The Soft-Sensing Of Aluminum Powder Rate Based On Support Vector Machine

In the real produce progress of industry, there are lots of progress parameters that can't be measure online. In order to solve such problems, soft-sensing techniques has emerged. Support Vector Machine is a learning technology based on structure risk minimization, its holds the strict theory and mathematics basement which doesn't exists local optimization, and doesn't rely on the quality and the number of training data excessively, so it fit to the complex industry progress of soft-sensing. In the process of atomized aluminum powder, the rate of aluminum powder is the key indicator to control production quality, but the variable can't be on-line measurement.This paper based on the practical control system development project. In order to establish the model, the key is to search for the factors that affect the rate of aluminum. We analyze the parameters that affect the rate of aluminum powder from the aluminum powder nitrogen atomization process and spray nozzle physical model. All of this prepares for establishing the model of soft-sensing.For the application of Support Vector Machine in the aluminum powder process, this paper search the impact of nuclear function and structure parameters and select the RBF kernel function. I analyze the methods of bilinear search and grid search and propose bilinear grid search algorithm. This paper preprocesses the data from on-site and selectsÏ…- SVR. We build up the soft-sensing model based on Support vector machine. The actual data simulation experiments confirm that the model has better accuracy ability than RBF. In the end, the model is applied in the host computer application development.