基于模糊神经的多路模温控制方法研究

发布时间：2018-04-10 05:14

本文选题：压铸模具　切入点：温度控制　出处：《沈阳理工大学》2017年硕士论文

【摘要】：近些年来,随着压铸行业的不断发展,即使在经济不是很景气的情况下,因为压铸模具的产品质量、生产工艺、产品的精密程度等有所提高,使得压铸市场仍保持比较健康的状态。根据市场调查显示,压铸模具的产值每年在稳步上升。汽车、航天等行业的迅猛发展使得对模具需求量大大提高,对我国压铸行业来说,机遇与挑战并存。然而我国压铸模具的制造水平与发达国家还是有一定差距,当务之急是要自主研发并改进模具的各项指标,而不是一味的去仿造国外产品。压铸模具产品的好坏,很主要的一个指标就是压铸过程中模具的温度,这对产品的质量及模具的寿命有着至关重要的影响。压铸模具温度的主要特点是冷却过程较慢,耗能大,如果得不到有效控制,严重时会直接影响产品的质量,这就要求在压铸过程中,温度要保持在一定的范围内。模具温度具有大延时、大惯性、非线性的特点,并且没有精确的数学模型,传统的控制方法不能达到精确控制的效果。基于这个问题,本论文设计了基于模糊神经网络的多路模具温度控制系统。因为压铸模具是由个多个模具块组合而成的,所以对每块模具设计了独立的循环冷却水回路,通过调节各回路冷却水阀门开通时间长短,来使模具降温,最终达到控温目的。模糊神经网络是近些年来控制领域研究的一个热点,模糊控制和神经网络都是基于无模型、非线性系统,其中模糊系统善于表达模糊、定性知识,缺乏自适应能力和自学习,而神经网络有很好的学习能力,但不能利用基于规则的知识,结合两者优点设计一种基于模糊神经网络控系统,通过MATLAB仿真,与模糊控制相比较,系统响应更快,精度更高,效果更好。
[Abstract]:In recent years, with the continuous development of the die casting industry, even when the economy is not very prosperous, the quality of the die casting die, the production process, the precision of the products have been improved.So that the die-casting market to maintain a relatively healthy state.According to market research, the output value of die-casting dies is rising steadily every year.With the rapid development of automobile, aerospace and other industries, the demand for dies is greatly increased, and opportunities and challenges coexist for the die casting industry in China.However, the manufacturing level of die casting dies in our country is still far from that in developed countries. The urgent task is to develop and improve the die indexes independently, instead of blindly copying foreign products.The temperature of die casting process is one of the most important indexes of die casting products, which has a vital effect on the quality of products and the life of dies.The main characteristic of die casting mold temperature is that the cooling process is slow and the energy consumption is large. If it is not effectively controlled, it will directly affect the quality of the product when it is not effectively controlled, which requires that the temperature should be kept within a certain range during the die casting process.The mold temperature has the characteristics of long delay, large inertia and nonlinear, and there is no accurate mathematical model. The traditional control method can not achieve the effect of accurate control.Based on this problem, a multi-channel mold temperature control system based on fuzzy neural network is designed in this paper.Because the die casting mould is composed of several die blocks, an independent circulating cooling water circuit is designed for each die. By adjusting the opening time of each circuit cooling water valve, the mold is cooled down and the temperature control is finally achieved.Fuzzy neural network (FNN) is a hot topic in the field of control in recent years. Both fuzzy control and neural network are based on modelless, nonlinear systems, in which fuzzy systems are good at expressing fuzzy, qualitative knowledge, lack of adaptive ability and self-learning.The neural network has good learning ability, but it can't use the rule-based knowledge to design a fuzzy neural network control system based on the advantages of both. Through MATLAB simulation, compared with the fuzzy control, the system response is faster and the accuracy is higher.It works better.
【学位授予单位】：沈阳理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG241;TP273

【参考文献】