自动化博览
自動化博覽
자동화박람
AUTOMAITON PANORAMA
2013年
z2期
83-86
,共4页
跟踪雷达%BP网络%故障诊断%专家知识
跟蹤雷達%BP網絡%故障診斷%專傢知識
근종뢰체%BP망락%고장진단%전가지식
Tracking Radar%BP Network%Fault Diagnosis%Expert Knowledge
跟踪雷达具有高度集成化、复杂化的特点，针对其故障定位难，维修难度大，容易发生误检或漏检的问题，构建基于改进BP神经网络的故障诊断专家系统，通过获取专家经验知识，使网络具有与专家解决此类领域问题相似的能力；采取正向推理机制，实现对故障诊断专家系统知识库中知识的选择。实验结果表明，跟踪雷达神经网络故障诊断专家系统诊断结果与实际情况一致，准确率高达85%。
跟蹤雷達具有高度集成化、複雜化的特點，針對其故障定位難，維脩難度大，容易髮生誤檢或漏檢的問題，構建基于改進BP神經網絡的故障診斷專傢繫統，通過穫取專傢經驗知識，使網絡具有與專傢解決此類領域問題相似的能力；採取正嚮推理機製，實現對故障診斷專傢繫統知識庫中知識的選擇。實驗結果錶明，跟蹤雷達神經網絡故障診斷專傢繫統診斷結果與實際情況一緻，準確率高達85%。
근종뢰체구유고도집성화、복잡화적특점，침대기고장정위난，유수난도대，용역발생오검혹루검적문제，구건기우개진BP신경망락적고장진단전가계통，통과획취전가경험지식，사망락구유여전가해결차류영역문제상사적능력；채취정향추리궤제，실현대고장진단전가계통지식고중지식적선택。실험결과표명，근종뢰체신경망락고장진단전가계통진단결과여실제정황일치，준학솔고체85%。
Tracking radar has the characteristics of high integration and complicated. To the problem of location and repair difficultly, a fault diagnosis expert system based on improved BP neural network was boiled, by acquiring expertise knowledge, making the network with the expert's ability to solve the problem of such areas similar, take the forward reasoning mechanism, to realize the choice of knowledge in the knowledge bas. The experimental results show that, the fault diagnosis expert system of neural network, it has the same results with the experts, the accuracy as high as 85%.
