Curve fitting using heuristics and bio-inspired optimization algorithms for experimental data processing in chemistry

This paper analyzes the application of different bio-inspired and heuristic techniques to the problem of curve fitting in experimental chemistry applications. Two different curve models are considered, the well-known Pearson VII function, used before in different curve fitting applications, and a novel hybrid model for transmittance curves, mainly used in the estimation of nitrates concentration in water samples. We describe the performance of an Evolutionary Programming algorithm, a Particle Swarm Optimization technique, a Variable Neighborhood Search algorithm and a Cooperative Coevolution approach in several curve fitting problems, including the processing of real curves such as X-ray diffraction patterns, differential scanning calorimetry curves and transmittance curves of contaminated water samples.