open source force-matching https://www.potfit.net/wiki/ Wed, 29 Apr 2026 13:29:05 +0000 FeedCreator 1.8 https://www.potfit.net/wiki/lib/exe/fetch.php?media=wiki:logo.png https://www.potfit.net/wiki/ https://www.potfit.net/wiki/doku.php?id=algorithms:diffevo&rev=1515605168 Differential Evolution ---------- The Differential Evolution algorithm can be used instead of simann in the optimization part of potfit. Enabling it with the evo compilation switch will create a binary that first uses Differential Evolution and then subsequently the $NP$$\boldsymbol x_{i,G=0}\;,i=1,2,\ldots,NP$$G$$NP-1$$\boldsymbol x_{i,G}\;i=1,2,\ldots,NP$$\boldsymbol v$$$\boldsymbol v_{i,G+1}=\boldsymbol x_{r_1,G}+F\left(\boldsymbol x_{r_2,G}-\boldsymbol x_{r_3,G}\right)$$$r_1,r_2,r_3 \in {1… anonymous@undisclosed.example.com (Anonymous) Wed, 10 Jan 2018 17:26:08 +0000 https://www.potfit.net/wiki/doku.php?id=algorithms:lsq&rev=1515603681 Method of Least Squares ---------- This page is a quote from the PhD Thesis of P. Brommer. ---------- Force Matching is closely related to determining the parameters $\xi_{i}$, $i=1,\ldots,n$ of a parametrised function $f_{\left\{\xi_{i}\right\}}(x)$ from a series of $m$ measurements, with $m\gg n$ $$f_{\left\{\xi_{i}\right\}}(x_{j}) = y_{j}, \quad j=1,\ldots,m.\qquad (2.51)$$ Due to measurement errors, this system can generally not be solved exactly. The best solution minimises the devia… anonymous@undisclosed.example.com (Anonymous) Wed, 10 Jan 2018 17:01:21 +0000 https://www.potfit.net/wiki/doku.php?id=algorithms:main&rev=1515603698 Algorithms ---------- This page is a quote from the PhD Thesis of P. Brommer. ---------- Force Matching As described above, classical MD calculations require effective potentials, which for many simple materials are matched to well-known experimental values. For structurally or chemically complex materials, there are generally more potential parameters to be determined, and the number of usable experimental measurements does not cover the increased need for data. In some cases, the exact st… anonymous@undisclosed.example.com (Anonymous) Wed, 10 Jan 2018 17:01:38 +0000 https://www.potfit.net/wiki/doku.php?id=algorithms:sidebar&rev=1550911594 Main page ---------- Download Publications Contact About potfit User Guide * Compiling potfit * Options * Running potfit * Parameters * Configuration files * Potential models * Potential files * Interactions * Algorithms * Simulated Annealing * Differential Evolution * Method of least squares * Output * Units * Utilities * Limitations Examples * Configurations * Potentials * Parameter files * Advanced Examples * Tutorials Potential Database… anonymous@undisclosed.example.com (Anonymous) Sat, 23 Feb 2019 08:46:34 +0000 https://www.potfit.net/wiki/doku.php?id=algorithms:simann&rev=1515603577 Simulated Annealing ---------- This page is a quote from the PhD Thesis of P. Brommer. ---------- If there is a disadvantage to Powell's method as described above, it is that it readily jumps to the closest local minimum available, ignoring any other, perhaps even better local minima. As the sequence of $Z(\boldsymbol\xi_{i})$$\Delta E,$$\Delta E>0$$P(\Delta E)=\exp(-\Delta E/k_{B}T)$$T$$\boldsymbol x$$\delta\boldsymbol x$$Z(\boldsymbol\xi)$$. $$T$$\boldsymbol v\in\mathbb{R}^n$$\boldsymbol x… anonymous@undisclosed.example.com (Anonymous) Wed, 10 Jan 2018 16:59:37 +0000