烟气轮机涡轮盘高温合金剩余寿命预测

本文针对已运行60000h的Waspaloy合金烟气轮机涡轮盘进行剩余寿命预测分析,采用不同试验条件下得到的waspaloy合金持久寿命数据对人工神经网络模型进行训练,得到预测精度较高的模型参数,建立温度、应力等服役条件与持久断裂寿命之间的人工神经网络模型,并利用该模型对Waspaloy合金涡轮盘的剩余寿命进行预测分析。结果表明,中间层节点个数为15时,所建立的人工神经网络模型对Waspaloy合金的持久断裂寿命具有最好的统计预测精度,并可以良好地表征Waspaloy合金剩余持久寿命与服役条件间复杂的非线性关系。     

基于人工神经网络的建筑电气节能评估模型

针对建筑电气节能评估模型问题和南京地区某生活小区的实际情况,提出了电气节能改造的方法。同时,进行了节能计算,建立了基于人工神经网络的生活小区建筑电气节能评估模型。计算结果和仿真研究表明,该节能方法效果明显,节能评估模型正确,具有较好的应用价值。     

用自适应神经元网络进行短期电力负荷预测

本文提出了一个用来进行电力负荷预测的基于自适应卡尔曼滤波算法的人工神经元网络（ＡＮＮ）。给出了网络算法和它的改进形式。在预测过程中，权值按最近的负荷行为自适应改变。测试结果表明，该方法具有较高的精度和较快的速度。     

天气敏感型神经网络在电网短期负荷预测中的应用

针对运城地区电网易受气象影响的特点，应用了一种具有天气敏感性的基于快速反应BP算法的神经网络预测模型，模型中合理地考虑了影响负荷变化的气温和降水量等主要气象因素，使其能够适应天气的变化。对运城地区的实际负荷进行预测，结果较好地满足了现场要求．从而验证了该模型和算法的有效性。     

异种钢焊接接头高温时效碳迁移的系统辨识与预测

针对奥氏体不锈钢和铁素体类耐热钢两类母材金属的特点及相应的焊接工艺 ,分别对最小二乘法、有限差分法和人工神经网络法 (ANN)中的BP算法在异种焊接接头高温时效碳迁移的系统辨识与预测结果进行了分析和比较 ,提出了用改进的算法进行异种钢焊接接头高温时效碳迁移的状态辨识及预测     

GMM/ANN混合说话人辨认模型

通过分析GMM(高斯混合模型)的说话人辨认系统的性能,提出了一种捕捉不同说话人交互信息的人工神经网络(ANN)方法,构成一个GMM/ANN混合说话人辨认系统。实验表明,GMM/ANN混合系统的说话人辨认能够取得比基于GMM和基于MLP(多层感知器)更高的辨认率。     

ADAPTIVE HARMONIC CANCELLA-TION APPLIED IN ELECTRO-HYDR-AULIC SERVO SYSTEM WITH ANN

The method for harmonic cancellation based on artificial neural network (ANN) is proposed. The task is accomplished by generating reference signal with frequency that should be eliminated from the output. The reference input is weighted by the ANN in such a way that it closely matches the harmonic. The weighted reference signal is added to the fundamental signal such that the output harmonic is cancelled leaving the desired signal alone. The weights of ANN are adjusted by output harmonic, which is isolated by a bandpass filter. The above concept is used as a basis for the development of adaptive harmonic cancellation (AHC) algorithm. Simulation results performed with a hydraulic system demonstrate the efficiency and validity of the proposed AHC control scheme.     

基于Online LS-SVM的钢铁件淬火硬度在线检测

为了实现钢铁件淬火硬度的在线电磁无损检测,提出了在线最小二乘支持向量机（online least square support vector machine）的建模方法。Online LS-SVM是以增量学习训练SVM,以减量学习减少样本数,实现小样本估计的训练方法。实验结果表明,Online LS-SVM不仅能实现钢铁件淬火硬度的在线电磁无损检测,而且具有学习速度快,泛化性能好,对样本依赖程度低的优点。     

Time series forecasting by neural networks: A knee point-based multiobjective evolutionary algorithm approach

In this paper, we investigate the problem of time series forecasting using single hidden layer feedforward neural networks (SLFNs), which is optimized via multiobjective evolutionary algorithms. By utilizing the adaptive differential evolution (JADE) and the knee point strategy, a nondominated sorting adaptive differential evolution (NSJADE) and its improved version knee point-based NSJADE (KP-NSJADE) are developed for optimizing SLFNs. JADE aiming at refining the search area is introduced in nondominated sorting genetic algorithm II (NSGA-II). The presented NSJADE shows superiority on multimodal problems when compared with NSGA-II. Then NSJADE is applied to train SLFNs for time series forecasting. It is revealed that individuals with better forecasting performance in the whole population gather around the knee point. Therefore, KP-NSJADE is proposed to explore the neighborhood of the knee point in the objective space. And the simulation results of eight popular time series databases illustrate the effectiveness of our proposed algorithm in comparison with several popular algorithms.     

基于PSO-BP的软件缺陷预测模型

软件缺陷检测旨在自动检测程序模块中是否包含缺陷,从而加速软件测试过程,提高软件系统的质量。针对传统软件缺陷预测模型被限制在一定的应用范围而影响其预测的准确性和适用性,提出了一种基于PSO-BP软件缺陷预测模型。该模型运用粒子群优化算法优化BP神经网络的权值和阈值,采用交叉验证的方式进行实验,并与传统的机器学习方法J48和BP神经网络等方法进行了比较。实验结果表明提出的方法具有较高的预测准确性。