基于灰色粒子群算法的温室环境多目标优化控制研究

发布时间：2018-05-12 07:16

本文选题：温室 + 模型　；参考：《安徽农业大学》2017年硕士论文

【摘要】：温室环境控制是作物监测的内容之一,对于其品质、产量等具有重要意义。合理控制温室设备,使温室内的环境参数更好的满足作物生长是目前温室环境控制需要解决的重要问题。为了合理调控温室环境控制设备,一定程度上节约用电成本。本文以安徽农业大学农萃园的茶树育苗温室为研究对象,通过引入人工控制因素,以扩展的自回归模型(Auto Regressive eXogenous,ARX)为基础,构建温度、湿度及耗电量多目标模型函数。在标准粒子群算法(Particle Swarm Optimization,PSO)基础上,结合灰色关联理论概念,面向温室环境进行多目标调控。主要研究内容和结果如下:(1)针对温室环境的空间特征,对茶苗温室进行多源信息采集。多点采集温室大棚环境温度和湿度信息,运用自适应加权融合估计算法对温室多源采集信息进行融合,完成大棚环境多源因子在数据层的融合。运用LabVIEW开发软件采集大棚环境信息,利用PH气象站采集大气环境信息。通过小波降噪和自适应加权融合估计算法对采集数据进行预处理,有效去除信息采集过程中存在的噪声,保证温湿度数据的可信度,为温室环境建模做准备。(2)构建育苗温室环境的温度、湿度及能耗模型。通过引入人工控制因素,围绕ARX模型结构,运用系统辨识的方法辨识出模型的结构和参数,构建育苗温室环境的温度、湿度模型。运用交叉验证的方式检验温度和湿度模型的准确性,仿真结果表明模型计算得到的温湿度与实测的温湿度变化趋势一致,说明ARX模型能有效模拟育苗温室内的温度和湿度信息;以调控机构运行消耗的电量为参考建立耗电量模型。(3)算法优化控制。通过引入灰色关联度理论,在标准PSO算法的基础上,将调控设备组合种类视为粒子的解,以温度模型、湿度模型及能耗模型为目标函数,以此完成温室环境控制的多目标优化控制。将本文算法优化得到的温湿度与线性加权和法、单目标PSO优化得到的结果相比对,发现选取本文方法不仅能够使温室环境的温湿度在作物适宜的生长范围之内,相对于其余两种优化方法在一定程度上节约了用电成本。
[Abstract]:Greenhouse environmental control is one of the contents of crop monitoring, which is of great significance to its quality and yield. It is an important problem for greenhouse environmental control to control the greenhouse equipment reasonably and make the environmental parameters of greenhouse better meet the crop growth. In order to control the greenhouse environment control equipment reasonably, save electricity cost to a certain extent. In this paper, based on the expanded autoregressive model (Auto Regressive eXogenous-ARX), the temperature, humidity and power consumption multiobjective model functions are constructed by introducing artificial control factors into the tea seedling greenhouse in the Sui Garden of Anhui Agricultural University. Based on the standard particle swarm optimization algorithm (PSO) and the concept of grey correlation theory, multi-objective regulation is carried out in greenhouse environment. The main contents and results are as follows: (1) according to the spatial characteristics of greenhouse, the tea seedling greenhouse is collected with multi-source information. Multi-point collection of greenhouse environment temperature and humidity information, using adaptive weighted fusion estimation algorithm to the greenhouse multi-source information fusion, the greenhouse environment multi-source factors in the data level fusion. The environment information of greenhouse is collected by LabVIEW software and atmospheric environment information is collected by PH weather station. The wavelet denoising and adaptive weighted fusion estimation algorithm are used to preprocess the collected data, which can effectively remove the noise existing in the process of information acquisition and ensure the reliability of the temperature and humidity data. The temperature, humidity and energy consumption model of greenhouse environment were constructed. By introducing artificial control factors around the structure of ARX model, the structure and parameters of the model are identified by the method of system identification, and the temperature and humidity models of greenhouse environment are constructed. The accuracy of the temperature and humidity model was verified by cross validation. The simulation results showed that the temperature and humidity calculated by the model were consistent with the measured temperature and humidity, which indicated that the ARX model could effectively simulate the temperature and humidity information in seedling raising greenhouse. The optimal control algorithm is established based on the model of power consumption. Based on the standard PSO algorithm, this paper introduces the grey correlation degree theory, regards the type of the control equipment as the solution of particles, and takes the temperature model, humidity model and energy consumption model as the objective functions. Thus the multi-objective optimal control of greenhouse environment control is completed. Comparing the temperature and humidity obtained by this method with the linear weighted sum method and the results obtained by single objective PSO optimization, it is found that the selection of this method can not only make the temperature and humidity of greenhouse environment within the suitable growth range of crops. Compared with the other two optimization methods, the cost of electricity is saved to some extent.
【学位授予单位】：安徽农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;S625

【参考文献】