Synthesis of Conceptual Designs for Sensors

Abstract

A computer-aided technique is developed in this thesis to systematically generate concepts for sensors of a wide variety. A database of building blocks, based on physical laws and effects that capture the transduction rules underlying the working principles of sensors, has been developed to synthesize concepts. The proposed method uses the database to first create a concept-space graph and then selects concepts that correspond to paths in the graph. This is in contrast to and more efficient than existing methods, such as, compositional synthesis and graph-grammar synthesis, where solution paths are laid out first and then a concept-space graph is generated. The research also explores an approach for synthesis of concepts for closed-loop sensors, where a quantity is sensed indirectly after nullifying its effect by using negative feedback. These sensors use negative feedback to increase the dynamic range of operation without compromising the sensitivity and resolution. According to the literature, generation of un-interesting solutions is a major drawback of the building block-based synthesis approaches. In the proposed approach, this shortcoming is mitigated substantially by using some rules. For a number of the concepts generated, in the sensor problems attempted, we found that those concepts were already implemented in existing patents; thus emphasising the usefulness of the concepts produced. The synthesis approach proposed new, feasible sensor concepts, thereby indicating its potential as a stimulator for enhancing creativity of designers.

Another important problem is to improve the robustness of designs. Robustness can be achieved by minimizing the side effects. Side effects are defined as unwanted effects that affect the intended working of the sensor. The research presents an algorithm that (a) predicts the potential side effects for the synthesized concepts of sensors; (b) aids in quantifying the magnitude of the side effects, thus helping the designer to predict the significant side effects; and (c) suggests ways to improve the robustness of the design.