Analytically Tuned Simulated Annealing Applied to the Protein Folding Problem
ICCS '07 Proceedings of the 7th international conference on Computational Science, Part II
Solving SAT problems with TA algorithms using constant and dynamic markov chains length
AAIM'05 Proceedings of the First international conference on Algorithmic Applications in Management
| Hi-index | 0.00 |
This paper presents a new approach named MultiQuenching Annealing (MQA) for the Protein Folding Problem (PFP). MQA has two phases: Quenching Phase (QP) and Annealing Phase (AP). QP is applied at extremely high temperatures when the higher energy variations can occur. AP searches for the optimal solution at high and low temperatures when the energy variations are not very high. The temperature during the QP is decreased by an exponential function. Both QP and AP are divided in several sub-phases to decrease the temperature parameter until a dynamic equilibrium is detected by measuring its quality solution. In addition, an efficient analytical method to tune the algorithm parameters is used. Experimentation presented in the paper shows that MQA can obtain high quality of solution for PFP.