Method For Predicting Disorder Of Tower Crane By Using Data Mining

摘要

Defining at least two or more state data relating to a tower crane associated with any failure of the tower crane, collecting the state data relating to the failure, collecting the collected state data relating to the tower crane over a threshold value Is transformed into a value of 1, and other values are proportionally converted with respect to a threshold value to have a range of values of 0 and 1, and a sigmoid function is used as a transfer function Calculating a probability of the failure as an output value by inputting the normalization data as an input value to the artificial neural network set by adjusting a connection weight by a learning process in advance, As the dependent variable, the time when the probability of the failure becomes 1 is regarded as an independent variable, The tower crane failure prediction method using the data mining, comprising the step of predicting a time to failure is provided. According to the method of predicting the tower crane failure using the data mining of the present invention, since the timing of occurrence of the failure of the tower crane can be accurately predicted, the tower crane does not fail during operation and the operation efficiency of the port is not lowered.