only create experimental projects and design experimental interface easily, can also edit and save the components in the experiment interfaces. At the same time, the software features meet the research and implementation of complex control algorithms so that users can analysis the control effect and the operation of state more better. Finally, this paper has a summary and outlook on the work.