(457c) Multi-Agent Optimization Framework (MAOP) for Synthesizing Optimal Radioactive Waste Blends

Berhane H. Gebreslassie and Urmila M. Diwekar

Center for Uncertain Systems, Tools for Optimization and Management (CUSTOM):Vishwamitra Research Institute, Crystal Lake, IL 60012 â?? USA

berhane@vri-custom.org; urmila@vri-custom.org

Abstract

The multi-agent optimization (MAOP) framework provides a way of combining various algorithms in one platform and exploits the strength possessed by each algorithm. Most of large scale process system engineering problems include nonlinear and non-convex problems. In this work, we propose a multi-agent optimization framework for solving complex large scale process system engineering problems. The framework uses a variety of different algorithmic agents which include the gradient based local optimizers and metaheuristic algorithms (efficient simulated annealing, efficient genetic algorithm and efficient ant colony optimization algorithms). Each agent encapsulates a particular problem-solving procedure. We investigate the effect of cooperation among agents of the multi-agent system combined into a framework designed to solve large scale combinatorial optimization problems. Computational experiments are carried out using benchmark problems and real world case study. The proposed methodology enables to improve the quality of solutions and the computational efficiency in comparison with non-cooperative multi-agent framework and standalone agents. The MAOP framework which includes the five major parts of the algorithm (figure 1) is given in the figure 2.

Fig. 1. The major parts of the MAOP algorithm and the information flow direction.

Fig. 2. The basic flow diagram of the heterogeneous MAOP algorithm

Case Study:

A real-world problem of synthesizing optimal waste blends is analyzed to test the applicability of this novel approach in addressing a general synthesis problem. The Hanford Site in southeastern Washington produced nuclear materials using various processes for nearly fifty years. Radioactive waste was produced as by-products of the processes. This waste can be retrieved and separated into high-level and low-level wastes and subsequently immobilized for future disposal. The high-level waste is converted into a glass form before disposal. However, the glass must meet both process-ability and durability restrictions. The process-ability conditions ensure that during processing, the glass melt has properties such as viscosity, electrical conductivity and liquidâ??s temperature such that they lie within the range of known to be acceptable for the vitrification process. Durability restrictions ensure that the resultant glass meets the quantitative criteria for disposal in a repository. There are also bounds on the composition of the various components in the glass. The radioactive waste and frit are mixed and heated to form a glass that satisfies the constraints related to the glass property and configuration.

Hanford has 177 tanks (50,000 to 1 million gallons) containing radioactive waste. These wastes vary widely in composition, and the glasses produced from these wastes will be limited by a variety of components. The minimum amount of frit would be used if all the high-level wastes were combined to form a single feed to the vitrification process. Because of the volume of waste involved and the time span over which it will be processed, this is logistically impossible. However, much of the similar benefit can be obtained by forming blends from sets of tanks. The problem is how to divide all the tanks into sets to be blended together such that the minimal amount of frit is used. Therefore, the goal of this work is to find the optimal blend configuration that minimizes the amount of frit added to convert the radioactive waste into glass. Minimizing the amount of frit added to a minimum amount has two explicit reasons.

Fig. 3. Processing of radioactive wastes to glass

References

1. Siirola JD, Steinar Hauan S, and Westerberg AW, 2003. Toward agent-based process systems engineering: proposed framework and application to non-convex optimization. Computers and Chemical Engineering 27: 1801-1811.
2. Gebreslassie BH, and Diwekar UM, 2015. Efficient ant colony optimization for computer aided molecular design: Case study solvent selection problem. Computers and Chemical Engineering 78: 1-9.
3. Diwekar UM, Xu W, 2005. Improved genetic algorithms for deterministic optimization and optimization under uncertainty. part I. algorithms development. Industrial and Engineering Chemistry Research 44(18) 7132-7137.
4. Venkatesh Narayan V, Diwekar UM, Hoza M. Synthesizing Optimal Waste Blends. Computers chem. Engng Vol. 20, Suppl., pp. SI443-SI448, 1996.
5. Chaudhuri P, Diwekar UM, 1999. Synthesis Approach to the Determination of Optimal Waste Blends under Uncertainty. AIChE Journal. 45: 1671-1687.