Pasar al contenido principal

DEEP LEARNING IN AIR POLLUTION FORECASTING

RTI2018-098160-B-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos I+D+i Retos de Investigación
Año convocatoria 2018
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD DE CADIZ

Resultados relacionados

Resultados totales (Incluyendo duplicados): 15
Encontrada(s) 1 página(s)

RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Contribución de congreso (ConferenceOutput)

Air Pollution forecasting using Long Short-Term Memory Networks in the Bay of Algeciras (Spain)

  • Rodríguez García, María Inmaculada
  • Carrasco García, María Gema
  • González Enrique, Francisco Javier
  • Ruiz Águilar, Juan Jesús
  • Turias Domínguez, Ignacio José
Continuing with similar studies developed in the Bay of Algeciras, a new method from deep learning has been used. Long Short-Term Memory (LSTMs) are applied to predict future concentrations of SO2, NO2and PM10in the port of Algeciras (Spain). Maritime data from the port of Algeciras are kindly provided by the Algeciras Bay Port Authority and meteorological and pollution data are hourly collected in twenty-one monitoring stations at different points of the Bay of Algeciras from 1st January 2017 to 31stDecember 2019 provided by Andalusian Regional Government. The structure of an LSTM consists of a deep net with feedback connections that can process entire sequences of data which makes it better to forecast relevant pieces of data in the sequence and preserve it for several instants of time due to it can therefore have both short-term (like basic Recurrent Networks) and long-term memory. In this work, LSTM models have been used to predict future values of SO2, NO2, and PM10. A cross-validation strategy was adopted to obtain generalization results for unseen patterns. Furthermore, a Bayesian regularization procedure for hyperparameter searching was applied to avoid overfitting. The results have been very promising, obtaining R values above 0.75 for all pollutants in the 2h-ahead predictions as well as above 0.6 in the 8h-ahead forecasts. These results have been more accurately obtained than using shallow artificial neural networks as authors used in previous studies.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

A Clustering-Based Hybrid Support Vector Regression Model to Predict Container Volume at Seaport Sanitary Facilities

  • Ruiz Águilar, Juan Jesús
  • Moscoso López, José Antonio
  • Urda, Daniel
  • González Enrique, Francisco Javier
  • Turias Domínguez, Ignacio José
An accurate prediction of freight volume at the sanitary facilities of seaports is a key factor to improve planning operations and resource allocation. This study proposes a hybrid approach to forecast container volume at the sanitary facilities of a seaport. The methodology consists of a three-step procedure, combining the strengths of linear and non-linear models and the capability of a clustering technique. First, a self-organizing map (SOM) is used to decompose the time series into smaller clusters easier to predict. Second, a seasonal autoregressive integrated moving averages (SARIMA) model is applied in each cluster in order to obtain predicted values and residuals of each cluster. These values are finally used as inputs of a support vector regression (SVR) model together with the historical data of the cluster. The final prediction result integrates the prediction results of each cluster. The experimental results showed that the proposed model provided accurate prediction results and outperforms the rest of the models tested. The proposed model can be used as an automatic decision-making tool by seaport management due to its capacity to plan resources in advance, avoiding congestion and time delays.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Air pollution relevance analysis in the bay of Algeciras (Spain)

  • Rodríguez García, María Inmaculada
  • González Enrique, Francisco Javier
  • Moscoso López, José Antonio
  • Ruiz Águilar, Juan Jesús
  • Turias Domínguez, Ignacio José
The aim of this work is to accomplish an in-depth analysis of the air pollution in the two main cities of the Bay of Algeciras (Spain). A large database of air pollutant concentrations and weather measurements were collected using a monitoring network installed throughout the region from the period of 2010-2015. The concentration parameters contain nitrogen dioxide (NO2), sulphur dioxide (SO2) and particulate matter (PM10). The analysis was developed in two monitoring stations (Algeciras and La Linea). The higher average concentration values were obtained in Algeciras for NO2 (28.850 mu g/m(3)) and SO2 (11.966 mu g/m(3)), and in La Linea for PM10 (30.745 mu g/m(3)). The analysis shows patterns that coincide with human activity. One of the goals of this work is to develop a useful virtual sensor capable of achieving a more robust monitoring network, which can be used, for instance, in the case of missing data. By means of trends analysis, groups of equivalent stations were determined, implying that the values of one station could be substituted for those in the equivalent station in case of failure (e.g., SO2 weekly trends in Algeciras and Los Barrios show equivalence). On the other hand, a calculation of relative risks was developed showing that relative humidity, wind speed and wind direction produce an increase in the risk of higher pollutant concentrations. Besides, obtained results showed that wind speed and wind direction are the most important variables in the distribution of particles. The results obtained may allow administrations or citizens to support decisions.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Forecasting air pollutants using classification models: a case study in the Bay of Algeciras (Spain)

