Intelligent Power Control
-
Gonzalez, A.J., Morris, R. A., Hagman P., "Electric Power
Control Using a Global Hybrid Approach," Intersociety
Energy Conversion Engineering Conference, 1995.
[
ABSTRACT
|
FULL DOCUMENT ] (Full document will be loaded onto a new page.)
There is strong evidence for the need for software systems capable of efficiently and robustly automating diagnostic and control tasks related to the generation, transmission and distribution of electrical power. Although researchers and developers have recognized the need for such a device, the vision, as yet, has not been realized. Two requirements, integral to the success of a global power controller, have represented major obstacles to be solved in realizing this objective. These are 1) the problem of automatic decision making in the face of complexity in the system being monitored, and 2) the problem of making power control decisions very quickly as well as reliably. This is because a power system is typically complex, and its behavior is a product of many interacting parameters. Techniques from artificial intelligence have been successfully utilized by others to advance the state of the art in automatic diagnosis and control. It is our firm belief that the best means of overcoming the technical obstacles will be the application of AI in the form of model-based reasoning. This belief is justified by concrete research results. Yet, the predominant sentiment among researchers in model-based to the model-based paradigm before it can be effectively applied to the electric power systems domain. In this paper, we describe the concept of hybrid reasoning, where traditional model-based reasoning techniques are combined with qualitative reasoning to achieve the greatest possible simplicity and, therefore, speed.