With the widespread use of shield tunneling technique

many accidents appear

and it is necessary to introduce fault diagnosis technique in shield tunneling to avoid similar accidents appear again. In essence

the fault diagnosis technique in shield tunneling is the problem of pattern recognition

which can be realized by means of the BP feed forward neural network. Combined with the production examples in Guangzhou region

the practical application of BP feed forward neural network to realize the fault diagnosis is analyzed and discussed. Computation example shows that it is feasible to use BP feed forward neural network to realize the fault diagnosis. Of course

the accumulation of the data of typical faults should be strengthened and increase the degree of separation among faults in order to the increase the effectiveness of the fault diagnosis.