  • Rodríguez García, María Inmaculada
  • Ribeiro Rodrigues, M.C.
  • González Enrique, Francisco Javier
  • Ruiz Águilar, Juan Jesús
  • Turias Domínguez, Ignacio José
The main goal of this work is to obtain reliable predictions of pollutant concentrations related to maritime traffic (SO2, PM10, NO2, NOX, and NO) in the Bay of Algeciras, located in Andalusia, the south of Spain. Furthermore, the objective is to predict future air quality levels of the principal maritime traffic-related pollutants in the Bay of Algeciras as a function of the rest of the pollutants, the meteorological variables, and vessel data. In this sense, three scenarios were analysed for comparison, namely Alcornocales Park and the cities of La Línea and Algeciras. A database of hourly records of air pollution immissions, meteorological measurements in the Bay of Algeciras region and a database of maritime traffic in the port of Algeciras during the years 2017 to 2019 were used. A resampling procedure using a five-fold cross-validation procedure to assure the generalisation capabilities of the tested models was designed to compute the pollutant predictions with different classification models and also with artificial neural networks using different numbers of hidden layers and units. This procedure enabled appropriate and reliable multiple comparisons among the tested models and facilitated the selection of a set of top-performing prediction models. The models have been compared using several quality classification indexes such as sensitivity, specificity, accuracy, and precision. The distance (d1) to the perfect classifier (1, 1, 1, 1) was also used as a discriminant feature, which allowed for the selection of the best models. Concerning the number of variables, an analysis was conducted to identify the most relevant ones for each pollutant. This approach aimed to obtain models with fewer inputs, facilitating the design of an optimised monitoring network. These more compact models have proven to be the optimal choice in many cases. The obtained sensitivities in the best models were 0.98 for SO2, 0.97 for PM10, 0.82 for NO2 and NOX, and 0.83 for NO. These results demonstrate the potential of the models to forecast air pollution in a port city or a complex scenario and to be used by citizens and authorities to prevent exposure to pollutants and to make decisions concerning air quality, 25 páginas
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Long Short-Term Memory Approach for Short-Term Air Quality Forecasting in the Bay of Algeciras (Spain)

  • Rodríguez García, María Inmaculada
  • Carrasco García, María Gema
  • González Enrique, Francisco Javier
  • Ruíz Aguilar, Juan Jesús
  • Turias Domínguez, Ignacio José
Predicting air quality is a very important task, as it is known to have a significant impact on health. The Bay of Algeciras (Spain) is a highly industrialised area with one of the largest superports in Europe. During the period 2017–2019, different data were recorded in the monitoring stations of the bay, forming a database of 131 variables (air pollutants, meteorological information, and vessel data), which were predicted in the Algeciras station using long short-term memory models. Four different approaches have been developed to make SO2 and NO2 forecasts 1 h and 4 h in Algeciras. The first uses the remaining 130 exogenous variables. The second uses only the time series data without exogenous variables. The third approach consists of using an autoregressive time series arrangement as input, and the fourth one is similar, using the time series together with wind and ship data. The results showed that SO2 is better predicted with autoregressive information and NO2 is better predicted with ships and wind autoregressive time series, indicating that NO2 is closely related to combustion engines and can be better predicted. The interest of this study is based on the fact that it can serve as a resource for making informed decisions for authorities, companies, and citizens alike.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Improving the Reliability of Network Intrusion Detection Systems Through Dataset Integration

