Based　on　dynamic　neural　network-Boltzmann　machines,　a　new　receding　optimal　control　method　of　penicillin　fermentation　processes　is　proposed.　First,　according　to　the　input　and　output　data　of　the　fed-batch　fermentation　processes,　the　subsection　state-space　models　are　built　by　system　identification　tools.　Secondly,　adopting　the　dynamic　neural　network-Boltzmann　machine　as　an　optimal　controller,　combining　with　the　penicillin　fermentation　process　and　its　state-space　model,　this　paper　structures　the　receding　optimization　closed　loop.　Simulation　experiment　results　show　that　using　this　method　the　outcome　concentration　can　increase　thirty　percent,　and　this　control　method　is　effectual.　©　2006　IEEE.