Fabio Moretti - Academia.edu (original) (raw)
Papers by Fabio Moretti
Technical and economic analysis of a Smart Public Lighting model
EPJ Web of Conferences, 2012
Model Predictive Control for Building Active Demand Response Systems
Energy Procedia, 2015
The role of data sample size and dimensionality in neural network based forecasting of building heating related variables
Energy and Buildings, 2016
tEnergy consumed in buildings represents a challenge in the context of reduction of greenhouse ga... more tEnergy consumed in buildings represents a challenge in the context of reduction of greenhouse gasesemission. For this reason and due to the growing interest in operative costs reduction the energy used bybuildings (tertiary and privates) for heating, ventilating, and air conditioning (HVAC) is even more inves-tigated. Due to the nature of the energy consumption profile a predictive optimization method is one ofthe solution the scientific literature spreads even more. However optimization techniques need a goodand reliable prediction of the variables of interest over a time horizon. This work focuses on methods toobtain a robust and reliable predictor based on artificial neural networks. For the optimization purposesthe neural model predicts total heating energy consumption (gas), internal air temperature and aggre-gated thermal discomfort 12 h ahead. Training and testing data are simulated using a simulator basedon heat, air and moisture model for building and systems evaluation (HAMBASE), by which a real officebuilding was modeled. Influence of training data sample size and selection of predictor inputs is exam-ined. Several combinations of early stopping condition and network complexity are tested for differenttraining sample sizes. It is observed that the early stopping mechanism is crucial especially but not onlyfor small training data, because it reliably overcomes overfitting problems. Surprisingly, relatively smallnetworks were sufficient or performed best, although examined range of training sample covered up tofive heating seasons. The use of a model tuning is thus supported by the results. Further, two strategiesof selection of suitable input variables are demonstrated. While the input selection does not degrade theprediction performance, it is able to reduce the dimensionality and thus to save computational, commu-nication, time, and data acquisition demands. The importance of inputs selection in HVAC modeling isthus pointed out and demonstrated.
Energy Procedia, 2014
In the building energy efficiency field, developing automatic and accurate fault detection and di... more In the building energy efficiency field, developing automatic and accurate fault detection and diagnosis methods is necessary in order to ensure optimal operations of systems and to save energy. In this paper first, fault detection analysis based on statistical methods where anomalies are detected through a comparison with neighborhood and averaged fault-free values and through a clustering technique is performed. Following the fault detection step, a fault diagnosis analysis based on fuzzy sets and fuzzy logic is implemented. Experimentation is carried out over a one day monitoring data set in December 2013 for the fan coil electric consumption of an actual office building located at ENEA 'Casaccia' Research Centre. Results show the effectiveness of proposed approaches in automatic detection and diagnosis of abnormal building fan coil electric consumption.
Building thermal optimisation through surrogate assisted evolutionary multi-objective algorithms
Urban traffic flow forecasting through statistical and neural network bagging ensemble hybrid modeling
Neurocomputing, 2015
ABSTRACT In this paper we show a hybrid modeling approach which combines Artificial Neural Networ... more ABSTRACT In this paper we show a hybrid modeling approach which combines Artificial Neural Networks and a simple statistical approach in order to provide a one hour forecast of urban traffic flow rates. Experimentation has been carried out on three different classes of real streets and results show that the proposed approach outperforms the best of the methods it puts together.
Primary and secondary biomass burning aerosols determined by proton nuclear magnetic resonance (H-NMR) spectroscopy during the 2008 EUCAARI campaign in the Po Valley (Italy)
Atmospheric Chemistry and Physics Discussions, 2013
Technical and economic analysis of a Smart Public Lighting model
EPJ Web of Conferences, 2012
Illuminazione pubblica adattiva: modellistica dei sistemi intelligenti
Sviluppo di una sperimentazione dimostrativa di" Smart Village" e metodi di progettazione
Advanced Street Lighting Control through Neural Network Ensembling
Metodologia di ottimizzazione multi-obiettivo della climatizzazione termica di edifici. Validazione su sistema di simulazione
Lecture Notes in Computer Science, 2014
Advances in Intelligent Systems and Computing, 2014
The paper demonstrates the importance of feature selection for recurrent neural network applied t... more The paper demonstrates the importance of feature selection for recurrent neural network applied to problem of one hour ahead forecasting of gas consumption for office building heating. Although the accuracy of the forecasting is similar for both the feed-forward and the recurrent network, the removal of features leads to accuracy reduction much earlier for the feed-forward network. The recurrent network can perform well even with 50% of features. This brings significant benefits in scenarios, where the neural network is used as a blackbox model of building consumption, which is called by an optimizer that minimizes the consumption. The reduction of input dimensionality leads to reduction of costs related to measurement equipment, but also costs related to data transfer.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2010
In this paper we present a study of the application of Evolutionary Computation methods to the op... more In this paper we present a study of the application of Evolutionary Computation methods to the optimisation of the start-up of a combined cycle power plant. We propose a multiobjective approach considering different objectives for the optimisation in order to reduce the pollution emissions and to maximise the efficiency of the plant. We compare a multiobjective evolutionary algorithm (NSGA-II) with 2 and 5 objectives on a software simulator and then we use different metrics to measure the performances. We show that NSGA-II algorithm is able to provide a set of solutions, defined as Pareto Front, that represent the best trade-off on the different objectives among those the decision maker can choose.
