SYNTHESIZING OPTIMAL WASTE BLENDS

Vitrification of tank wastes to form glass is a technique that will be used for the disposal of high-level waste at Hanford. Process and sto rage economics show that minimizing the total number of glass logs pro duced is the key to keeping cost as low as possible. The amount of gla ss produced can be reduced by blending of the wastes. The optimal way to combine the tanks to minimize the volume of glass can be determined from a discrete blend calculation. However, this problem results in a combinatorial explosion as the number of tanks increases. Moreover, t he property constraints make this problem highly nonconvex where many algorithms get trapped in local minima. In this paper we examine the u se of different combinatorial optimization approaches to solve this pr oblem. A two-stage approach using a combination of simulated annealing and nonlinear programming (NLP) is developed. The results of differen t methods such as the heuristics approach based on human knowledge and judgment, the mixed integer nonlinear programming (MINLP) approach wi th GAMS, and branch and bound with lower bound derived from the struct ure of the given blending problem are compared with this coupled simul ated annealing and NLP approach.