In　order　to　optimize　fuel　injection　results　of　the　electronically-controlled　pressure-accumulation　system　for　diesel　engines,　the　optimization　of　its　parameters　including　structure　parameters　and　function　parameters　is　required.　The　number　of　related　parameters　is　large　and　complex　relation　between　them　is　presented.　According　to　numerical　simulation　of　the　working　process,　based　on　the　simple　genetic　algorithm　(SGA),　some　types　of　improved　genetic　algorithms　are　applied　to　optimize　the　parameters.　In　this　paper,　the　process　of　the　parameter　optimization　designed　for　each　improved　genetic　algorithm　(GA)　is　explained　briefly　and　discussed　in　many　respects.　The　optimized　results　show　that,　in　the　total　number　of　the　largest　fitness,　AGA　is　355　times　more　than　SGA,　AORGA　is　392　times　more　than　it,　and　REGA　is　397　times　more　than　it.