Indoor lighting fault detection and diagnosis using a data fusion approach
Energy Production and Management in the 21st Century, 2014
Science and Technology of Nuclear Installations, 2011
The present paper describes the main features and an application to a real Nuclear Power Plant (N... more The present paper describes the main features and an application to a real Nuclear Power Plant (NPP) of an Integrated Software Environment (in the following referred to as "platform") developed at University of Pisa (UNIPI) to perform Pressurized Thermal Shock (PTS) analysis. The platform is written in Java for the portability and it implements all the steps foreseen in the methodology developed at UNIPI for the deterministic analysis of PTS scenarios. The methodology starts with the thermal hydraulic analysis of the NPP with a system code (such as Relap5-3D and Cathare2), during a selected transient scenario. The results so obtained are then processed to provide boundary conditions for the next step, that is, a CFD calculation. Once the system pressure and the RPV wall temperature are known, the stresses inside the RPV wall can be calculated by mean a Finite Element (FE) code. The last step of the methodology is the Fracture Mechanics (FM) analysis, using weight functions, aimed at evaluating the stress intensity factor (KI) at crack tip to be compared with the critical stress intensity factor KIc. The platform automates all these steps foreseen in the methodology once the user specifies a number of boundary conditions at the beginning of the simulation.
Important Source of Marine Secondary Organic Aerosol from Biogenic Amines
Environmental Science & Technology, 2008
Relevant concentrations of dimethyl- and diethylammonium salts (DMA+ and DEA+) were measured in s... more Relevant concentrations of dimethyl- and diethylammonium salts (DMA+ and DEA+) were measured in submicrometer marine aerosol collected over the North Atlantic during periods of high biological activity (HBA) in clean air masses (median concentration (minimum-maximum)=26(6-56) ng m(-3)). Much lower concentrations were measured during periods of low biological activity (LBA): 1 (<0.4-20) ng m(-3) and when polluted air masses were advected to the sampling site: 2 (<0.2-24) ng m(-3). DMA+ and DEA+ are the most abundantorganic species, second only to MSA, detected in fine marine particles representing on average 11% of the secondary organic aerosol (SOA) fraction and a dominant part (35% on average) of the water-soluble organic nitrogen (WSON). Several observations support the hypothesis that DMA+ and DEA+ have a biogenic oceanic source and are produced through the reaction of gaseous amines with sulfuric acid or acidic sulfates. Moreover, the water-soluble fraction of nascent marine aerosol particles produced by bubble-bursting experiments carried out in parallel to ambient aerosol sampling over the open ocean showed WSON, DMA+, and DEA+ concentrations always below the detection limit, thus excluding an important primary sea spray source.
Source Attribution of Water-Soluble Organic Aerosol by Nuclear Magnetic Resonance Spectroscopy
Environmental Science & Technology, 2007
The functional group compositions of atmospheric aerosol water-soluble organic compoundswere obta... more The functional group compositions of atmospheric aerosol water-soluble organic compoundswere obtained employing proton nuclear magnetic resonance (1H NMR) spectroscopy in a series of recent experiments in several areas of the world characterized by different aerosol sources and pollution levels. Here, we discuss the possibility of using 1H NMR functional group distributions to identifythe sources of aerosol in the different areas. Despite the limited variability of functional group compositions of atmospheric aerosol samples, characteristic 1H NMR fingerprints were derived for three major aerosol sources: biomass burning, secondary formation from anthropogenic and biogenic VOCs, and emission from the ocean. The functional group patterns obtained in areas characterized by one of the above dominant source processes were then compared to identify the dominant sources for samples coming from mixed sources. This analysis shows that H NMR spectroscopy can profitably be used as a valuable tool for aerosol source identification. In addition, compared to other existing methodologies, it is able to relate the source fingerprints to integral chemical properties of the organic mixtures, which determine their reactivity and their physicochemical properties and ultimately the fate of the organic particles in the atmosphere.
