【作者】 王英；

【导师】 曹军； 孙丽萍；

【作者基本信息】 东北林业大学 ， 检测技术与自动化装置， 2013， 硕士

【摘要】 当今全球森林资源日益减少,如何提高林业资源的利用率已成为木材加工业面临的难题之一。木材干燥是木材加工处理的重要环节,运用完善的干燥设备和先进干燥技术不但能大量节约木材资源、提高木材品质,还能起到节能环保作用。高频真空联合木材干燥是一种干燥速度快、能源消耗低、环境污染小的新型联合干燥技术。该技术具备高频干燥和真空干燥各自的优点：在减少木材干燥过程中由于局部温度过高而发生开裂或烧焦可能性的同时,缩短了干燥时间：另外,还保证了木材的干燥品质。所以该技术在木材干燥领域有广阔的应用价值和发展前景。由于木材干燥过程具非线性、时变性及强耦合性等特点,依据传统的理论基础很难建立精确的数学模型,进而很难达到自动控制的效果。本文以BP神经网络为基础,采用遗传算法(Genetic Algorithm, GA)和将其与模拟退火算法(Simulated Annealing,SA)相结合的遗传退火算法(SAGA)优化BP网络模型,为高频真空联合木材干燥过程建立了温度控制模型和干燥基准模型。并通过仿真实验验证得到：基于SAGA优化BP神经网络的温度控制模型和干燥基准模型都有较高的预测精度；与GA优化BP网络模型相比,更适合描述复杂的木材干燥过程,为设计精确的空制系统和优化干燥基准奠定了基础。在木材高频真空联合干燥过程的理论分析基础上,基于已建立的木材干燥模型,设计了木材干燥模糊控制器和模糊神经网络控制器。在MATLAB/SIMULINK环境下对两种控制方法进行了仿真实验,结果表明木材干燥的模糊神经网络控制有更好的控制效果,即温度上升更快,控制精度更高,稳定性更好,这对于木材干燥领域从手工、半自动控制向全自动控制的迈进起到了推动作用。更多还原

【Abstract】 Today’s global forest resources dwindling, how to improve the forest resource utilization has become one of the difficulties faced by the wood processing industry. Wood drying is an important link in wood processing, with perfect drying equipment and advanced drying technology; it can not only save timber resources, improve the quality of the wood, but also have an energy conservation and environmental protection effect.High-frequency vacuum combined wood drying is a new union wood drying technology which is using less time, less power and lower pollution to the environment. This technology has the respective advantages of high frequency drying and steam drying, it’s not only reducing the possibility of cracking occurs or burning because of the local temperature is too high in the wood drying process, but also shorten the drying time. At the same time, it’s also ensured the drying quality of the wood. So in the wood drying area, this technology has broad application value and development prospects.Wood drying process has the characteristics of nonlinear, time-varying and strong couplings, o it’s difficult to using the traditional theoretical to establish a precise mathematical model for hitting the effects of automation-control. This paper is using the Genetic Algorithms and genetic algorithm that is the main process of SAGA algorithm to establish the temperature control model and drying baseline model for High-frequency vacuum combined wood drying, which is based on BP neural network. Simulation results show the following conclusions: temperature control model and drying baseline model are both having higher prediction accuracy which is based on the BP neural network that is optimized by the SAGA algorithm. Comparing with the BP neural network that is optimized by the GA algorithm, SAGA algorithm is more appropriate for describing the complicated process of wood drying, and it’s also the essential preparation of designing the precise control system and optimized drying baseline model.The wood drying model has been built through the theoretical analysis of High-frequency vacuum combined wood drying. And this paper also has designed wood drying fuzzy controller and fuzzy Neural Network Controller. The two control methods were simulated in MATLAB/SIMULINK environment and got the result:Fuzzy Neural Network Control could get better control effect in the wood drying process, for instance, the temperature could rise faster, for the field of wood drying from manual, semi-automatic control to automatic control forward has played a role in promoting.更多还原