Energy inefficiency diagnosis in industrial process through one-class machine learning techniques

In the era of Industry 4.0, the ease of access to precise measurements in real-time and the existence of machine-learning (ML) techniques will play a vital role in building practical tools to isolate inefficiencies in energy-intensive processes. This paper aims at developing an abnormal event diagnosis (AED) tool based on ML techniques for monitoring the operation of industrial processes. This tool makes it easier for operators to accomplish their tasks and to make quick and accurate decisions to ensure highly efficient processes. One of the most popular ML techniques for AED is the multivariate statistical control (MSC) method; it only requires the dataset of the normal operating conditions (NOC) to detect and identify the variables that contribute to abnormal events (AEs). Despite the popularity of MSC, it is challenging to select the appropriate method for detecting and isolating all possible abnormalities a complex industrial process can experience. To address this limitation and improve efficiency, we have developed a generic methodology that integrates different ML techniques into a unified multiagent based approach, the selected ML techniques are supposed to be built using only the normal operating condition. For the sake of demonstration, we chose a combination of two ML methods: principal component analysis and k-nearest neighbors (k-NN). The k-NN was integrated into the proposed multiagent to take into account the nonlinearity and multimodality that frequently occur in industrial processes. In addition, we modified a k-NN method proposed in the literature to reduce computation time during real-time detection and isolation. Finally, the proposed methodology was successfully validated to monitor the energy efficiency of a reboiler located in a thermomechanical pulp mill.

     

Main-Chain Nematic Side-Chain Smectic Composite Liquid Crystalline Elastomers

A composite liquid crystalline elastomer is designed, combining main-chain and side-chain mesogenic polymers in the network, and resulting in micro-phase separated regions of nematic and smectic ordering in the macroscopically homogeneous elastomer. A range of different fractions of the components is explored, from fully nematic main-chain system, across to fully smectic side-chain elastomer. Thermal phase transitions of both phases coexisting in the material are detected by calorimetry, and the nematic/smectic structure investigated by X-ray scattering. The tensile stress–strain data reveal the key effect of such a multi-phase composite, where the nematic fraction adds ductility while the smectic fraction increases the modulus and mechanical stiffness. Varying the composition, the authors are able to optimize the mechanical properties of this material type.     

Experimental analysis of fuel from fish processing industry waste in a diesel engine

In the present work, biofuel derived from industrial fish processing industry waste is used in diesel engines to study its suitability . Biofuel from industry fish waste is produced through catalytic cracking, and its quality has been improved through distillation. A single cylinder 4.5 kW at 1500 rpm was used to find the suitability of biofuel and undistilled biofuel in diesel engine. Experimental results show that the brake thermal efficiency of biofuel and undistilled biofuel is similar. Brake thermal efficiency for diesel, undistilled biofuel and biofuel is 29.98, 32.12 and 32.4%, respectively, at 80% load. Carbon monoxide, unburnt hydrocarbons, particulate matter and oxides of nitrogen emissions increase with undistilled biofuel compared to biofuel. There is a small reduction in carbon dioxide emission with undistilled biofuel compared to biofuel. Even though the cylinder pressure is high with undistilled biofuel, the intensity of premixed combustion is lower than distilled biofuel. The ignition delay and combustion duration increase with undistilled biofuel. Finally, it is concluded that the fuel derived from fish processing industry waste can be used as a fuel for diesel engine after distillation.     

Relevance feedback revisited: dealing with content and structure in XML documents

Relevance feedback (RF) is a technique that allows to enrich an initial query according to the user feedback. The goal is to express more precisely the user’s needs. Some open issues arise when considering semi-structured documents like XML documents. They are mainly related to the form of XML documents which mix content and structure information and to the new granularity of information. Indeed, the main objective of XML retrieval is to select relevant elements in XML documents instead of whole documents. Most of the RF approaches proposed in XML retrieval are simple adaptation of traditional RF to the new granularity of information. They usually enrich queries by adding terms extracted from relevant elements instead of terms extracted from whole documents. In this article, we describe a new approach of RF that takes advantage of two sources of evidence: the content and the structure. We propose to use the query term proximity to select terms to be added to the initial query and to use generic structures to express structural constraints. Both sources of evidence are used in different combined forms. Experiments were carried out within the INEX evaluation campaign and results show the effectiveness of our approaches.     

Spatiotemporal evolution of the El Biar landslide (Algiers): new field observation data constrained by ground-penetrating radar investigations

Bulletin of Engineering Geology and the Environment - A better understanding of the spatiotemporal evolution of landslides in urban zones is a key factor in assessing the risk of future slides...     

