I am in a quandary and have reserched here and never really found an answer so maybe I will get lucky [on top of researching like crazy elsewhere]

I need to compare the thermal behavior of two fluids over time - one fluid will be used as a "simulation" fluid in shipping box qualifications because the fluid it is simulating cannot be sacrificed for routine experimenation

- what we are looking at is data with the x-axis being time and the y-axis being temperature of the two fluids. They are placed in a hot environment than allowed to cool to a cooler environment and then will be warmed back to the hot enviroment to compare the rate of cooling as a function of time.

We intend to run triplicate tests of both the actual product and the simulation fluid. I am looking for a way to determine whether or not the cooling curves are statistically "equivalent"

We have already considered the following:

2 Way ANOVA with replication - only problem is time/temperature are not independant - not sure it is valid to ignore the time/temperature data and only compare the product and simulation fluid data. This to me seems the most logical approach and the results match some visual data from another test. But it is a "bastardized" approach.

Paired t-tests for all pair combinations [e.g. 1 vs 2, 1 vs 3., 2 vs 3.....]. I did look at that with some existing data and have some concerns that very similar data [visually] is still statistically different at even the 90% confidence interval. In this case 3 replicates of the exact same fluid subjected to the same treatment. Is the paired t-test really an appropriate measure?

Pearson's Correlation - it was interesting but clearly this appears to match the visual data comparisons I have done but I am not sure how to use it and how to apply any statistical significance or if that is a valid test.

Any suggestions - I have talked with some statistics folks who can "compare the numbers for me" but good experimental design requires that I define my statistical method and acceptance criteria before collecting the data.