Technical and economic analysis of a Smart Public Lighting model
EPJ Web of Conferences, 2012
Model Predictive Control for Building Active Demand Response Systems
Energy Procedia, 2015
The role of data sample size and dimensionality in neural network based forecasting of building heating related variables
Energy and Buildings, 2016
tEnergy consumed in buildings represents a challenge in the context of reduction of greenhouse ga... more tEnergy consumed in buildings represents a challenge in the context of reduction of greenhouse gasesemission. For this reason and due to the growing interest in operative costs reduction the energy used bybuildings (tertiary and privates) for heating, ventilating, and air conditioning (HVAC) is even more inves-tigated. Due to the nature of the energy consumption profile a predictive optimization method is one ofthe solution the scientific literature spreads even more. However optimization techniques need a goodand reliable prediction of the variables of interest over a time horizon. This work focuses on methods toobtain a robust and reliable predictor based on artificial neural networks. For the optimization purposesthe neural model predicts total heating energy consumption (gas), internal air temperature and aggre-gated thermal discomfort 12 h ahead. Training and testing data are simulated using a simulator basedon heat, air and moisture model for building and systems evaluation (HAMBASE), by which a real officebuilding was modeled. Influence of training data sample size and selection of predictor inputs is exam-ined. Several combinations of early stopping condition and network complexity are tested for differenttraining sample sizes. It is observed that the early stopping mechanism is crucial especially but not onlyfor small training data, because it reliably overcomes overfitting problems. Surprisingly, relatively smallnetworks were sufficient or performed best, although examined range of training sample covered up tofive heating seasons. The use of a model tuning is thus supported by the results. Further, two strategiesof selection of suitable input variables are demonstrated. While the input selection does not degrade theprediction performance, it is able to reduce the dimensionality and thus to save computational, commu-nication, time, and data acquisition demands. The importance of inputs selection in HVAC modeling isthus pointed out and demonstrated.
Energy Procedia, 2014
In the building energy efficiency field, developing automatic and accurate fault detection and di... more In the building energy efficiency field, developing automatic and accurate fault detection and diagnosis methods is necessary in order to ensure optimal operations of systems and to save energy. In this paper first, fault detection analysis based on statistical methods where anomalies are detected through a comparison with neighborhood and averaged fault-free values and through a clustering technique is performed. Following the fault detection step, a fault diagnosis analysis based on fuzzy sets and fuzzy logic is implemented. Experimentation is carried out over a one day monitoring data set in December 2013 for the fan coil electric consumption of an actual office building located at ENEA 'Casaccia' Research Centre. Results show the effectiveness of proposed approaches in automatic detection and diagnosis of abnormal building fan coil electric consumption.
Building thermal optimisation through surrogate assisted evolutionary multi-objective algorithms
Urban traffic flow forecasting through statistical and neural network bagging ensemble hybrid modeling
Neurocomputing, 2015
ABSTRACT In this paper we show a hybrid modeling approach which combines Artificial Neural Networ... more ABSTRACT In this paper we show a hybrid modeling approach which combines Artificial Neural Networks and a simple statistical approach in order to provide a one hour forecast of urban traffic flow rates. Experimentation has been carried out on three different classes of real streets and results show that the proposed approach outperforms the best of the methods it puts together.
Primary and secondary biomass burning aerosols determined by proton nuclear magnetic resonance (H-NMR) spectroscopy during the 2008 EUCAARI campaign in the Po Valley (Italy)
Atmospheric Chemistry and Physics Discussions, 2013
Technical and economic analysis of a Smart Public Lighting model
EPJ Web of Conferences, 2012
Illuminazione pubblica adattiva: modellistica dei sistemi intelligenti
Sviluppo di una sperimentazione dimostrativa di" Smart Village" e metodi di progettazione
Advanced Street Lighting Control through Neural Network Ensembling
Metodologia di ottimizzazione multi-obiettivo della climatizzazione termica di edifici. Validazione su sistema di simulazione
Lecture Notes in Computer Science, 2014
Advances in Intelligent Systems and Computing, 2014
The paper demonstrates the importance of feature selection for recurrent neural network applied t... more The paper demonstrates the importance of feature selection for recurrent neural network applied to problem of one hour ahead forecasting of gas consumption for office building heating. Although the accuracy of the forecasting is similar for both the feed-forward and the recurrent network, the removal of features leads to accuracy reduction much earlier for the feed-forward network. The recurrent network can perform well even with 50% of features. This brings significant benefits in scenarios, where the neural network is used as a blackbox model of building consumption, which is called by an optimizer that minimizes the consumption. The reduction of input dimensionality leads to reduction of costs related to measurement equipment, but also costs related to data transfer.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2010
In this paper we present a study of the application of Evolutionary Computation methods to the op... more In this paper we present a study of the application of Evolutionary Computation methods to the optimisation of the start-up of a combined cycle power plant. We propose a multiobjective approach considering different objectives for the optimisation in order to reduce the pollution emissions and to maximise the efficiency of the plant. We compare a multiobjective evolutionary algorithm (NSGA-II) with 2 and 5 objectives on a software simulator and then we use different metrics to measure the performances. We show that NSGA-II algorithm is able to provide a set of solutions, defined as Pareto Front, that represent the best trade-off on the different objectives among those the decision maker can choose.
