Evolution and forecasting of PM10 concentration at the Port of Gijon (Spain).

The name PM10 refers to small particles with a diameter of less than 10 microns. The present research analyses different models capable of predicting PM10 concentration using the previous values of PM10, SO2, NO, NO2, CO and O3 as input variables. The information for model training uses data from January 2010 to December 2017. The models trained were autoregressive integrated moving average (ARIMA), vector autoregressive moving average (VARMA), multilayer perceptron neural networks (MLP), support vector machines as regressor (SVMR) and multivariate adaptive regression splines. Predictions were performed from 1 to 6 months in advance. The performance of the different models was measured in terms of root mean squared errors (RMSE). For forecasting 1 month ahead, the best results were obtained with the help of a SVMR model of six variables that gave a RMSE of 4.2649, but MLP results were very close, with a RMSE value of 4.3402. In the case of forecasts 6 months in advance, the best results correspond to an MLP model of six variables with a RMSE of 6.0873 followed by a SVMR also with six variables that gave an RMSE result of 6.1010. For forecasts both 1 and 6 months ahead, ARIMA outperformed VARMA models.

Cyanotoxin level prediction in a reservoir using gradient boosted regression trees: a case study.

Cyanotoxins are a type of cyanobacteria that is poisonous and poses a health threat in waters that could be used for drinking or recreational purposes. Thus, it is necessary to predict their presence to avoid risks. This paper presents a nonparametric machine learning approach using a gradient boosted regression tree model (GBRT) for prediction of cyanotoxin contents from cyanobacterial concentrations determined experimentally in a reservoir located in the north of Spain. GBRT models seek and obtain good predictions in highly nonlinear problems, like the one treated here, where the studied variable presents low concentrations of cyanotoxins mixed with high concentration peaks. Two types of results have been obtained: firstly, the model allows the ranking or the dependent variables according to its importance in the model. Finally, the high performance and the simplicity of the model make the gradient boosted tree method attractive compared to conventional forecasting techniques.

Evaluation of ground calcite/water heavy media cyclone suspensions for production of residual plastic concentrates.

Viable recycled residual plastic (RP) product(s) must be of sufficient quality to be reusable as a plastic or source of hydrocarbons or fuel. The varied composition and large volumes of such wastes usually requires a low cost, high through-put recycling method(s) to eliminate contaminants. Cyclone separation of plastics by density is proposed as a potential method of achieving separations of specific types of plastics. Three ground calcite separation medias of different grain size distributions were tested in a cylindrical cyclone to evaluate density separations at 1.09, 1.18 and 1.27 g/cm3. The differences in separation recoveries obtained with these medias by density offsets produced due to displacement of separation media solid particles within the cyclone caused by centrifugal settling is evaluated. The separation density at which 50% of the material of that density is recovered was found to increase from 0.010 to 0.026 g/cm3 as the separation media density increased from 1.09 to 1.27 g/cm3. All separation medias were found to have significantly low Ep95values of 0.012-0.033 g/cm3. It is also demonstrated that the presence of an excess content of <10 µm calcite media particles (>75%) resulted in reduced separation efficiencies. It is shown that the optimum separations were achieved when the media density offset was 0.03-0.04 g/cm3. It is shown that effective heavy media cyclone separations of RP denser than 1.0 g/cm3 can produce three sets of mixed plastics containing: PS and ABS/SAN at densities of >1.0-1.09 g/cm3; PC, PMMA at a density of 1.09-1.18 g/cm3; and PVC and PET at a density of >1.27 g/cm3.

Comparative Study of Neural Network Frameworks for the Next Generation of Adaptive Optics Systems.

Many of the next generation of adaptive optics systems on large and extremely large telescopes require tomographic techniques in order to correct for atmospheric turbulence over a large field of view. Multi-object adaptive optics is one such technique. In this paper, different implementations of a tomographic reconstructor based on a machine learning architecture named "CARMEN" are presented. Basic concepts of adaptive optics are introduced first, with a short explanation of three different control systems used on real telescopes and the sensors utilised. The operation of the reconstructor, along with the three neural network frameworks used, and the developed CUDA code are detailed. Changes to the size of the reconstructor influence the training and execution time of the neural network. The native CUDA code turns out to be the best choice for all the systems, although some of the other frameworks offer good performance under certain circumstances.

Successful sulfur recovery in low sulfurate compounds obtained from the zinc industry: Evaporation-condensation method.

The improvement of an evaporation-condensation method allows for successful recovery of elemental sulfur from sulfide concentrates from the zinc industry. Elemental sulfur can be obtained with this method in samples with a low (60%) sulfur content. The effects of heating temperature between 150°C and 250°C and heating time up to 120min on the recovery of sulfur are also studied. Elemental sulfur obtained in this way is of high purity and therefore, there is no need for further purification. The treatment of these industrial residues would help removing sulfur from the environment.

Radon Mitigation Approach in a Laboratory Measurement Room.

