Medium Density Fiberboard Hot-pressing Parameters Optimization Control System Research

The hot pressing process of the Medium density fibreboard (MDF) MDF as an important production process link, the selection of pressing parameters directly influences the performance of MDF products and processing costs.The density fiberboard processing process energy consumption is high in the solution, the production cycle is long, the production end product performance parameter does not gather the people's desires in the question, is worth studying very much. At present, the thermo-compression technological parameter's optimization did not have the application in the actual production, but carries on the thermo-compression experiment cost to consume high, the parameter selection is complex.This topic is needs to develop one based on the neural network thermo-compression process simulation model, realizes through the thermo-compression parameter's multi-objective optimization model's solution reduces the production energy consumption the goal, and establishes a thermo-compression parameter management system management system, uses for to manage the thermo-compression process the parameter and the end product performance parameter match database as well as the management user database, realizes to the parameter active control and the selection.This article has analyzed the characteristics of BP neural network,,applies the density fiberboard thermo-compression parameter optimization this algorithm to select and to predict that has realized the density fiberboard thermo-compression parameter data mining, will train the good neural network model to fuse in the multi-objective optimization question, enables in the density fiberboard thermo-compression parameter optimal process to be able to apply the multi-objective optimization selection thermo-compression parameter directly.Through the establishment of the density fiberboard pressing parameters management system can quickly find the material parameters(the specific pressing temperature, pressing time, pressing pressure) in a particular hot medium density fiberboard production of finished products out of the performance,such as elastic modulusthe amount of water swelling, internal bond strength, etc., can also query the corresponding electricity consumption, to provide the necessary staff consultation, staff based on this information to select the appropriate processing parameters, optimization of pressing production process.Help to improve productivity, reduce production costs, make database information management more scientific.This is characterized by BP neural network model from the experimental data obtained learning samples and standards through training to achieve predictable, you can not predict the actual production process used sheet metal forming parameters on the properties of MDF;Using multi-objective optimization algorithm can be fast and reasonable process of formulating a hot temperature, pressure and time parameters, along with the hot and information feedback, the system through self-learning, and constantly improve the database to improve their decision-making, So that decision-making results more reasonable;Parameter management system running on the Windows platform, using SQL Server as the database server, development environment using Microsoft Visual C++ 6.0 (hereinafter referred to as VC), using MFC framework, and use MATLAB as a simulation tool.Experiments show that this method can quickly develop a reasonable pressing parameters, greatly reducing the application of the system parameters of the hot pressing process to choose the time and reduce the labor of workers, while reducing processing costs, not only to overcome the parameter selection process, human factors Impact, but also effectively control the processing quality.