An Intelligent Measurement Method of the Flatness Errors on the Surface of Arm Cylinder with Medium and Small Tonnage Crane

In order to solve the problems of inconvenient and complication in planeness measuring on medium and small tonnage crane arm’s welding surface, a model based on the combination between support vector machine (SVM) and fuzzy logical theory (FLT) was established in planeness measuring system. In the model, the influencing factor of welding deformation and the displacement of welding face were set as input and output, respectively. Meanwhile, the parameters of kernel function and penalty factor in the model were optimized using genetic algorithm theory where the error function was defined as the evaluation index. In the end, the kernel function and penalty factor ($\sigma=1.8$ and c=26) were confirmed, respectively. When the model was used in practical application, the results showed that the model had a better effort compared to traditional measuring method and least squares support vector machine (LSSVM) method. Therefore, the model proposed in this study was an effective measuring method.