Radon gas is the second leading cause of lung cancer, causing thousands of deaths annually. It can be a problem for people or animals in houses, workplaces, schools or any building. Therefore, its mitigation has become essential to avoid health problems and to prevent radon from interfering in radioactive measurements. This study describes the implementation of radon mitigation systems at a radioactivity laboratory in order to reduce interferences in the different works carried out. A large set of radon concentration samples is obtained from measurements at the laboratory. While several mitigation methods were taken into account, the final applied solution is explained in detail, obtaining thus very good results by reducing the radon concentration by 76%.

Hybrid Intelligent System to Perform Fault Detection on BIS Sensor During Surgeries.

This paper presents a new fault detection system in hypnotic sensors used for general anesthesia during surgery. Drug infusion during surgery is based on information received from patient monitoring devices; accordingly, faults in sensor devices can put patient safety at risk. Our research offers a solution to cope with these undesirable scenarios. We focus on the anesthesia process using intravenous propofol as the hypnotic drug and employing a Bispectral Index (BISTM) monitor to estimate the patient's unconsciousness level. The method developed identifies BIS episodes affected by disturbances during surgery with null clinical value. Thus, the clinician-or the automatic controller-will not take those measures into account to calculate the drug dose. Our method compares the measured BIS signal with expected behavior predicted by the propofol dose provider and the electromyogram (EMG) signal. For the prediction of the BIS signal, a model based on a hybrid intelligent system architecture has been created. The model uses clustering combined with regression techniques. To validate its accuracy, a dataset taken during surgeries with general anesthesia was used. The proposed fault detection method for BIS sensor measures has also been verified using data from real cases. The obtained results prove the method's effectiveness.

A Hybrid Algorithm for Missing Data Imputation and Its Application to Electrical Data Loggers.

The storage of data is a key process in the study of electrical power networks related to the search for harmonics and the finding of a lack of balance among phases. The presence of missing data of any of the main electrical variables (phase-to-neutral voltage, phase-to-phase voltage, current in each phase and power factor) affects any time series study in a negative way that has to be addressed. When this occurs, missing data imputation algorithms are required. These algorithms are able to substitute the data that are missing for estimated values. This research presents a new algorithm for the missing data imputation method based on Self-Organized Maps Neural Networks and Mahalanobis distances and compares it not only with a well-known technique called Multivariate Imputation by Chained Equations (MICE) but also with an algorithm previously proposed by the authors called Adaptive Assignation Algorithm (AAA). The results obtained demonstrate how the proposed method outperforms both algorithms.

A New Missing Data Imputation Algorithm Applied to Electrical Data Loggers.

Nowadays, data collection is a key process in the study of electrical power networks when searching for harmonics and a lack of balance among phases. In this context, the lack of data of any of the main electrical variables (phase-to-neutral voltage, phase-to-phase voltage, and current in each phase and power factor) adversely affects any time series study performed. When this occurs, a data imputation process must be accomplished in order to substitute the data that is missing for estimated values. This paper presents a novel missing data imputation method based on multivariate adaptive regression splines (MARS) and compares it with the well-known technique called multivariate imputation by chained equations (MICE). The results obtained demonstrate how the proposed method outperforms the MICE algorithm.

A hybrid PCA-CART-MARS-based prognostic approach of the remaining useful life for aircraft engines.

Prognostics is an engineering discipline that predicts the future health of a system. In this research work, a data-driven approach for prognostics is proposed. Indeed, the present paper describes a data-driven hybrid model for the successful prediction of the remaining useful life of aircraft engines. The approach combines the multivariate adaptive regression splines (MARS) technique with the principal component analysis (PCA), dendrograms and classification and regression trees (CARTs). Elements extracted from sensor signals are used to train this hybrid model, representing different levels of health for aircraft engines. In this way, this hybrid algorithm is used to predict the trends of these elements. Based on this fitting, one can determine the future health state of a system and estimate its remaining useful life (RUL) with accuracy. To evaluate the proposed approach, a test was carried out using aircraft engine signals collected from physical sensors (temperature, pressure, speed, fuel flow, etc.). Simulation results show that the PCA-CART-MARS-based approach can forecast faults long before they occur and can predict the RUL. The proposed hybrid model presents as its main advantage the fact that it does not require information about the previous operation states of the input variables of the engine. The performance of this model was compared with those obtained by other benchmark models (multivariate linear regression and artificial neural networks) also applied in recent years for the modeling of remaining useful life. Therefore, the PCA-CART-MARS-based approach is very promising in the field of prognostics of the RUL for aircraft engines.

Using artificial neural networks for open-loop tomography.

Modern adaptive optics (AO) systems for large telescopes require tomographic techniques to reconstruct the phase aberrations induced by the turbulent atmosphere along a line of sight to a target which is angularly separated from the guide sources that are used to sample the atmosphere. Multi-object adaptive optics (MOAO) is one such technique. Here, we present a method which uses an artificial neural network (ANN) to reconstruct the target phase given off-axis references sources. We compare our ANN method with a standard least squares type matrix multiplication method and to the learn and apply method developed for the CANARY MOAO instrument. The ANN is trained with a large range of possible turbulent layer positions and therefore does not require any input of the optical turbulence profile. It is therefore less susceptible to changing conditions than some existing methods. We also exploit the non-linear response of the ANN to make it more robust to noisy centroid measurements than other linear techniques.