The designing of electromagnetic devices is a two-part process. First a broad experience-based decision is made on the generic class of the device to be used. Then the design process takes a very specific analytical approach where the method is highly specialised, mathematical and uses sophisticated design tools. This part makes the generic device very specific by defining through rigorous analysis its dimensions, materials etc.
The main objective of this project is to find out the methods by which the first part of the design process can be automated. This involves capturing, representing and manipulating the engineering domain knowledge.
We will be looking for ways in which the engineering design knowledge can be represented, ways in which it can be manipulated to give meaningful solutions and ways in which the knowledge base can be searched efficiently.
So as to narrow the project down with strict focus on the conceptual methodology, the project is confined to the first part of the Design Process in the design of motors.
The objectives of the project could be defined as:
1. Apply AI techniques to represent and manipulate engineering design knowledge.
2. Find ways to capture the Engineering Design Knowledge especially the motor design rules.
3. Integrate the traditional AI based methods with the fairly new Object Oriented concepts and produce new representation and reasoning methodologies.
4.
Create an Engineering Assistant that will automate the first part of the
design process. Once the requirement specification is given, this Engineering Assistant will solve the
problem of finding the generic class of the device.
5. Create a prototype for a language, which can be used for knowledge representation and inference. This language will try to overcome the limitations/drawbacks of the languages that were surveyed. New concepts will be developed to overcome these limitations and will be implemented in this language.