Deploying　wave　energy　converters　(WECs)　in　an　array　can　reduce　costs　due　to　shared　infrastructure,　operations,　and　maintenance.　Nevertheless,　the　close　placement　of　WEC　devices　introduces　new　challenges　for　control　design,　as　interactive　effects,　such　as　radiation　and　diffraction,　must　be　considered　within　an　energy　maximisation　control　framework　to　achieve　optimal　overall　performance.　Model　predictive　control　(MPC),　based　on　an　integrated　model　that　captures　all　hydrodynamic　interactions,　however,　typically　results　in　significant　computational　challenges,　particularly　given　that　many　devices　exhibit　nonlinear　hydrodynamic　behaviour.　To　address　this　issue,　this　paper　first　develops　a　＇Neighbour-to-Neighbour　(N2N)＇　distributed　nonlinear　MPC　scheme,　wherein　the　energy　maximisation　problem　for　the　array　is　calculated　distributively　by　the　individual　controllers　for　each　WEC　device.　The　array-level　energy　maximisation　objective　is　achieved　through　frequent　device-to-device　communications.　However,　for　certain　array　designs,　the　computational　cost　of　the　required　communication　may　still　be　excessive.　Therefore,　we　develop　another　＇Coordinated＇　distributed　scheme,　which　incorporates　an　array-level　coordinator　into　the　framework,　to　significantly　reduce　the　computational　load　associated　with　communication.　The　two　schemes　are　benchmarked　through　a　demonstrative　numerical　simulation　for　arrays　with　various　numbers　of　devices,　illustrating　the　necessity　to　account　for　interactive　effects　and　verifying　the　efficacy　of　the　distributed　approach　in　retaining　most　of　the　performance,　while　providing　advantages　in　computational　efficiency　and　scalability.