At linear.com, there is an application note about hardware you can build, for simulating SMPS spikes. Perhaps that would be helpful.
First you need to learn about, or fully define, the nature of the voltage and current spikes that need to be investigated. For example, on what conductors can they appear? Are they on one conductor only, with respect to a second conductor that is "zero"? And which one(s)? If a pair of conductors is involved, could the current or voltage be either "common mode" or "differential mode"? And, for any and all of those cases, what are the ranges of the amplitudes of the voltages and currents. And for each of the cases, what is the desired result? So you need all of the requirements, and then also the sources of the requirements. THEN you can begin to define how to test, and determine if each requirement is met.