P300　is　an　important　control　system　signal　in　the　brain,　so　there　is　an　urgent　need　and　practical　significance　to　work　on　the　efficient　classification　of　P300　event-related　potentials.　In　this　article,　we　design　a　convolutional　neural　network　CNNnet　based　on　chaotic　adaptive　particle　swarm　optimization　(CAPSO)　algorithm　for　efficient　and　accurate　detection　and　classification　of　P300　EEG　signals.　The　chaotic　adaptive　particle　swarm　optimization　algorithm　uses　Logistic　chaotic　mapping　to　initialize　the　initial　position　of　particles,　and　adopts　a　dynamic　adaptive　weighting　strategy.　Compared　with　traditional　particle　swarm　optimization　algorithms,　it　can　effectively　improve　the　optimization　speed　and　convergence　speed　of　particles.　The　experimental　results　show　that　compared　with　other　P300　detection　neural　networks　and　traditional　particle　swarm　optimization　algorithms,　this　algorithm　has　faster　convergence　speed　and　higher　convergence　accuracy,　and　can　effectively　avoid　the　problem　of　particle　swarm　falling　into　local　optima.　©　2023　ACM.