The Researching Of Quality Modeling In The Complex Industrial Production Process Of Steel Rolling Products

Complex industrial production process of steel rolling is a complicated system, which composed of multiple subsystems process, possesses the characteristics of highly nonlinear, uncertainty, time-varying, large time delay, strong-coupling and multi-parameter. In order to realize the quality control, model building for complex industrial production process product of steel rolling is very necessary. But because of the complexity and industrial noise pollution in complex industrial production process, modeling for its product quality is very difficult. In the recent years, model building for its product quality with different neural networks is carried out by many domestic and foreign scholars. And that lives up to certain effect. But it still has certain distance in the aspect of the models built instructing production practice. In this paper, on the basis of the characteristics of the steel rolling production process, the product quality models of the industrial production process of the steel rolling get built by means of BP neural network and its improved algorithm with wavelet and genetic algorithm, after that, some results with practical significance get obtained. The main work and contents of this paper are as follows:(1) In view of complex production process of steel rolling, the problem of quality modeling for its products is analyzed, the factors that influence complex production process products of steel rolling by using correlation analysis method are researched and analyzed in detail, thirty two main factors that influence the quality of complex industrial production process products of steel rolling are determined, after that, the foundation of model building and quality control for production process products of steel rolling is laid.(2) The research for the products quality modeling methods of the steel rolling production process based on the neural network is carried out. On the basis of preprocessing for the steel rolling products data, model building for the steel rolling production process products quality with BP neural network and its improved algorithm by wavelet is carried out. And its instance simulation and analysis are carried out. The validity and reliability are indicated by the results of simulation and its analysis.(3) On the basis of analyzing the wavelet neural network's structure with Improved Genetic Algorithm, the products quality modeling methods based on WNN optimized IGA are proposed. The method has the advantage of Genetic Algorithm's global search capability and WNN Algorithm's simple and useful feature. The validity and reliability are indicated by the results of simulation and its analysis.