ÕªÒª£ºÕë¶Ô¹ö¶¯Öá³ÐµÄ¹ÊÕÏÌص㣬Ìá³öÁËÒ»ÖÖ½«IMFÄÜÁ¿ÓëRBFÉñ¾­ÍøÂçÏà½áºÏµÄ·½·¨ÓÃÓÚ¹ÊÕÏÕï¶Ï¡£¸Ã·½·¨Ê×ÏÈÀûÓþ­Ñéģ̬·Ö½â£¨EMD£©·½·¨£¬°ÑÕñ¶¯ÐźŷֽâΪÈô¸É¸öIMF·ÖÁ¿,ÔÙÓÃÖØÒªµÄIMF·ÖÁ¿ÇóµÃIMFÄÜÁ¿ÌØÕ÷ÏòÁ¿£¬×îºó½«ÌØÕ÷ÏòÁ¿ÊäÈëRBFÉñ¾­ÍøÂç½øÐйÊÕÏģʽ·ÖÀࡣͨ¹ý¶Ô¹ö¶¯Öá³ÐµÄÕý³£×´Ì¬¡¢ÄÚȦ¹ÊÕÏ¡¢¹ö¶¯Ìå¹ÊÕϺÍÍâȦ¹ÊÕÏÐźŵķÖÎö½á¹û±íÃ÷£¬¸Ã·½·¨Äܹ»×¼È·¡¢ÓÐЧµØʶ±ðÕâЩ¹ÊÕÏ¡£

 Abstract£ºAccording to the characteristics of rolling bearing fault,This paper put forward a fault diagnosis method combining IMF energy and RBF neural network. This method firstly use the empirical mode decomposition (EMD) method to decompose the vibration signal into some IMF component, then garnish with important IMF component to obtain characteristic vector of IMF energy , finally put feature vector into RBF neural network fault to classify the fault pattern.Through the signal analysis of the normal state, inner ring fault, roller ring fault and outer ring fault , showing that the method can accurately, effectively identify these fault.