A novel approach to analyse incomplete design of experiments – Application to the study of the influence of operational parameters on the performance of a solid oxide fuel cell based micro-combined heat and power system

A SOFC based commercial μ-CHP system is characterized by Electrochemical Impedance Spectroscopy, using a 2⁴ full factorial test plan. The studied factors are: natural gas input power, ratio between oxygen and natural gas flow rates at the reformer inlet, stack average temperature and average operating cell voltage. Six replicates are performed in the domain centre. We performed equivalent circuit analysis and extracted three responses from each spectrum: ohmic resistance together with the two parameters of the CPE used in the model.However, one of our experiment is an outlier. To circumvent this problem, two methods described in the literature were applied: recalculation of missing response and introduction of a dynamic variable. Due their unsatisfactory results, we developed an innovative approach combining an iterative fitting of the multilinear model underlying any factorial design and an N-way ANOVA. Our method is successfully validated on the different 2⁴⁻¹ fractional designs deriving from the full factorial one.The only impacted response is the ohmic resistance. It increases as temperature decreases or as applied voltage increases. It is impacted by a strong synergistic effect of pressure and temperature and a compensating effect of pressure and applied voltage. No significant quadratic effect is observed.     

Rule-based OneClass-DS learning algorithm

One-class learning algorithms are used in situations when training data are available only for one class, called target class. Data for other class(es), called outliers, are not available. One-class learning algorithms are used for detecting outliers, or novelty, in the data. The common approach in one-class learning is to use density estimation techniques or adapt standard classification algorithms to define a decision boundary that encompasses only the target data. In this paper, we introduce OneClass-DS learning algorithm that combines rule-based classification with greedy search algorithm based on density of features. Its performance is tested on 25 data sets and compared with eight other one-class algorithms; the results show that it performs on par with those algorithms.     

A survey of outlier detection in high dimensional data streams

The rapid evolution of technology has led to the generation of high dimensional data streams in a wide range of fields, such as genomics, signal processing, and finance. The combination of the streaming scenario and high dimensionality is particularly challenging especially for the outlier detection task. This is due to the special characteristics of the data stream such as the concept drift, the limited time and space requirements, in addition to the impact of the well-known curse of dimensionality in high dimensional space. To the best of our knowledge, few studies have addressed these challenges simultaneously, and therefore detecting anomalies in this context requires a great deal of attention. The main objective of this work is to study the main approaches existing in the literature, to identify a set of comparison criteria, such as the computational cost and the interpretation of outliers, which will help us to reveal the different challenges and additional research directions associated with this problem. At the end of this study, we will draw up a summary report which summarizes the main limits identified and we will detail the different directions of research related to this issue in order to promote research for this community.     

Using intelligent data analysis to detect abnormal energy consumption in buildings

This paper describes a novel method for detecting abnormal energy consumption in buildings based on daily readings of energy consumption and peak energy consumption. The method uses outlier detection to determine if the energy consumption for a particular day is significantly different than previous energy consumption. For buildings with abnormal energy consumption, the amount of variation from normal is determined using robust estimates of the mean and standard deviation. This new data analysis method will reduce operating costs by detecting problems that previously would have gone unnoticed. Also, operators should save time by not having to manually detect faults or diagnose false alarms. The new data analysis method has successfully detected high-energy consumption in many buildings. This paper presents field test results for buildings that had the following problems: (1) chiller failure and a poor control strategy, (2) poor design of ventilating and air-conditioning equipment, and (3) improper operation of equipment following a change in the electrical panel.     

基于GPS/SINS紧耦合系统的历史故障次优融合检测法

为提高GPS/SINS紧耦合系统对于软故障的检测能力,在新息外推法的基础上提出一种历史故障次优融合算法。通过对历史故障进行加权次优融合,形成检验统计量进行故障检测,并提出野值诊断方法,解决该算法易受野值影响的问题。在GPS/SINS紧耦合系统基础上进行对比仿真,结果表明,历史故障次优融合算法在故障量越小的情况下,软故障检测效果越明显,而且能更好地消除野值对故障检测的影响。     

Research on outlier detection algorithm for express logistics

To detect the problems of time delay, path error and destination error in express logistics process effectively, a novel outlier detection algorithm for express logistics is proposed in this paper. To test the detection results, the express logistics system operating model is built to test the detection results. Experiment results show that the proposed algorithm is well applied to the express logistics data with multi-attribute characteristics, and can work well in detecting the abnormal conditions of express logistics.     

