Fast　accurate　and　reliable　identification　of　photovoltaic　(PV)　model　parameters　based　on　measured　current-voltage　(IV)　characteristic　curves　is　significant　for　the　analysis,　evaluation　and　diagnosis　of　the　operating　status　of　in-situ　PV　arrays　to　optimize　solar　energy　conversion.　Although　many　techniques　have　been　proposed,　it　is　still　challenging　to　achieve　both　fast　and　accurate　parameters　identification　with　high　reliability.　In　this　paper,　based　on　a　new　eagle　strategy,　an　improved　adaptive　Nelder-Mead　simplex　(NMS)　hybridized　with　the　artificial　bee　colony　(ABC)　metaheuristic,　EHA-NMS,　is　proposed　to　improve　parameters　identification　of　PV　models.　The　proposed　novel　eagle　strategy　consists　of　three　cascaded　stages:　coarse　exploration,　coarse　exploitation　and　fine　exploitation,　through　which　the　strong　global　exploration　of　ABC　and　the　powerful　local　exploitation　of　NMS　merits　are　combined　and　the　high　computation　burden　of　ABC　and　the　high　probability　of　being　trapped　in　local　minima　of　NMS　drawbacks　are　alleviated.　The　EHA-NMS　is　compared　with　some　state-of-the-art　algorithms　on　three　benchmark　problems　of　model　parameters　identification　of　a　R.T.C　France　solar　cell　and　Photowatt-PWP201　PV　module　which　are　commonly　adopted　in　the　literature.　The　intensive　experiment　result　and　analysis　show　that　the　EHA-NMS　outperforms　other　state-of-the-art　techniques　especially　in　terms　of　convergence　and　reliability.　Due　to　the　high　computation　efficiency,　the　EHA-NMS　can　be　easily　ported　to　embedded　systems　to　realize　online　real-time　parameters　identification　of　PV　models.　©　2016　Elsevier　Ltd