  • Magán Carrión, Roberto
  • Urda, Daniel
  • Díaz Cano, Ignacio
  • Dorronsoro Díaz, Bernabé
This work presents Reliable-NIDS (R-NIDS), a novel methodology for Machine Learning (ML) based Network Intrusion Detection Systems (NIDSs) that allows ML models to work on integrated datasets, empowering the learning process with diverse information from different datasets. We also propose a new dataset, called UNK22. It is built from three of the most well-known network datasets (UGR'16, USNW-NB15 and NLS-KDD), each one gathered from its own network environment, with different features and classes, by using a data aggregation approach present in R-NIDS. Therefore, R-NIDS targets the design of more robust models that generalize better than traditional approaches. Following R-NIDS, in this work we propose to build two well-known ML models for reliable predictions thanks to the meaningful information integrated in UNK22. The results show how these models benefit from the proposed approach, being able to generalize better when using UNK22 in the training process, in comparison to individually using the datasets composing it. Furthermore, these results are carefully analyzed with statistical tools that provide high confidence on our conclusions. Finally, the proposed solution is feasible to be deployed in network production environments, not usually taken into account in the literature., 16 páginas
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Artificial Neural Networks, Sequence-to-Sequence LSTMs, and Exogenous Variables as Analytical Tools for NO2 (Air Pollution) Forecasting: A Case Study in the Bay of Algeciras (Spain)

  • González Enrique, Francisco Javier
  • Ruiz Águilar, Juan Jesús
  • Moscoso López, José Antonio
  • Urda, Daniel
  • Deka, Lipika
  • Turias Domínguez, Ignacio José
This study aims to produce accurate predictions of the NO2 concentrations at a specific station of a monitoring network located in the Bay of Algeciras (Spain). Artificial neural networks (ANNs) and sequence-to-sequence long short-term memory networks (LSTMs) were used to create the forecasting models. Additionally, a new prediction method was proposed combining LSTMs using a rolling window scheme with a cross-validation procedure for time series (LSTM-CVT). Two different strategies were followed regarding the input variables: using NO2 from the station or employing NO2 and other pollutants data from any station of the network plus meteorological variables. The ANN and LSTM-CVT exogenous models used lagged datasets of different window sizes. Several feature ranking methods were used to select the top lagged variables and include them in the final exogenous datasets. Prediction horizons of t + 1, t + 4 and t + 8 were employed. The exogenous variables inclusion enhanced the model's performance, especially for t + 4 (rho approximate to 0.68 to rho approximate to 0.74) and t + 8 (rho approximate to 0.59 to rho approximate to 0.66). The proposed LSTM-CVT method delivered promising results as the best performing models per prediction horizon employed this new methodology. Additionally, per each parameter combination, it obtained lower error values than ANNs in 85% of the cases.
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Contribución de congreso (ConferenceOutput)

Prediction of container filling for the selective waste collection in Algeciras (Spain)

  • Rodríguez López, Juana Carmen
  • Rodríguez García, María Inmaculada
  • Moscoso López, José Antonio
  • Ruiz Águilar, Juan Jesús
  • Alcántara Pérez, José Manuel
  • Turias Domínguez, Ignacio José
14th Conference on Transport Engineering: 6th – 8th July 2021, The aim of this study is to create an intelligent system that improves the efficiency of garbage collection, (cardboard waste, in this particular case). The number of cardboard containers to be collected each day will be determined based on a prediction made on the filled volume recorded in each container. It will be reflected in the cost and fuel savings, reducing emissions and contributing to environmental sustainability. These results will allow planning the sequence of waste removal, which means the optimal collection route considering restrictive parameters such as the type of truck, the location of containers, collection times by zones, and the availability of working staff. A filling prediction system is proposed based on real historical data provided by the current waste collection company in Algeciras (ARCGISA). To achieve this objective, an intelligent system is designed using predictive analytics and several methods based on machine learning, modelling the collection system as a classification model, comparing the results from a statistical point of view (using sensitivity, specificity, etc.). The results obtained with the best-Tested method indicate an improvement average rate of 26% in sensitivity performance index and 67% in specificity performance index. Currently, waste collection is carried out without predictive analysis. The relevance of an efficient waste collection system is becoming increasingly important. Achieving optimal waste collection will result in improved service to citizens, cost savings for the administration, and significant environmental improvements. © 2021 Elsevier B.V.. All rights reserved
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Tesis doctoral (DoctoralThesis)