Characterization of host response to bruchids (Callosobruchus chinensis and C. maculatus) in 39 genotypes belongs to 12 Cajanus spp. And assessment of molecular diversity inter se

Thirty nine genotypes belong to 12 Cajanus species were characterized for host response to bruchids (C. chinensis and C. maculatus) using no-choice based in vivo and in vitro antibiosis assay, and these assays revealed seven genotypes of primary gene pool (C. cajan) were more susceptible to bruchid infestation as compared to the genotypes of secondary and tertiary gene pools. Among all the genotypes, C. cajanifolius acc. ICPW-31 showed higher degree of resistance to both the bruchids. The molecular diversity using CDDP and SCoT markers showed wide range of genetic variations among the 12 species of Cajanus, and was supported by estimates of Nei's genetic distance and fixation index (F_ST) based haplotype matrix. The genetic structure showed clustering of 39 genotypes into eight distinct groups (K = 8) on the basis of their allelic composition, and among them C. cajanifolius acc. ICPW 31 and ICPW 30 showed close affinity with the cultivars of C. cajan by sharing several alleles. AMOVA analyses showed the existence of higher extent of genetic variation both at the genotype (37%) and species (63%) level in the genus Cajanus. The dendrogram and the principal coordinate analysis (PCoA) placed the 39 genotypes into six major clusters at par with their sectional classification. MCheza and ARLEQUIN based outlier analysis revealed 13 common loci under balancing selection, which are supposed to be involved in maintenance of genetic polymorphism either at species or genotype level, and are assigned to putative gene families such as KNOX, WRKY, ERF, MYB, ABP1 and MADS. The in vivo and in vitro bruchid assay vis-à-vis the DNA marker based molecular diversity analysis affirmed the possible use of C. cajanifolius acc. ICPW-31 as donor genotype for the introgression of bruchid resistance allele(s) into cultivated genetic background.     

Local Influence of Predictors in Multiple Linear Regression

Local influence diagnostics can be used to assess the influence of predictor values in multiple linear regression. For n observations and k regressors, an eigenanalysis of an nk ×nk matrix is required to assess the influence on the estimated coefficients. We provide the analytic expressions for the eigenvectors and show that they are easily computed, describe influence on the parameter estimates of a principal components regression, and are related to leverage, outliers, and added-variables plots. The results indicate that multicollinearity and overfitting contribute to a fitted model's sensitivity, leading to strategies for model assessment and selection.     

Machine Learning Supports Robust Operation of Thermosiphon Reboilers

The analysis of process and equipment operational data in chemical engineering regularly requires a high level of expert knowledge. This work presents a Machine Learning-based approach to evaluate and interpret process data to support robust operation of a thermosiphon reboiler. By applying an outlier detection, potentially interesting and unstable operating conditions can be identified quickly. A multidimensional regression allows to forecast the circulating mass flow. The results obtained fit well into the current state of research and manual evaluation of thermosiphon reboilers.     

Efficient and flexible algorithms for monitoring distance-based outliers over data streams

Anomaly detection is considered an important data mining task, aiming at the discovery of elements (known as outliers) that show significant diversion from the expected case. More specifically, given a set of objects the problem is to return the suspicious objects that deviate significantly from the typical behavior. As in the case of clustering, the application of different criteria leads to different definitions for an outlier. In this work, we focus on distance-based outliers: an object x is an outlier if there are less than k objects lying at distance at most R from x. The problem offers significant challenges when a stream-based environment is considered, where data arrive continuously and outliers must be detected on-the-fly. There are a few research works studying the problem of continuous outlier detection. However, none of these proposals meets the requirements of modern stream-based applications for the following reasons: (i) they demand a significant storage overhead, (ii) their efficiency is limited and (iii) they lack flexibility in the sense that they assume a single configuration of the k and R parameters. In this work, we propose new algorithms for continuous outlier monitoring in data streams, based on sliding windows. Our techniques are able to reduce the required storage overhead, are more efficient than previously proposed techniques and offer significant flexibility with regard to the input parameters. Experiments performed on real-life and synthetic data sets verify our theoretical study.