共查询到19条相似文献,搜索用时 359 毫秒
1.
针对现有的以概率统计理论为基础的方法和模糊神经网络法必须建立在大量统计数据基础之上,以及模糊信息扩散估计法可能对器件失效阈值估计过高的问题,提出将模糊信息处理技术用于对原始实验数据的处理,得到训练样本,在此基础上利用支持向量机回归预测一定功率的高功率微波辐照条件下电子器件的损伤概率。仿真结果表明:该方法与模糊神经网络法都较好地给出了预测结果,但该方法具有更高的精度(均方根误差为7.40610-5),并且克服了在样本数据减半的小样本情况下模糊神经网络法可能出现野值的缺陷。 相似文献
2.
3.
4.
单粒子效应是航天电子器件失效的重要原因,研究其物理过程对航天电子器件寿命预测、器件抗辐照加固有着重要的意义。现有的模型多从线性能量沉积的角度来解释单粒子翻转,因此无法解释单粒子效应地面实验模拟过程中的温度效应。建立了一个新模型,从高能带电离子与材料相互作用的物理过程出发,通过解三维的热扩散方程,计算出能量在材料中沉积、交换、扩散,得到电子和晶格温度的空间分布以及时间演化过程。推断出离子辐照过程中导致的自由电子浓度和收集电荷随LET的变化关系。此模型解释了单粒子效应中随着器件温度升高,单粒子效应截面增加的现象。 相似文献
5.
提出一种对光参量呈非单调下降规律的LED灯珠可靠性进行评价的方法.采用加速寿命实验获得光通量退化数据,利用指数叠加形式的退化模型对光通维持率退化数据进行拟合,与指数模型拟合效果相比,该模型具有更好的效果.用MATLAB软件计算样品的伪失效寿命,通过KolmogorovSmirnov检验法得到两个公司样本伪失效寿命分布分别服从对数正态分布和威布尔分布,以相应分布参量评估产品可靠性得到两个公司样本的伪失效寿命分别为5 328.37h和4 758.35h.该方法对参量呈非单调下降规律的LED器件可靠性的评估具有参考价值. 相似文献
6.
提出了一种针对室内场景的轻量化端到端单目深度估计神经网络。首先设计了新型自适应深度分布估计模块,可针对不同输入图像估计差异化的深度范围,使得网络更好地预测室内物体的相对位置关系,恢复出的深度图像能获得更接近真实值的像素分布。其次,在深度估计的过程中,通过基于平面系数的深度间接表示形式加入平面隐式约束,可以在场景的平面区域得到更平滑的深度估计结果。在NYU Depth-v2数据集上的多项实验结果表明,提出方法能满足较高分辨率下的实时性要求,同时能以更少的参数恢复出质量更高、更完整的室内深度图像,有助于实现更加准确的三维重建效果。 相似文献
7.
提出一种模糊边界模块化神经网络(FBMNN)的混沌时间序列预测方法,该方法先对混沌时间序列观测点重构的相空间进行模块化划分,划分点的选取由遗传算法自动寻优.然后定义一个模糊隶属度函数,在划分边界一侧按照一定的模糊隶属度设定模糊边界带,通过模糊化处理,解决了各模块划分点附近预测结果的跳跃问题.最后每一模块,及其模糊边界的样本点都对应一个递归神经网络进行训练,通过预测合成模块输出结果.该方法对三个混沌时间序列基准数据集Mackey-Glass,Lorenz,Henon进行实验,结果表明该方法有效地提高了混沌时间序列预测效果.
关键词:
模糊边界
模块化神经网络
混沌时间序列
预测 相似文献
8.
9.
本文利用模糊神经网络直接从实验数据中提取规则,进行材料性能建模与预测,作为应用示例,建立了基于成份和组织参数的灰铁预测模型,与多元统计分析,模糊回归和广义回归网络所得的结果相比,该方法所得的模型具有学习精度高,且具有更好的泛化能力。 相似文献
10.
空间辐射环境会对电子器件产生辐射损伤。由于商用器件性能普遍优于抗辐射加固器件,所以从商用器件中筛选出抗辐射性能优异的器件将在一定程度上提高空间电子系统的可靠性。结合数学回归分析与物理应力实验的方法,研究了集成电路抗辐射性能无损筛选技术。通过不同的外界能量注入及总剂量辐照实验,探究电路典型参数的应变情况与电路耐辐射性能的关系,并确定其辐射敏感参数;建立预测电路抗辐射性能的多元线性回归方程,并对应力条件下的回归方程进行辐照实验验证。结果显示,物理应力实验与数学回归分析结合的筛选方法减小了实验值与预测值的偏差,提高了预估方程的拟合优度和显著程度,使预估方程处于置信区间。 相似文献
11.
12.
