Protein adsorption at solid surfaces has received intense focus due to its high relevance to biotechnological applications. In alternative to experimental approaches, computational methods such as molecular dynamics (MD) simulations are frequently employed to simulate the protein adsorption process and to study molecular interactions at the interfacial region. However, a successful simulation of the adsorption process depends largely on the initial adsorbed protein orientation on the surface. To avoid sampling protein trajectory which will eventually fail to adsorb, a workaround is to first determine the preferred orientations of the protein relative to the surface and use them as starting structures in MD simulations. Here, we present the first application of particle swarm optimization (PSO) to search for the low energy docking poses of a protein molecule on a solid surface. Performing rigid-body translation and rotation of the protein with energy minimization and force field-based scoring function, our search algorithm successfully located the low energy conformations of the lysozyme molecule on a hydrophobic PTFE surface. Study from a rigorously testing of three different boundary conditions shows that when a PSO particle has moved outside the positional limit for translation along Z, re-positioning it at the boundary could have a higher chance to locate lower energy orientations. It also shows that using larger swarm size is easier to find a low energy orientation. The predicted conformations are subsequently clustered to identify unique protein adsorption conformations and the predicted adsorption sites identified from each cluster are agreeable to previously published long MD simulation results. This study proves that our method provides a reliable way to predict the optimal protein orientations useful for computational studies of protein-surface interactions.