Indoor lighting fault detection and diagnosis using a data fusion approach
Energy Production and Management in the 21st Century, 2014
Science and Technology of Nuclear Installations, 2011
The present paper describes the main features and an application to a real Nuclear Power Plant (N... more The present paper describes the main features and an application to a real Nuclear Power Plant (NPP) of an Integrated Software Environment (in the following referred to as "platform") developed at University of Pisa (UNIPI) to perform Pressurized Thermal Shock (PTS) analysis. The platform is written in Java for the portability and it implements all the steps foreseen in the methodology developed at UNIPI for the deterministic analysis of PTS scenarios. The methodology starts with the thermal hydraulic analysis of the NPP with a system code (such as Relap5-3D and Cathare2), during a selected transient scenario. The results so obtained are then processed to provide boundary conditions for the next step, that is, a CFD calculation. Once the system pressure and the RPV wall temperature are known, the stresses inside the RPV wall can be calculated by mean a Finite Element (FE) code. The last step of the methodology is the Fracture Mechanics (FM) analysis, using weight functions, aimed at evaluating the stress intensity factor (KI) at crack tip to be compared with the critical stress intensity factor KIc. The platform automates all these steps foreseen in the methodology once the user specifies a number of boundary conditions at the beginning of the simulation.
Important Source of Marine Secondary Organic Aerosol from Biogenic Amines
Environmental Science & Technology, 2008
Relevant concentrations of dimethyl- and diethylammonium salts (DMA+ and DEA+) were measured in s... more Relevant concentrations of dimethyl- and diethylammonium salts (DMA+ and DEA+) were measured in submicrometer marine aerosol collected over the North Atlantic during periods of high biological activity (HBA) in clean air masses (median concentration (minimum-maximum)=26(6-56) ng m(-3)). Much lower concentrations were measured during periods of low biological activity (LBA): 1 (<0.4-20) ng m(-3) and when polluted air masses were advected to the sampling site: 2 (<0.2-24) ng m(-3). DMA+ and DEA+ are the most abundantorganic species, second only to MSA, detected in fine marine particles representing on average 11% of the secondary organic aerosol (SOA) fraction and a dominant part (35% on average) of the water-soluble organic nitrogen (WSON). Several observations support the hypothesis that DMA+ and DEA+ have a biogenic oceanic source and are produced through the reaction of gaseous amines with sulfuric acid or acidic sulfates. Moreover, the water-soluble fraction of nascent marine aerosol particles produced by bubble-bursting experiments carried out in parallel to ambient aerosol sampling over the open ocean showed WSON, DMA+, and DEA+ concentrations always below the detection limit, thus excluding an important primary sea spray source.
Source Attribution of Water-Soluble Organic Aerosol by Nuclear Magnetic Resonance Spectroscopy
Environmental Science & Technology, 2007
The functional group compositions of atmospheric aerosol water-soluble organic compoundswere obta... more The functional group compositions of atmospheric aerosol water-soluble organic compoundswere obtained employing proton nuclear magnetic resonance (1H NMR) spectroscopy in a series of recent experiments in several areas of the world characterized by different aerosol sources and pollution levels. Here, we discuss the possibility of using 1H NMR functional group distributions to identifythe sources of aerosol in the different areas. Despite the limited variability of functional group compositions of atmospheric aerosol samples, characteristic 1H NMR fingerprints were derived for three major aerosol sources: biomass burning, secondary formation from anthropogenic and biogenic VOCs, and emission from the ocean. The functional group patterns obtained in areas characterized by one of the above dominant source processes were then compared to identify the dominant sources for samples coming from mixed sources. This analysis shows that H NMR spectroscopy can profitably be used as a valuable tool for aerosol source identification. In addition, compared to other existing methodologies, it is able to relate the source fingerprints to integral chemical properties of the organic mixtures, which determine their reactivity and their physicochemical properties and ultimately the fate of the organic particles in the atmosphere.