This paper presents an application of fuzzy and neural network techniques to a vision-guided closed loop control for stationary luminous flames. The image processing technique is used to analyze and identify the process states. Fuzzy control strategy avoids the difficulty in establishing a mathematical model for an ill-defined process. Expert knowledge and training patterns can be incorporated into fuzzy rules, which are represented in the form of neurons. The use of a neural network makes it easy to increase the number of control parameters and provides the system the possibility to adjust its performance automatically. 相似文献
13.
Bad meteorological conditions may reduce the reliability of power communication equipment, which can increase the distortion possibility of fault information in the communication process, hence raising its uncertainty and incompleteness. To address the issue, this paper proposes a fault diagnosis method for transmission networks considering meteorological factors. Firstly, a spiking neural P system considering a meteorological living environment and its matrix reasoning algorithm are designed. Secondly, based on the topology structure of the target power transmission network and the action logic of its protection devices, a diagnosis model based on the spiking neural P system considering the meteorological living environment is built for each suspicious fault transmission line. Following this, the action messages of protection devices and corresponding temporal order information are used to obtain initial pulse values of input neurons of the diagnosis model, which are then modified with the gray fuzzy theory. Finally, the matrix reasoning algorithm of each model is executed in a parallel manner to obtain diagnosis results. Experiment results achieved out on IEEE 39-bus system show the feasibility and effectiveness of the proposed method. 相似文献
14.
In this paper, an adaptive fuzzy neural controller (AFNC) for a class of unknown chaotic systems is proposed. The proposed
AFNC is comprised of a fuzzy neural controller and a robust controller. The fuzzy neural controller including a fuzzy neural
network identifier (FNNI) is the principal controller. The FNNI is used for online estimation of the controlled system dynamics
by tuning the parameters of fuzzy neural network (FNN). The Gaussian function, a specific example of radial basis function,
is adopted here as a membership function. So, the tuning parameters include the weighting factors in the consequent part and
the means and variances of the Gaussian membership functions in the antecedent part of fuzzy implications. To tune the parameters
online, the back-propagation (BP) algorithm is developed. The robust controller is used to guarantee the stability and to
control the performance of the closed-loop adaptive system, which is achieved always. Finally, simulation results show that
the AFNC can achieve favourable tracking performances. 相似文献
15.
16.
The theories of intelligent information processing are urgently needed for the rapid development of modem science. In this paper, a novel fuzzy chaotic neural network, which is the combination of fuzzy logic system, artificial neural network system, and chaotic system, is proposed. We design its model structure which is based on the Sigmoid map, derive its mathematical model, and analyse its chaotic characteristics. Finally the relationship between the accuracy of map and the membership function is illustrated by simulation. 相似文献
17.
针对葡萄酒品质预测模型难以建立的问题,提出一种基于模糊递归小波神经网络的葡萄酒品质预测模型。利用葡萄酒物理化学指标和品酒师打分作为模型的输入输出,采用梯度下降算法在线学习隶属函数层中心、宽度和小波函数平移因子、伸缩因子、自反馈权重因子以及输出层权值。仿真实验时,首先利用Mackey-Glass混沌时间序列进行了性能测试,然后利用UCI数据集葡萄酒品质数据对所建立的品质预测模型进行了验证。结果显示,与多层感知器、径向基函数神经网络等传统前馈神经网络相比,构建的模糊递归小波神经网络品质预测模型具有更高的预测精度,更加适合于葡萄酒的品质预测。 相似文献
18.
分析了快递超网络和电子元件超网络的相继故障扩散方式, 结合超图理论提出了2-section 图分析法和线图分析法, 并仿真分析了无标度超网络耦合映像格子的相继故障进程. 结果表明: 无标度超网络对外部攻击表现出了既鲁棒又脆弱的特性. 针对相继故障的不同扩散方式, 无标度超网络的相继故障行为表现出不同的特点. 超网络的相继故障行为和超网络的超度以及超边度分布有密切的联系, 也和超网络中超边的个数有关. 通过和同规模的Barabasi-Albert (BA)无标度网络对比, 在同一种攻击方式下同规模的无标度超网络都比BA 无标度网络表现出了更强的鲁棒性. 另外, 基于超边扩散的相继故障进程比基于节点扩散的相继故障进程更加缓慢. 相似文献
19.
Base on the principle of the superposition of waves, active noise control is achieved by adaptively tuning a secondary source which produces an anti-noise of equal amplitude and opposite phase with primary source. This paper presents the study on the acoustic attenuation in a duct by using the combination of fuzzy neural network with error back propagation algorithm to control secondary source. The most important advantage of fuzzy inference system is that the structured knowledge is represented in the form of fuzzy IF-THEN rules. But it lacks the ability to accommodate the change of external environments. Combining neural network with fuzzy system can help in this tuning process by adapting fuzzy sets and creating fuzzy rules. The performance of attenuation and control error can be measured by the microphone placed in the downstream of duct. The results of this study, show that the acoustic attenuation by 40 dB for pure-tone noise and nearly 30 dB for dual-tones noise are obtained. 相似文献