Determinación del efecto del tráfico marítimo en la estimación de la calidad del aire de una ciudad-puerto, Analysis of the effect of maritime traffic on the estimation of air quality in a port-city

  • Rodríguez García, María Inmaculada
Predecir la calidad del aire es una tarea muy importante ya que se sabe que tiene un impacto significativo en la salud. La Bahía de Algeciras (España) es una zona altamente industrializada con uno de los mayores puertos de Europa. Durante el periodo 2010-2019, se registraron diferentes datos en las estaciones de monitorización de la Bahía, formando una base de datos de 131 variables de concentraciones de contaminantes atmosféricos, de información meteorológica y datos de tonelada-buque de barcos en la Bahía de Algeciras. En esta tesis se desarrolló un análisis en tres fases. La primera fase consistió en un diagnóstico de la calidad del aire, la segunda en una estimación-predicción de la calidad del aire y la tercera en una predicción de la calidad del aire. En la primera fase, el primer paso fue desarrollar un análisis en profundidad realizado durante los años 2010 a 2015 del alcance de la contaminación atmosférica en las dos principales ciudades y más pobladas de la Bahía de Algeciras (Algeciras y La Línea). Es una zona donde coexisten varios grandes focos de contaminación atmosférica como varias industrias químicas y carreteras. Además, el puerto de Algeciras es uno de los más importantes de Europa, y la ciudad de La Línea está afectada por el aeropuerto de Gibraltar, lo que contribuye al aumento de la contaminación en el Estrecho de Gibraltar. Se ha desarrollado un completo análisis estadístico para conocer aspectos relevantes de la contaminación atmosférica en este escenario concreto. Por un lado, los valores medios de concentración más elevados se obtuvieron en Algeciras para NO2 (28,850 μg/m3 ) y en La Línea para SO2 (11,966 μg/m3 ) y PM10 (30,745 μg/m3 ). El análisis mostró patrones que coinciden con la actividad humana. Por otra parte, se desarrolló un cálculo de riesgos relativos que mostró que la humedad relativa, la velocidad del viento y la dirección del viento producen un aumento del riesgo de mayores concentraciones de contaminantes. Además, los resultados obtenidos mostraron que la velocidad y la dirección del viento son las variables más importantes en la distribución de partículas. En la segunda fase, el objetivo era obtener predicciones fiables de las concentraciones de contaminantes relacionados con el tráfico marítimo (SO2, PM10, NO2, NOX y, NO) durante los años 2017 a 2019. Se analizaron tres escenarios diferentes de la Bahía de Algeciras que fueron las localizaciones del Parque de los Alcornocales, así como las ciudades de La Línea y Algeciras. Estos escenarios permitieron comparar los resultados entre ellos. El objetivo fue predecir los niveles futuros de calidad del aire de los principales contaminantes relacionados con el tráfico marítimo en la Bahía de Algeciras en función del resto de contaminantes, las xvii variables meteorológicas y, una base de datos de buques. Se diseñó un procedimiento de remuestreo aleatorio utilizando un procedimiento de 5-CV (5-validación cruzada) y 20 repeticiones en cada experimento para asegurar la capacidad de generalización de los modelos probados, para calcular las predicciones de contaminantes con diferentes modelos de clasificación (incluyendo árboles de decisión, máquinas de soporte vectorial, ensembles, entre otros) y también con redes neuronales artificiales (RNAs) utilizando diferentes números de capas y unidades ocultas. El procedimiento propuesto de remuestreo aleatorio permite realizar comparaciones múltiples adecuadas y robustas de los modelos y permitió seleccionar un grupo de los mejores modelos de predicción. Los modelos se compararon utilizando varios índices de calidad de la clasificación, como la sensibilidad, la especificidad, la exactitud y la precisión. También se utilizó la distancia (d1) al clasificador perfecto (1, 1, 1, 1) como característica discriminante, para seleccionar los mejores modelos. Además, se realizó un análisis de relevancia para conocer qué variables son las más importantes para cada contaminante y se diseñaron modelos con menor número de entradas en una red de monitorización más óptima. Las sensibilidades obtenidas en los mejores modelos fueron de 0,98 para SO2, 0,97 para PM10, 0,82 para NO2 y NOX, y 0,83 para NO. Los resultados obtenidos demuestran el potencial de los modelos para predecir la contaminación en una ciudad portuaria o en un escenario de contaminación atmosférica complejo. En la tercera fase, se predijeron los datos disponibles durante los años 2017 a 2019 en la estación de Algeciras utilizando modelos de Long-Short Term Memory (LSTM). Se desarrollaron cuatro enfoques diferentes para realizar previsiones de SO2 y NO2 a 1 hora y a 4 horas en Algeciras. El primero utiliza las 130 variables exógenas restantes. El segundo utiliza únicamente los datos de series temporales sin variables exógenas. El tercer enfoque consiste en utilizar un arreglo de series temporales autorregresivas como entrada y el cuarto es similar utilizando las series temporales junto con datos de viento y barcos. Los resultados mostraron que el SO2 se predice mejor con información autorregresiva y el NO2 se predice mejor con series temporales autorregresivas de barcos y viento, lo que indica que el NO2 está estrechamente relacionado con los motores de combustión y puede predecirse mejor. Uno de los objetivos