Research On Model Building And Optimization Of MDF Continuous Hot-Pressing Process

Hot-pressing process is one of the key procedures in the production of medium density fiber-board (MDF). With the decrease resource of timber and the increase demand of MDF, it is of great important to improve the utilization ratio of energy and resource. This thesis mainly talks about the MDF hot-pressing automatic control system, and takes the support vector machine (SVM) and neural network intelligent algorithm as the basis tools. By data analysis, and deep understanding of the working mechanism and operation process of the hot-pressing process, this thesis gives some theoretical analysis and establishes two intelligent algorithms for the MDF hot-pressing system process, based on which this thesis also provides some feasible schemes for optimizing the MDF performance indexes. The main results of this thesis are as follows:1. First, we study the process and analyze the parameters of hot-pressing process. Then, based on the SVM theory with the existing linear and nonlinear kernel functions, we find the optimization problems for establishing the relations models with respect to the properties of the slab, technical parameters in hot-pressing process and the performance indexes of MDF.2. For the generated optimization problems used to establish the relation models of hot-pressing process, we design a new intelligent algorithm with rigorous theoretical proof. First, we add the regularization item into the built optimization problem to find the sparse solutions, which reduces the control complexity of the hot-pressing process. Penalty function is also used to solve the constraints, which brings the intelligent algorithm well optimal properties and brief structure. By the smoothing techniques, we propose an intelligent algorithm for solving the optimization problem generated by the indefinite kernel SVM theory. Next, we reform the original SVM opti-mization problem in another way and improve the former intelligent algorithm when the relation model of MDF hot-pressing process is built based on the positive kernel SVM theory. The new algorithm owns the better dynamic and optimal performance. Finally, based on the existing data set, we build the relation models for three kinds of MDF with different thickness, which takes the hot-pressing pressure, hot-pressing temperature, hot-pressing time and moisture content of slab as the argument variables, and the modulus of rupture (MOR), modulus of elasticity (MOE) and internal bonding strength (IB) indexes of MDF as the dependent variables. The numerical results show the good fitting of the built models for the data sets.3. Based on the built relation models for the hot-prossing process, this thesis establishes an-other optimization model and propose an relative intelligent algorithm for reducing the product cost and improving the labor efficiency by selecting the appropriate slab and controlling the hot-pressing process reasonably. First, we bring forward some classes of optimization models which take the hot-pressing pressure, hot-pressing temperature, hot-pressing time and moisture content of slab as the argument variables and aim to optimize MOR, MOE and IB of MDF. Some def-initions for the kernel functions are given, which let the corresponding optimization models not only can be solved effectively, but also have the strong application values. Second, we design an effective intelligent algorithm for solving these optimization problems, where its dynamical and optimal properties are also proved theoretically. Finally, we consider two classes of parameter optimizing problems, one aims at optimizing one performance index of MDF, another is for com-prehensively optimizing the three performance indexes of MDF. Based on the obtained relation models of the continuous hot-pressing system for the three kinds of MDF with different thickness, we give a theoretical suggestion on the setting of hot-pressing pressure, hot-pressing temperature, hot-pressing time and moisture content of slab to let the MDF meet the given requirements.4. In order to improve the practical values of this thesis, based on the obtained theoretical results, we design a simple simulation software for the MDF continuous hot-pressing automatic control system in MATLAB software. On the basis of the imported date set, this simulation soft-ware not only can build up the relation models with respect to the necessary technical parameters and MDF performance indexes, but also can provide a reliability assessment on the main factors in the hot-pressing process for the staff. So this simulation software is useful for improving the qual-ities of the MDF, reducing the energy consumption and improving the utilization rate of materials.