針對(duì)傳統(tǒng)專家系統(tǒng)和神經(jīng)網(wǎng)絡(luò)的各自特點(diǎn)，將兩者有機(jī)結(jié)合，構(gòu)造了一種基于神經(jīng)網(wǎng)絡(luò)
的混合型故障診斷專家系統(tǒng)。提出了神經(jīng)網(wǎng)絡(luò)子模塊NNM 的概念并分析了NNM 的具體實(shí)現(xiàn)方法。應(yīng)用該故障診斷模型對(duì)某內(nèi)燃機(jī)組供油系統(tǒng)故障進(jìn)行了模擬驗(yàn)證，試驗(yàn)結(jié)果表明神經(jīng)網(wǎng)絡(luò)與專家系統(tǒng)結(jié)合是一種有效的診斷方法。
關(guān)鍵詞：神經(jīng)網(wǎng)絡(luò)，NNM，故障診斷，專家系統(tǒng)
Abstract: According to the characters of traditional expert system and neural network, a neutral-network-based mixed fault diagnosis expert system was designed. A concept of neural network model(NNM) was given, and the certainly implementary method of NNM was analysed. This method was applied in fault diagnosis of Internal Combustion Engines, the results show this method is an effective diagnostic method.
Key words: Neural network, NNM, fault diagnosis, expert system