de esta tesis doctoral es desarrollar un sensor virtual útil capaz de conseguir una red de vigilancia más robusta, que pueda utilizarse en caso de datos faltantes. Además, puede servir como sistema de apoyo a la toma de decisiones por parte de las autoridades, que podría ser utilizado por los ciudadanos para prevenir la exposición a contaminantes y por las empresas para tomar decisiones sobre la calidad del aire., Predicting air quality is a very important task as it is known to have a significant impact on health. The Bay of Algeciras (Spain) is a highly industrialised area with one of the largest super-ports in Europe. During the period 2010-2019, different data were recorded in the monitoring stations of the Bay, forming a database of 131 variables (air pollutants, meteorological information, and vessel data). A three stage analysis was developed in this Thesis. The first stage was an air quality diagnosis, the second stage was an estimation-prediction of the air quality, and the third stage was an air quality forecasting. In the first stage, the first step was to develop an in-depth analysis accomplished during the years 2010 to 2015 of the scope of pollution in the two main cities of the Bay of Algeciras (Algeciras and La Línea). Surrounded by several industries and roads, this area is highly polluted. Besides, Algeciras port is one of the most important ports in Europe, and La Línea city is affected by Gibraltar airport, which contributes to the increase of air pollution in The Strait of Gibraltar. A complete statistical analysis was developed in order to gain knowledge about relevant aspects of air pollution in this particular scenario. On the one hand, the higher average concentration values were obtained in Algeciras for NO2 (28.850 μg/m3 ) and in La Línea for SO2 (11.966 μg/m3 ) and PM10 (30.745 μg/m3 ). The analysis showed patterns that coincide with human activity. On the other hand, a calculation of relative risks was developed showing that relative humidity, wind speed and wind direction produce an increase in the risk of higher pollutant concentrations. Besides, obtained results showed that wind speed and wind direction are the most important variables in the distribution of particles. In the second stage, the objective was to obtain reliable predictions of pollutant concentrations related to maritime traffic (SO2, PM10, NO2, NOX and, NO) during the years 2017 to 2019. Three scenarios were analysed which involve the locations of Alcornocales Park, as well as the cities of La Línea and Algeciras. These scenarios allowed us to compare the results. The objective was to predict future air quality levels of the principal maritime traffic-related pollutants in the Bay of Algeciras as a function of the rest of the pollutants, the meteorological variables and, a vessel database. A randomised resampling procedure was designed using a 5-fold cross-validation procedure and 20 repetitions in each experiment to ensure the generalisation capabilities of the tested models to compute the pollutant xix predictions with different classification models (including Trees, Ensembles, SVM, among others) and also with artificial neural networks (ANNs) using different numbers of hidden layers and units. The proposed procedure of random resampling permits adequate and robust multiple comparisons of the tested models and allowed us to select a group of best prediction models. The models were compared using several quality classification indexes such as sensitivity, specificity, accuracy, and precision. The distance (d1) to the perfect classifier (1, 1, 1, 1) was also used as a discriminant feature, which allowed us to select the best models. Furthermore, a relevance analysis was performed in order to know which variables are the most relevant to each pollutant and to design models with less number of inputs in a more optimal monitoring network. These models seem to be the best in a number of scenarios. The obtained sensibilities in the best models were 0.98 for SO2, 0.97 for PM10, 0.82 for NO2 and NOX, and 0.83 for NO. The obtained results demonstrate the potential of the models to forecast air pollution in a port city or a complex scenario. The third stage, data available during the years 2017 to 2019 were predicted in the Algeciras station using Long-Short Term Memory (LSTM) models. Four different approaches have been developed to make SO2 and NO2 forecasts 1h and 4h in Algeciras. The first uses the remaining 130 exogenous variables. The second uses only the time series data without exogenous variables. The third approach consists of using an autoregressive time series arrangement as input and the fourth one is similar using the time series together with wind and ship data. The results showed that SO2 is better predicted with autoregressive information and NO2 is better predicted with ships and wind autoregressive time series, indicating that NO2 is closely related to combustion engines and can be better predicted. One of the goals of this Thesis is to develop a useful virtual sensor capable of achieving a more robust monitoring network, which can be used, for instance, in the case of missing data. Besides, it can serve as a decision support system for authorities, that could be used by citizens to prevent exposure to pollutants and companies to make air quality decisions., 233 páginas
RODIN. Repositorio de Objetos de Docencia e Investigación de la Universidad de Cádiz
Publicación Artículo científico (JournalArticle)