Road crash fatality rates in France: A comparison of road user types,taking account of travel practices

Background

Travel practices are changing: bicycle and motorized two-wheeler (MTW) use are rising in some of France’s large cities. These are cheaper modes of transport and therefore attractive at a time of economic crisis, but they also allow their users to avoid traffic congestion. At the same time, active transport modes such as walking and cycling are encouraged because they are beneficial to health and reduce pollution. It is therefore important to find out more about the road crash risks of the different modes of transport. To do this, we need to take account of the number of individuals who use each, and, even better, their travel levels.

Method

We estimated the exposure-based fatality rates for road traffic crashes in France, on the basis of the ratio between the number of fatalities and exposure to road accident risk. Fatality data were obtained from the French national police database of road traffic casualties in the period 2007–2008. Exposure data was estimated from the latest national household travel survey (ENTD) which was conducted from April 2007 to April 2008. Three quantities of travel were computed for each mode of transport: (1) the number of trips, (2) the distance traveled and (3) the time spent traveling. Annual fatality rates were assessed by road user type, age and sex.

Results

The overall annual fatality rates were 6.3 per 100 million trips, 5.8 per billion kilometers traveled and 0.20 per million hours spent traveling. The fatality rates differed according to road user type, age and sex. The risk of being killed was 20 to 32 times higher for motorized two-wheeler users than for car occupants. For cyclists, the risk of being killed, both on the basis of time spent traveling and the number of trips was about 1.5 times higher than for car occupants. Risk for pedestrians compared to car occupants was similar according to time spent traveling, lower according to the number of trips and higher according to the distance traveled. People from the 17–20 and 21–29 age groups and those aged 70 and over had the highest rates. Males had higher rates than females, by a factor of between 2 and 3.

Conclusion

When exposure is taken into account, the risks for motorized two-wheeler users are extremely high compared to other types of road user. This disparity can be explained by the combination of speed and a lack of protection (except for helmets). The differential is so great that prevention measures could probably not eliminate it. The question that arises is as follows: with regard to public health, should not the use of MTW, or at least of motorcycles, be deterred? The difference between the fatality risk of cyclists and of car occupants is much smaller (1.5 times higher); besides, there is much room for improvements in cyclist safety, for instance by increasing the use of helmets and conspicuity equipment. Traffic calming could also benefit cyclists, pedestrians and perhaps moped users.     

Hydrogen production by hybrid system and its conversion by fuel cell in Algeria; Djanet

In this work we present a scenario of wind and solar energy production and seasonal energy storage producing Hydrogen in Djanet (East-South of Algeria). In addition we suppose assume the use of a set of fuel cells which are connected to the grid to provide a supply of energy when needed afterwards. The aim of this primary study is giving an alternative solution for the electric production in Djanet, which is mainly based on diesel generator. For that we made an investigation to highlight the potential of renewable energy production in this region. To ascertain feasibility of one hybrid system, we made energetic assessment considering the real climatic conditions of Djanet.     

A comparative evaluation of filter-based feature selection methods for hyper-spectral band selection

Band selection (dimensionality reduction) plays an essential role in hyper-spectral image processing and applications. This article presents a unified comparison framework for systematic performance comparison of filter-based feature selection models and conducts a comparative evaluation of four methods: maximal minimal associated index (MMAIQ), mutual information-based max-dependency criterion (mRMR), relief feature selection (Relief-F), and correlation-based feature selection (CFS) for hyper-spectral band selection. The evaluation is based on the performance of effectiveness, robustness, and classification accuracy, which involves five measuring indices: class separability, feature entropy, feature stability, feature redundancy, and classification accuracy. Three images acquired by different sensors were used to investigate the performance of the metrics. Experimental results show the best results for MMAIQ for all data sets in terms of used measurements, except for feature stability where mRMR and Relief-F exhibit their superiority.     

Study of microstructure, phases and microhardness of metallic coatings deposited by flame thermal spray

The recharging technique by thermal spraying offers the opportunity of renovating the worn surface parts of a machine element to give it again a new technical life despite its previous degradation in service. This process has consequently interesting economic impacts. In order to improve the adherence between 100Cr6 steel deposits and the substrate material (left worn crankshafts), company SNC ATRA of Béjaïa uses at present a composite formed by (100Cr6 steel/molybdenum bond coat of 0.2 mm thick/crankshaft substrate). As a matter of fact, it is shown in the present work that the molybdenum bond coat is not appropriate since, for the 0.2 mm thickness, lateral cracks are observed in the middle of the bond coat. On the other hand, our experiment is that a deposit of 100Cr6 steel projected directly on the substrate seems more promising since no gaps or cracks were detected at the “deposit/substrate” interface of this two-material composite. Lastly, phase analysis using X-ray diffraction confirmed that during spraying process, a stable -phase (bcc) of 100Cr6 wire was transformed to a new phase of γ-phase (fcc). The coatings exhibited the higher microhardness which would contribute to increase wear resistance.