Fitting potential energy surface of reactive system via genetic algorithm

Wiliam Ferreira da Cunha,Luiz Fernando Roncaratti,Ricardo Gargano,Geraldo Magela e Silva

DOI: https://doi.org/10.48550/arXiv.physics/0511131

2005-11-15

Abstract:In this work, we present a new fitting of the Na+HF potential energy surface (PES) utilizing a new optmization method based in Genetic Algorithm. Topology studies, such as isoenergetic contours and Minimum Energy Path(MEP), show that the quality of this new PES is comparable to the best PES of literature. These facts, suggests that this new approach can be utilized as new tool to fit PES of reactive systems.

Computational Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to fit the potential energy surface (PES) of the reaction system, especially the potential energy surface of the Na+HF reaction, through the Genetic Algorithm Optimization Technique (GAOT). Specifically, the author hopes to find a mathematical model that can accurately describe the energy change during the Na+HF reaction process by this method.

### Main problems
1. **High - precision fitting of the potential energy surface**: Traditional fitting methods may not be able to well capture the energy changes in complex reaction systems, especially for reactions with high internal energy and bent transition states. Therefore, the author hopes to develop a new optimization method to improve the fitting accuracy.
2. **Global minimization problem**: Finding the global minimum in a multi - dimensional space is a challenge, especially in the presence of multiple local minima. The genetic algorithm is selected as an optimization tool because of its advantages in dealing with such problems.

### Solutions
The author proposes an optimization technique based on the genetic algorithm (GAOT) for fitting the potential energy surface of the Na+HF reaction. The specific steps are as follows:

- **Selecting an appropriate function form**: Use the Bond Order (BO) polynomial expansion as a trial function and combine the energy data obtained from ab initio calculations.
- **Encoding and decoding**: Encode the parameters into binary strings and gradually optimize these parameters through the operations of the genetic algorithm (selection, recombination, mutation).
- **Dynamic search space**: Adjust the search range and precision of the parameters according to the optimization results of the previous generations to accelerate convergence and improve the fitting accuracy.
- **Verification and comparison**: Verify the effectiveness of the new method by drawing isopotential energy contour maps and the minimum energy path (MEP) and comparing them with the existing BO5 and GSA potential energy surfaces.

### Conclusions
Through the above methods, the author has successfully fitted the potential energy surface of the Na+HF reaction, and the results show that the quality of the potential energy surface obtained by the GAOT method is comparable to that of the current best BO5 and GSA potential energy surfaces. This indicates that GAOT can be used as a new tool for fitting the potential energy surfaces of other complex reaction systems.

### Formula summary
- **Mean square deviation**:
\[
S=\sum_{p = 1}^{np}\delta_p^2=\sum_{p = 1}^{np}(e_p - e_p')^2
\]
where \(e_p'=V([a],\vec{r}_p)\).

- **Polynomial expansion of the three - body term**:
\[
V(R_{\text{NaF}},R_{\text{HF}},R_{\text{NaH}})=\sum_{x = 0}^{5}\sum_{y = 0}^{5}\sum_{z = 0}^{5}a_{xyz}\eta_{\text{NaF}}^x\eta_{\text{HF}}^y\eta_{\text{NaH}}^z
\]
where \(x + y+z\leq6\), and at least two exponents are not zero.

- **Transformation formula**:
\[
\eta_M = e^{-\beta_M(R_M - R_{eM})}
\]

Through these formulas and methods, the author demonstrates the effectiveness and potential of GAOT in fitting the potential energy surfaces of complex reaction systems.