The checkout and forecast for the product performance online

When the product is being manufactured, its quality will be checked and forecasted online. First of all the performance parameters of the products are chosen during the tentative course of the production. Secondly, by the help of keeping down most information of the raw data, wiping of the relativity among the datum and decreasing the influences of disturbance etc, the ANNs and ANFIS simulation model are founded. After analysis and discussion, ANNs is not reasonable, but ANFIS does well. Then through utilizing section taxonomy the ANFIS output data are divided into two sections, which correspond with the actual output of the speed sensor. Finally the section taxonomy and the ANFIS model are employed to check the products and forecast the quality. Since during the course of the production the performance parameters will be measured online, the coming wasters can be forecasted, the performance parameters should be adjusted online in the factory and the quality can be assured, the wasters can be avoided, the cost of the production can be decreased and the production time can be saved. The similar products of the customers’ requirements can be produced with the section taxonomy and the ANFIS model. The simulation examples demonstrate the accuracy and efficiency of the present method.