Multi-access　edge　computing　(MEC)　technology　is　a　potential　solution　to　the　conflict　between　resource-constrained　unmanned　aerial　vehicles　(UAVs)　and　various　emerging　computationally　intensive,　delay-sensitive　applications　of　UAVs.　In　this　paper,　we　study　the　joint　optimization　problem　of　distributed　computation　offloading　and　resource　allocation　in　MEC-supported　UAV　application　scenarios,　where　the　task　of　each　UAV　can　be　divided　and　offloaded　to　multiple　servers　deployed　in　ground　wireless　access　points　(WAPs)　for　parallel　processing.　To　enhance　the　quality　of　user　experience　(QoE)　while　avoiding　excessive　optimization,　we　propose　a　new　user　satisfaction　function　to　evaluate　the　task　processing　level　of　MEC　network,　which　respectively　scores　the　actual　task　processing　delay　and　energy　consumption　based　on　different　criteria　determined　by　task　requirements.　We　aim　to　trade　off　delay　and　energy　consumption　by　jointly　optimizing　server　selection,　computing　resource　allocation,　transmission　power　control,　and　task　allocation　to　achieve　maximum　user　satisfaction.　The　joint　optimization　problem　is　formulated　as　a　mixed-integer　nonlinear　programming　(MINLP)　problem.　To　reduce　the　complexity　of　the　joint　optimization　problem,　it　is　decomposed　into　three　sub-problems　based　on　the　divide-and-conquer　approach:　server　computing　resource　allocation,　transmission　power　control,　and　task　offloading　decision.　We　propose　a　three-layer　intelligent　optimization　(TIO)　algorithm,　which　employs　the　Lagrange　multiplier　method,　the　AdaDelta　gradient　descent　method,　and　our　proposed　multivariate　decoupling-particle　swarm　optimization　(MD-PSO)　algorithm　to　solve　the　above　three　sub-problems　alternately　for　gradually　approaching　the　maximum　user　satisfaction.　The　numerical　results　show　that　our　proposed　algorithm　is　significantly　superior　to　the　benchmark　algorithms　in　terms　of　solving　speed,　accuracy,　and　stability.　©　2023　IEEE.