苹果可溶性固形物含量的检测方法——基于可见光近红外光谱技术

为了快速检测苹果的可溶性固形物(SSC)含量,采用可见光近红外光谱技术,结合主成分分析(PCA)和BP神经网络技术,来建立苹果SSC的预测模型.获取苹果样本在345～1039 nm波段的漫反射光谱,采用DPS数据处理系统对其进行主成分分析,并提取出累计可信度大于95 %的5个新主成分.建立一个3层的BP神经网络模型,并将这5个新的主成分作为BP神经网络模型的输入量,其结果是98 %以上预测样本的预测相对误差在5 %以下.该研究表明,采用近红外光谱技术来建立苹果可溶性固形物的预测模型是可行的.     

基于神经网络逆控制方法的水轮发电机组非线性控制器仿真研究

基于水轮发电机组刚性水击模型和神经网络逆系统控制方法,在分析励磁与导叶系统可逆性的基础上提出了一种新的水轮发电机组解耦控制器设计方法.理论分析和仿真结果表明该控制策略较好的解决了水轮发电机组控制时存在的多变量、非线性、强耦合问题;实现了系统的动态解耦控制,具有较好的适应性和鲁棒性,因此可有效增强输电系统的暂态稳定性.     

静止无功补偿器的智能自适应PID控制器设计

将神经网络和模糊控制与有着广泛应用ＰＩＤ控制相结合,设计了一种静止无功补偿器的智能自适应ＰＩＤ控制器。利用神经网络实现系统模型辨识,采用模糊逻辑和神经网络相结合对ＰＩＤ控制器参数动态寻优。使ＳＶＣ的控制既具有模糊控制的简单,有效的非线性控制作用,又具有神经网络的自学习,自适应能力。     

基于物联网的生猪运动行为及猪舍环境远程监控系统

In intensive and large-scale livestock breeding, how to evaluate welfare of each individual effectively has become one hotspot of welfare breeding.In this study, a remote monitor system based on computer vision technology and wireless sensor networks was proposed.In the system, pig videos in the piggery were captured by IP cameras and transmitted remotely to PC systems for objects tracking and trajectory extraction, and the piggery environmental data such as temperature, humidity and harmful gases were collected by wireless sensor networks(WSN) nodes.Traditional Camshift tracking algorithm was improved.The average root mean square errors of traditional and improved algorithm were 78.07 and 32.29, indicating the improved algorithm reduced error and improved tracking stability.The results of the real-time monitoring of environmental parameters by WSN showed that the hydrogen and ammonia concentrations, and the ambient temperature were all kept in the normal range, while the carbon dioxide concentration was exceeded, and the air humidity was slightly high.Through data mining, precise welfare monitoring and breeding management were realized.     

基于神经网络方法的土壤流失量预测研究

生产建设项目土壤流失量的预测直接关系到建设项目的水土保持分析、评价和防治措施体系的布局,目前常用的预测方法因其局限性、不合理性以及精度差等问题往往难以实现准确预测。将人工神经网络的BP算法引入到土壤流失量预测中,将降雨侵蚀力、土壤可蚀性、坡长、坡度、水土保持措施作为影响土壤流失量的主要因子,并以17个生产建设项目水土保持监测实例作为学习样本和检测样本,建立了基于神经网络方法的土壤流失量预测模型。预测结果表明,该模型拟合和预测精度高,具有很强的应用价值,能够满足工程应用需要。     

Research on immunization strategies based on independent sets of complex networks

According to the concept of independent set, this paper proposes a new immunization strategy of complex networks which immunizes the nodes in an independent set with maximum degree, called targeted immunization in the independent set. When the total degree of immune nodes in independent set is equal to the total degree of targeted immunization nodes in the whole network, the immunization strategy of independent set is more effective than the targeted immunization strategy of the whole network. Based on the network structure, this paper explains the reason. Comparing the random immunization of independent sets with the random immunization in the whole network, test results indicate that more nodes do not be removed with high degree for the random immunization of independent sets. For classic susceptible-infected(SI) model, all nodes in the network are only two states: susceptible or infected. In the study of immunization strategies, comparing SI model with SIR, SIS models. The use of SI model will be more favorable.     

基于地下水合理埋深的井渠结合灌区水资源联合调控

以泾惠渠灌区为例,针对井渠结合灌区地下水超采及地下水位上升而导致的农田灌溉水环境等问题,从水位调控水量的角度出发,提出了基于地下水位合理埋深的水资源调控模式。在确定不同水文地质单元和不同植被类型条件下地下水位合理埋深上下限的基础上,结合PSO-RBF神经网络对地下水位埋深预测的结果,设计了基本、节水两种水资源联合方案,对不同保证率下的灌区水资源进行了联合调控。结果表明:泾惠渠灌区地下水位合理埋深上限介于1.76~3.50m,下限介于8.7~25.0m,不同水文地质单元、植被类型的水位埋深上下限值差异较大;水资源联合调控时,局部地区出现地下水超采,需加强节水灌溉和多水源的联合调控。     

水果表面缺陷自动检测系统中的人工智能方法研究

通过计算机视觉技术获取了带有各种表面缺陷的苹果图像并进行预处理，采用自适应特征聚类（ＳＡＦＣ）神经网络与模糊加权决策树（ＦＷＤＴ）相结合的方法实现了缺陷区域的准确检测和详细分类。实验结果表明，用人工智能方法进行表面缺陷检测，具有良好的抗噪容错能力并能有效地克服传统图像分割方法适应性差的缺点，提高判别准确率和分类精度。     

基于Hopfeild神经网络的油菜籽脱皮冷榨压榨系数识别

在油菜籽脱皮侧限排油一维冷榨试验基础上，采用半固态饱和物料和滤饼物料物理模型分别建立两个理论压榨比模型，运用Hopfeild神经网络方法识别压榨系数。识别结果表明：脱皮油菜籽在压力低于10 MPa段为半固态状饱和物理模型，在压力高于20 MPa段则为滤饼状物理模型，对整个压榨过程，采用分段计算物理模型可有效提高整体压榨模型精度。脱皮冷榨的压榨系数远小于未脱皮冷榨的压榨系数。