Evaluating the impact of different Feature as a Counter data aggregation approaches on the performance of NIDSs and their selected features

  • Magán Carrión, Roberto
  • Urdal Muñoz, Daniel
  • Díaz Cano, Ignacio
  • Dorronsoro Díaz, Bernabé
There is much effort nowadays to protect communication networks against different cybersecurity attacks (which are more and more sophisticated) that look for systems’ vulnerabilities they could exploit for malicious purposes. Network Intrusion Detection Systems (NIDSs) are popular tools to detect and classify such attacks, most of them based on ML models. However, ML-based NIDSs cannot be trained by feeding them with network traffic data as it is. Thus, a Feature Engineering (FE) process plays a crucial role transforming network traffic raw data onto derived one suitable for ML models. In this work, we study the effects of applying one such FE technique in different ways on the performance of two ML models (linear and non-linear) and their selected features. This is the Feature as a Counter approach. The derived observations are computed from either with the same number of raw samples, (batch-based approaches) or by aggregating them by time intervals (timestamp-based approach). Results show that there is no significant differences between the proposed approaches neither in the performance of the models nor in the selected features that validate our proposal making it feasible to be widely used as a standard FE method.
Repositorio Institucional de la Universidad de Burgos (RIUBU)
Publicación Contribución de congreso (ConferenceOutput)

Prediction of container filling for the selective waste collection in Algeciras (Spain)

