In　recent　years,　bidirectional　DC-AC　converters　have　been　widely　used　in　power　electronic　circuits.　However,　long-term　operation　at　high　frequency　and　high　power　will　lead　to　converter　aging　and　component　parameters　degradation.　Therefore,　the　accurate　identification　of　the　component　parameters　of　the　converter　is　of　great　significance　for　the　stable　and　efficient　operation　of　the　power　system.　In　this　paper,　a　parameter　identification　method　based　on　model　and　improved　BPNN　(Back　Propagation　Neural　Network)　is　proposed.　Firstly,　the　mathematical　model　of　the　converter　is　established　by　analyzing　the　topology,　and　the　state　equations　of　the　inductor　current　and　the　capacitor　voltage　are　derived　as　the　characteristic　equations.　The　improved　BPNN　is　used　to　fit　the　characteristic　equations,　and　the　function　of　the　relevant　component　parameters　is　used　as　the　adjustable　weight　value　of　the　improved　BPNN.　The　output　of　the　improved　BPNN　is　compared　with　the　measured　reference　value.　When　the　fitting　error　is　small　enough,　the　values　of　inductance,　capacitance,　ESR　and　other　parameters　can　be　directly　extracted　from　the　weights.　The　simulation　results　show　that　the　method　is　feasible　and　accurate.　©　2022,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.