A Multi-Phase Transport Model (AMPT) has been extensively used to understand the physics behind the experimental observation. Like other models, the outcome of the AMPT model depends on the initial parameters. Therefore, one needs to choose suitable initial parameters before using the model. Lund string fragmentation function has been used to create quark-antiquarks pairs in the AMPT model. The Lund string fragmentation parameters determine the yields and transverse momentum ($p_{T}$) spectra of particles produced in nucleus-nucleus collisions. The values of Lund string fragmentation parameters were determined by fitting the charged particle yield and $p_{T}$ spectra measured in the experiment. In this paper, we have shown the yield of strange quarks carrying hadrons, e.g. $phi$ mesons, are more sensitive to Lund string fragmentation parameters compared to non-strange pions. Using $phi$ meson spectra measured at RHIC, we have obtained new sets of Lund parameters for Au+Au collisions at $sqrt{s_{mathrm {NN}}}$ = 11.5, 39, and 200 GeV. We have found that using the same set of parameters, we can explain $phi$-meson yield at $sqrt{s_{mathrm {NN}}}$ = 39 and 200 GeV, however, we need a different set of parameters for $sqrt{s_{mathrm {NN}}}$ = 11.5 GeV. This suggests that at low energy, $sqrt{s_{mathrm {NN}}}$ = 11.5 GeV, the underlying mechanism for particle production is different compared to top RHIC energies. We have also predicted invariant yield of $pi$ and $phi$ mesons as a function of $p_{T}$ in U+U collisions at $sqrt{s_{mathrm {NN}}}$ = 196 GeV to be measured by STAR experiment.