Here we can see the results graphed. The simulated code doesn't provide us with a stateTime at this point, so it assumed that this is the equivalent of State 1, stateTime 0.
It is encouraging here to see the double peaks; that is usually a good visual indication that the simulation is generating the results that we should expect. For comparison, below we show the graph that we are using for our benchmarks (http://en.wikipedia.org/wiki/Fission_product_yield).
Comparing the two graphs highlights some issues with our Simulation.
- The States doesn't match up well. The Benchmark graph shows results for 1, 10, 100, and 1000 years. Ours output shows only the results directly following fission.
- The values are not "spiking" on the elements where it would be expected. Indeed, the benchmark doesn't show a value for Pm (Prometheum), yet our results show a high frequency of Pm resulting from the decay of U-235
- Our values are measured in absolute values, where their values are measured as a fission yield percentage. Without the prior tow issues so prominent, we would correct this issue and compare the results a second time.
- Semantic issue, but the labels on our x-axis don't match theirs. This isn't a critical break in the process, but could certainly lead to some confusion in interpreting the results.


No comments:
Post a Comment