My aim is to make a power consumption data logger and analyzer that is non-intrusive and can be hooked anywhere to monitor the power consumption on that line. I've come across clamp sensors which uses the principle of electromagnetic induction to measure the current on the line. These products are usually in the form of a complete meter. For my engineering project, I want to implement this manually by myself. I guess one can make a clamp sensor by winding a wire, but I'm not well versed in the technical details of doing so. How can I build one that is accurate, and how can a design an appropriate amplifier circuit? The idea is to supply this small secondary current to op-amp circuit, and then use a microcontroller with built-in ADC to measure and calculate the different electrical characteristics.
For a conventional current transformer, you can refer to Colonel Wm. T. McLyman's book Transformer and Inductor Design Handbook. There is a small section on current transformers. You should be able to find it at the library.
The principles of current transformers are similar to any other transformer, but the details (flux density, core materials etc.) will likely differ considerably if you want good accuracy.
Another option would be to wind a Rogowski coil, however the Rogowski coil responds to the derivative of current, so there are certain advantages and disadvantages. It's air core, so no worries about linearity. On the other hand, the signal processing requires integration.
I would suggest considering making a planar Rogowski coil using prototype PCB technology.