Faced　with　the　characteristics　of　non-linearity　and　time-varying　delay　of　the　controlled　object　in　the　actual　industrial　control　process,　it　is　difficult　for　the　parameters　of　the　conventional　PID　to　change　in　real　time　to　achieve　the　ideal　control　effect.　In　the　field　programmable　logic　gate　array　(FPGA)　integrated　development　environment,　the　BP　neural　network　self-tuning　PID　controller　is　designed.　In　the　construction　of　neural　network,　the　parallel　processing　ability　of　FPGA　and　the　deep　pipeline　technology　are　used　to　realize　the　serial　and　parallel　connection　of　the　network,　and　all　the　weights　of　each　layer　of　the　network　are　stored　in　registers　to　realize　cyclic　storage.　The　method　of　Modelsim　and　Simulink　is　used　to　simulate　and　verify　the　system　design.　The　co-simulation　shows　that　the　design　scheme　is　effective　and　feasible,　and　has　the　characteristics　of　reliable　performance,　good　control　effect　and　strong　usability.　©　2022,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.