  • Rodríguez López, Juana Carmen
  • Moscoso López, José Antonio
  • Ruiz Aguilar, Juan Jesús
  • Rodríguez García, Inmaculada
  • Alcántara Pérez, Jose Manuel
  • Turias Domínguez, Ignacio J.
Trabajo presentado en: R-Evolucionando el transporte, XIV Congreso de Ingeniería del Transporte (CIT 2021), realizado en modalidad online los días 6, 7 y 8 de julio de 2021, organizado por la Universidad de Burgos, The aim of this study is to create an intelligent system that improves the efficiency of garbage collection, (cardboard waste, in this particular case). The number of cardboard containers to be collected each day will be determined based on a prediction made on the filled volume recorded in each container. It will be reflected in the cost and fuel savings, reducing emissions and contributing to environmental sustainability. These results will allow planning the sequence of waste removal, which means the optimal collection route considering restrictive parameters such as the type of truck, the location of containers, collection times by zones, and the availability of working staff. A filling prediction system is proposed based on real historical data provided by the current waste collection company in Algeciras (ARCGISA). To achieve this objective, an intelligent system is designed using predictive analytics and several methods based on machine learning, modelling the collection system as a classification model, comparing the results from a statistical point of view (using sensitivity, specificity, etc.). The results obtained with the best-tested method indicate an improvement average rate of 26% in sensitivity performance index and 67% in specificity performance index. Currently, waste collection is carried out without predictive analysis. The relevance of an efficient waste collection system is becoming increasingly important. Achieving optimal waste collection will result in improved service to citizens, cost savings for the administration, and significant environmental improvements., This work is part of the research project RTI-2018-098160-B-I00 supported by 'MICINN. Programa Estatal de I+D+i Orientada a 'Los Retos de la Sociedad'. Data used in this work have been kindly provided by ARCGISA. Colaboration between ARCGISA and University of Cádiz was supported with Fundación del Campus Tecnológico de Algeciras (FCTA).
Repositorio Institucional de la Universidad de Burgos (RIUBU)
Publicación Contribución de congreso (ConferenceOutput)

Comparison of maritime transport influence of SO2 levels in Algeciras and Alcornocales Park (Spain)

  • Rodríguez García, Inmaculada
  • Moscoso López, José Antonio
  • Ruiz Aguilar, Juan Jesús
  • González Enrique, Francisco Javier
  • Rodríguez López, Juana Carmen
  • Turias Domínguez, Ignacio J.
Trabajo presentado en: R-Evolucionando el transporte, XIV Congreso de Ingeniería del Transporte (CIT 2021), realizado en modalidad online los días 6, 7 y 8 de julio de 2021, organizado por la Universidad de Burgos, The main aim of this work was to measure the influence of the volume of shipping over the Sulphur dioxide (SO2) concentration in the air pollution in two monitoring stations located at Algeciras city and Alcornocales Park developing the same analysis in these two locations. The target is to demonstrate the assumption that Algeciras is more affected by SO2 than Alcornocales Park which is 30 km far away from Algeciras Port. A multiple regression approach has been applied using wind data: wind direction (degrees) and wind speed (km/h) recorded in two weather stations, together with the volume of the gross tonnage per hour (GT/h) of vessels in the Bay of Algeciras to estimate SO2 concentration values in the two stations Algeciras and Alcornocales. The database contains records of hourly samples of these variables during the year 2019. Different artificial neural networks (ANNs) models were compared and the results showed that SO2 in Algeciras station could be better explained than the same pollutant in Alcornocales station. On the other hand, ANNs produced better results than linear models which means that nonlinear models fit best the data. A cross- validation procedure has been applied in order to assure the generalization capabilities of the tested models. The results showed that in Algeciras a more reliable estimation could be done reaching a correlation estimation between the model and the target (real) values of SO2. This fact highlights the major influence of maritime transport in the Bay of Algeciras, This work is part of the research project RTI2018-098160-B-I00 supported by 'MICINN’ Programa Estatal de I+D+i Orientada a 'Los Retos de la Sociedad'. Data used in this work have been kindly provided by the Algeciras Port Authority and the Andalusian Regional Government.
Repositorio Institucional de la Universidad de Burgos (RIUBU)
Publicación Artículo científico (JournalArticle)

Artificial Neural Networks, Sequence-to-Sequence LSTMs, and Exogenous Variables as Analytical Tools for NO2 (Air Pollution) Forecasting: A Case Study in the Bay of Algeciras (Spain)

