Editorial: Special Issue on High Performace Distributed Computing

The Journal of Supercomputing -     

Calibration of a Photodiode Array Spectrometer Against the Copper Point

This paper describes a method to calibrate photodiode array spectrometers in the spectral radiance mode using a fixed-point blackbody as a reference source. Fixed-point blackbodies are characterized by their excellent emissivity, uniformity, and stability, which make them superior to both conventional standard lamps and variable temperature blackbodies. The temperature of these fixed points is accurately determined being traceable to either the International Temperature Scale (ITS-90) or thermodynamically through radiometric standards. The potential advantage of the fixed-point traceability chain is that it can be universally reproduced without recourse to any hierarchical calibrations or standards. The paper presents the calibration system and discusses the limitations of such an approach. The method used obtained an uncertainty of around 1.4 % ( \(k = 2\) ) associated with radiance responsivity across the spectral region from 550 nm to 1050 nm, which is comparable to what is readily achieved with a lamp-tile or lamp-illuminated spherical source.     

A robust neural network model for fault detection in the presence of mislabelled data

Several data-driven methodologies for process monitoring and detection of faults or abnormalities have been developed for the safety of processing systems. The effectiveness of data-based models, however, is impacted by the volume and quality of training data. This work presents a robust neural network model for addressing the mislabelled and low-quality data in detecting faults and process abnormalities. The approach is based on harnessing data quality features along with supervisory labels in the network training. The data quality has been computed using the Mahalanobis distances and trusted centres of each class of data such as normal and faulty data. The method has been examined for detecting abnormalities in two case studies; a continuous stirred tank heater problem for detecting leaks and the Tennessee Eastman chemical process for detecting step and sticking faults. The performance of the proposed robust artificial neural networks (ANN) model is evaluated in terms of accuracy, fault detection rate, false alarm rate, and classification index at varying extents of mislabelling, namely, 1%, 5%, and 10% mislabelled data. The proposed model demonstrates higher detection performance, especially at increased labels of mislabelled data where the performance of the conventional ANN is severely impacted. The proposed methodology can be advantageous in handling mislabelled and low-quality data issues which are crucial in the data-driven modelling of processing systems.     

Perspectives in pancreatic pain

This review describes some of the mechanisms which are thought to be important in the causation of pain in chronic pancreatitis. Both medical and surgical techniques for treating this pain are described.     

Fetal retinoblastoma

Fetal tumors are sporadic entities and in some cases can be diagnosed prenatally. A team approach, incorporating the skills of obstetricians, radiation therapists, pediatric medical oncologists and hematologists, neonatal surgeons, rehabilitation specialists, and social workers, is imperative to ensure that the patient receives the best treatment. Retinoblastoma is a relatively uncommon tumor of childhood that arises in the retina. The estimated annual incidence is between 1 in 15,000 and 1 in 34,000. Although it may occur at any age, it most often arises in younger children, with 80% of cases diagnosed before the age of 5 years. We report a case of prenatally diagnosed fetal retinoblastoma.     

Rapid monitoring of the spoilage of minced beef stored under conventionally and active packaging conditions using Fourier transform infrared spectroscopy in tandem with chemometrics

Fourier transform infrared (FTIR) spectroscopy was exploited to measure biochemical changes within fresh minced beef in an attempt to rapidly monitor beef spoilage. Minced beef packaged either aerobically, under modified atmosphere and using an active packaging were held from freshness to spoilage at 0, 5, 10, and 15 °C. Frequent FTIR measurements were collected directly from the sample surface using attenuated total reflectance, in parallel the total viable counts of bacteria, the sensory quality and the pH were also determined.     

Anodic polarization of tungsten in neutral and alkaline solutions under conditions of anode film growth

Anodic polarization of W has been studied in neutral (Na₂SO₄, NaCl) and alkaline solutions (ammonium borate, Na₂CO₃, Na₃PO₄, and NaOH) by the galvanostatic method under conditions which lead to anode film growth. The rate of oxide growth has been calculated as a function of current density, nature and concentration of electrolyte. Film growth to potentials above 100 V is achieved in alkaline solutions, provided the solution is dilute and the current density is high. The applicability of the high field ionic conduction has been verified from the dependence of both reciprocal capacity and potential on the logarithm of current density. Data are recorded for the coulometric increase of oxide thickness. The constants of the exponential law, and the electric field.     

Burnability of cement raw materials at rapid calcination conditions

Cement raw materials (clay, limestone and pyrite cinders) were mixed, pressed into pellets, burnt at high temperatures, cooled and the resulting clinkers analysed for free CaO content. The effect of the following parameters and their interactions on the free CaO was studied: top temperature of burning (in three levels: 1300, 1350, 1400°C), soaking time at top temperature (in three levels: 5, 30 and in some cases 60 min), method of heating (in two levels: rapid and normal), method of cooling (in two levels, rapid and slow), lime saturation factor of the mix, LSF (in three levels, 0.95, 0.85, 0.76). Experiments were planned to enable a statistical evaluation (calculation of error variance at various levels of significance). Rapid calcination improves the burnability of cement raw materials, as contrasted to normal calcination. This indicates that rapid calcination enables the use of lower calcination temperatures. A long soaking at top temperature is unnecessary in case of rapid calcination. The free CaO content of the clinker is affected primarily by the LSF of the mix and the top temperature of firing; all other parameters are less effective.     

The role of the experimental data base used to identify material parameters in predicting the cyclic plastic response of an austenitic steel

The first objective of this paper investigates the influence of the previous strain history on ratcheting. New tests were performed where different strain-controlled histories have been applied prior to ratcheting tests. It is demonstrated that under the same conditions, one can observe ratcheting, plastic shakedown or elasticity according to the prior strain-controlled history. The second objective points out the correlation between the experimental data base devoted to the identification of the material parameters and the quality of the predictions in cyclic plasticity. The results suggest that the choice of the tests should be closely linked to the capabilities of the model. In particular, the presence of non proportional strain-controlled tests in the data base may be not a good choice if the model itself is not able to represent explicitly such a character. All tests considered here were performed on 304L SS at room temperature.     

Electric load forecasting by using dynamic neural network

Electrical energy is fundamental for the wellbeing and for the economic development of any country. However, all countries must ensure access to essential resources and ensure the continuity of its supply. Due to the non-storable nature of electrical energy, the amount of consumed active power should always be equal the produced active power just to avoid power system frequency deviation problem. In order to keep the relationship production–consumption relation in compliance with different standards and to secure profitable operations of power system, electric load consumption must be predicted and controlled instantaneously. Several statistical and classical techniques are proposed in the literature but unfortunately all these methods are not accurate in a satisfactory manner. In this paper, a dynamic neural network is used for the prediction of daily power consumption. The suitability and the performance of the proposed approach is illustrated and verified with simulations on load data collected from French Transmission System Operator (RTE) website. The obtained results show that the accuracy and the efficiency are improved comparatively to conventional methods widely used in this field of research.