Sts Cranes Working Condition Recognition Technology Based On Fuzzy Neural Network And Its Application Research

Because the STS cranes have the development trend of becoming more large scale and high efficient, the paper devotes to research some technologies about the status monitoring and failure diagnosis of the crane.The data of the paper are collected by the HMSAS system invented by the Dept. of Mechanical Engineering of Shanghai Maritime University. Collect the stress data of the six positions on the crane when it is working. There are some interference on the machine and for some factitious reason some data collected are not correct. So an ANSYS model is built to calculate the stress in theory. To compare the collected data with the theoretical data and delete the data which are far from the theoretical value. Then the data become more accuracy.During the pattern recognition, the collected data are divided into two parts. One part is used to train the net. Another part is used to test the recognition capability of the net. Considering the problem the paper wants to solve, when build the net blend the fuzzy logic in it. Because the RBF net has the advantage of fast study speed and precise recognition, so in the paper use RBF net to solve the problem. It is found that the RBF net is better than the BP net on these two sides. Base on the RBF net use two ways to combine with the fuzzy logic to make two net. Training these two net and use them to recognize the work pattern. It is found that the first net has faster study speed and the second net can recognize more precisely.