  • González Enrique, Francisco Javier
  • Ruiz Aguilar, Juan Jesús
  • Moscoso López, José Antonio
  • Urda Muñoz, Daniel
  • Deka, Lipika
  • Turias Domínguez, Ignacio J.
This study aims to produce accurate predictions of the NO2 concentrations at a specific station of a monitoring network located in the Bay of Algeciras (Spain). Artificial neural networks (ANNs) and sequence-to-sequence long short-term memory networks (LSTMs) were used to create the forecasting models. Additionally, a new prediction method was proposed combining LSTMs using a rolling window scheme with a cross-validation procedure for time series (LSTM-CVT). Two different strategies were followed regarding the input variables: using NO2 from the station or employing NO2 and other pollutants data from any station of the network plus meteorological variables. The ANN and LSTM-CVT exogenous models used lagged datasets of different window sizes. Several feature ranking methods were used to select the top lagged variables and include them in the final exogenous datasets. Prediction horizons of t + 1, t + 4 and t + 8 were employed. The exogenous variables inclusion enhanced the model’s performance, especially for t + 4 (ρ ≈ 0.68 to ρ ≈ 0.74) and t + 8 (ρ ≈ 0.59 to ρ ≈ 0.66). The proposed LSTM-CVT method delivered promising results as the best performing models per prediction horizon employed this new methodology. Additionally, per each parameter combination, it obtained lower error values than ANNs in 85% of the cases., This research was funded by MICINN (Ministerio de Ciencia e Innovación-Spain), grant number RTI2018-098160-B-I00, and grant “Ayuda para Estancias en Centros de Investigación del Programa de Fomento e Impulso de la actividad Investigadora de la Universidad de Cádiz”.
Repositorio Institucional de la Universidad de Burgos (RIUBU)
Publicación Artículo científico (JournalArticle)

Towards a Reliable Comparison and Evaluation of Network Intrusion Detection Systems Based on Machine Learning Approaches

  • Magán-Carrion, Roberto
  • Urda Muñoz, Daniel
  • Díaz-Cano, Ignacio
  • Dorronsoro, Bernabé
Presently, we are living in a hyper-connected world where millions of heterogeneous devices are continuously sharing information in different application contexts for wellness, improving communications, digital businesses, etc. However, the bigger the number of devices and connections are, the higher the risk of security threats in this scenario. To counteract against malicious behaviours and preserve essential security services, Network Intrusion Detection Systems (NIDSs) are the most widely used defence line in communications networks. Nevertheless, there is no standard methodology to evaluate and fairly compare NIDSs. Most of the proposals elude mentioning crucial steps regarding NIDSs validation that make their comparison hard or even impossible. This work firstly includes a comprehensive study of recent NIDSs based on machine learning approaches, concluding that almost all of them do not accomplish with what authors of this paper consider mandatory steps for a reliable comparison and evaluation of NIDSs. Secondly, a structured methodology is proposed and assessed on the UGR’16 dataset to test its suitability for addressing network attack detection problems. The guideline and steps recommended will definitively help the research community to fairly assess NIDSs, although the definitive framework is not a trivial task and, therefore, some extra effort should still be made to improve its understandability and usability further., The authors would like to acknowledge the Spanish Ministerio de Ciencia, Innovación y Universidades and ERDF for the support provided under contracts RTI2018-100754-B-I00 (iSUN) and RTI2018-098160-B-I00 (DEEPAPFORE).
Repositorio Institucional de la Universidad de Burgos (RIUBU)
Publicación

A Clustering-Based Hybrid Support Vector Regression Model to Predict Container Volume at Seaport Sanitary Facilities

  • Ruiz Aguilar, Juan Jesús
  • Moscoso López, José Antonio
  • Urda Muñoz, Daniel
  • González Enrique, Francisco Javier
  • Turias Domínguez, Ignacio J.
An accurate prediction of freight volume at the sanitary facilities of seaports is a key factor to improve planning operations and resource allocation. This study proposes a hybrid approach to forecast container volume at the sanitary facilities of a seaport. The methodology consists of a three-step procedure, combining the strengths of linear and non-linear models and the capability of a clustering technique. First, a self-organizing map (SOM) is used to decompose the time series into smaller clusters easier to predict. Second, a seasonal autoregressive integrated moving averages (SARIMA) model is applied in each cluster in order to obtain predicted values and residuals of each cluster. These values are finally used as inputs of a support vector regression (SVR) model together with the historical data of the cluster. The final prediction result integrates the prediction results of each cluster. The experimental results showed that the proposed model provided accurate prediction results and outperforms the rest of the models tested. The proposed model can be used as an automatic decision-making tool by seaport management due to its capacity to plan resources in advance, avoiding congestion and time delays., This research was funded by MICINN (Ministerio de Ciencia e Innovación-Spain), grant number RTI2018-098160-B-I00.