The　stator　resistance,　d-axis　inductance,　q-axis　inductance　and　rotor　permanent　magnet　flux　linkage　of　inner　permanent　magnet　synchronous　motors　(IPMSM)　will　be　changed　easily　at　the　moment,　which　can　deteriorate　operation　performance　of　the　motor　control　system.　To　settle　this　issue,　multi-parameter　online　identification　method　of　the　IPMSM　based　on　trapezoidal-wave　current　injection　was　proposed.　Forgetting　factor　recursive　least　square　was　adopted　to　identify　the　stator　resistance,　d-axis　inductance,　q-axis　inductance　and　rotor　permanent　magnet　flux　linkage　of　the　IPMSM　and　the　problem　of　under-ranking　in　simultaneous　online　identification　of　four　parameters　was　solved.　Compared　with　the　square-wave　current　injection　method,　the　proposed　method　reduces　the　torque　ripple　of　the　motor.　To　achieve　fast　and　accurate　convergence　of　the　identification　parameters,　the　condition　of　persistence　excitation　was　studied.　To　reduce　influence　of　inverter　nonlinearity　on　parameter　identification　accuracy,　the　inverter　nonlinearity　model　in　the　two-phase　synchronous　rotation　coordinate　system　was　built　and　analyzed.　Simulation　and　experiment　results　demonstrate　effectiveness　of　the　proposed　method.　©　2022,　Harbin　University　of　Science　and　Technology　Publication.　All　right　reserved.