Benchmark value determination of energy efficiency indexes for coal-fired power units based on data mining methods

The operational optimisation of coal-fired power units is important for saving energy and reducing losses in the electric power industry. One of the key issues is how to determine the benchmark values of the energy efficiency indexes of the units. Therefore, a new framework for determining these benchmark values is proposed, based on data mining methods. First, the energy efficiency key performance indicators (KPIs) associated with the net coal consumption rate (NCCR) were selected based on the domain knowledge. Second, the decision-making samples with minimal NCCR were acquired with the fuzzy C-means (FCM) clustering algorithm, and the corresponding clustering centres were employed as the benchmark values. Finally, based on the support vector regression (SVR) algorithm, the target values of the NCCR were obtained with the KPIs as input, and the energy saving potential was evaluated by comparing the target values with the historical values of the NCCR. An actual on-duty 1000 MW unit was taken as study unit, and the results show that the energy saving potential is remarkable when the operators adjust the KPIs based on the calculated benchmark values.     

基于加权中智C均值算法的变压器故障诊断模型

为提高变压器故障诊断准确度,提出了一种基于加权中智C均值算法的变压器故障诊断方法。该方法利用基于样本相似度的加权方法对样本特征进行加权,再引入中智理论对样本的分布重新分配,建立起基于加权中智C均值算法的变压器故障诊断模型。研究结果表明,该方法不仅弥补了传统FCM相同权重分配的不足,有效提高了故障诊断的准确率,且诊断结果产生的中智点对故障的变化预测具有重要意义。     

联合模糊聚类和多样本SSO优化SVR的配电通信网络流

针对电力通信网络流量预测问题，提出了一种联合模糊聚类和多样本群居蜘蛛优化（social spider optimization，SSO）SVR的配电网络流量预测方案。针对配电网络流量时序非线性、周期性特点，首先采用自适应模糊聚类方法（adaptive fuzzy C-means clustering）对流量样本数据进行预处理，实现了样本数据聚类自动划分，有效降低了流量数据短相关性对预测精度的影响；然后利用多样本SSO优化算法（multi-sample social spider optimization algorithm）对SVR预测模型参数进行优化，通过引入多样本、网格迭代进化策略，从而得到不同流量数据聚类对应的最佳SVR参数组合；最后，运用多样本SSO优化SVR模型对预测数据进行预测分析。仿真结果表明，同ARIMA、神经网络等配电网路流量预测模型相比，提出的预测方案预测精度提高了18.8%~34.1%。     

模糊数学在零件分组中的应用研究

在充分研究减法聚类法、模糊C均值聚类法和最大隶属度原则的基础上,针对机械零件分组中的模糊性,将以上的方法或者原则应用到其中,并结合工程实际情况作适当修改,使其更符合工程实际。通过实例验证了此方法的正确性和实用性,并分析了这种方法的优缺点。     

H2-selective mixed matrix membranes modeling using ANFIS,PSO-ANFIS,GA-ANFIS

The novel contribution of the current study is to employ adaptive neuro-fuzzy inference system (ANFIS) for evaluation of H₂-selective mixed matrix membranes (MMMs) performance in various operational conditions. Initially, MMMs were prepared by incorporating zeolite 4A nanoparticles into polydimethylsiloxane (PDMS) and applied in gas permeation measurement. The gas permeability of CH₄, CO₂, C₃H₈ and H₂ was used for ANFIS modeling. In this manner, the H₂/gas selectivity as the output of the model was modeled to the variations of feed pressure, nanofiller contents and the kind of gas, which were defined as input (design) variables. The proposed method is based on the improvement of ANFIS with genetic algorithm (GA) and particle swarm optimization (PSO). The PSO and GA were applied to improve the ANFIS performance. To determine the efficiency of PSO-ANFIS, GA-ANFIS and ANFIS models, a statistical analysis was performed. The results revealed that the PSO-ANFIS model yields better prediction in comparison to two other methods so that root mean square error (RMSE) and coefficient of determination (R²) were obtained as 0.0135 and 0.9938, respectively. The RMSE and R² values for GA-ANFIS were 0.0320 and 0.9653, respectively, and for ANFIS model were 0.0256 and 0.9787, respectively.     

Traffic state recognition using static images and FCM

Accurate recognition of the traffic condition can proactively alert drivers who will enter the congested road to avoid congestion, so that the degree of congestion will not be increased. And it is also the basis to make scientific decision on active traffic managements, and conducive to alleviate congestion, improve the traffic efficiency, save energy and reduce emission. In this paper, the traffic surveillance videos are sampled every three minutes to build static image database, and the road area is marked as the region of interest (ROI), and then ROI images are normalized in terms of angle and scale. The three image features in ROI, i.e., average gradient, corner and long edge number, are then extracted. Finally, the fuzzy C-means clustering (FCM) method is used to classify the traffic condition into two classifications, i.e., flowing traffic and congestion. Experimental results show that the proposed algorithm can effectively identify the traffic condition involved in the image by the accuracy of 98％. Moreover, compared with the video-based approaches, this method greatly reduces the implementation cost.     

基于改进半监督模糊C-均值聚类的发动机磨损故障诊断

为解决在少量油液样本条件下发动机磨损故障诊断难的问题,提出一种改进半监督模糊C-均值聚类算法(Improvedsemi-supervised fuzzy c-means clustering algorithm,ISS-FCM).定义一种优化的目标函数,将无标签样本与训练样本间的平均距离度量考虑在内并赋予其一定权值,以...     

一种面向订单剩余完工时间预测的SOM-FWFCM特征选择算法#br#

准确的订单剩余完工时间预测有助于动态调整生产计划、优化制造过程,以满足订单产品按时交付的需求。订单剩余完工时间受到车间物料、设备、在制品等各类生产要素的综合影响,相关数据具有典型的大量、多维、高冗余的特点,有效的特征选择能够获得更高的预测精度。在构建候选特征集的基础上,提出了一种基于自组织映射（SOM）网络特征加权模糊C均值（FWFCM）的特征选择算法。通过构建SOM网络初始化FWFCM的聚类中心,减少后者对初始聚类中心的依赖;基于互信息计算特征权重,实现导向性特征聚类,根据聚类结果选择特征代表,构成高质量关键特征子集。以某机加工车间的生产数据为例,通过与其他4种特征选择算法的对比分析,验证了所提算法的有效性。     

尺度方向自适应的减法聚类视频运动目标定位

针对减法聚类算法对视频运动目标进行定位时无法获取目标尺度及方向参数的问题,本文提出了一种可获取待定位目标尺度及方向参数的视频运动目标定位算法.该算法在减法聚类算法预定位目标位置及获得目标个数的基础上,进一步采用模糊C均值聚类对目标前景样本进行归类,最后通过对目标前景样本协方差矩阵特征值和特征向量的分析获得目标的尺度及方向参数,从而实现对视频运动目标的定位.实验结果表明,所提出的方法与原减法聚类定位方法相比可获得更合理的目标定位结果.     

基于小波和模糊C-均值聚类算法的汽油机爆震诊断研究

提出了一种小波变换和模糊C-均值聚类相结合的汽油机爆震诊断方法,并对某汽油机进行了爆震试验,利用小波变换方法从汽油机振动信号中提取了轻微爆震特征。以振动信号小波变换后的特征子带信号为样本信号,以特征子带信号的4个时域参数作为燃烧模式特征值,利用模糊C-均值聚类方法对燃烧模式进行分类,用贴近度方法进行爆震诊断。结果表明:小波结合模糊C-均值聚类方法能够利用振动信号有效地分类燃烧模式,并能诊断出轻微爆震。