| 1. |
- Fast, Magnus, et al.
(författare)
-
A Novel Approach For Gas Turbine Condition Monitoring Combining Cusum Technique And Artificial Neural Network
- 2009
-
Ingår i: Proceedings Of The Asme Turbo Expo 2009, Vol 1. - 9780791848821 ; , s. 567-574
-
Konferensbidrag (refereegranskat)abstract
- Investigation of a novel condition monitoring approach, combining artificial neural network (ANN) with a sequential analysis technique, has been reported in this paper. For this purpose operational data from a Siemens SGT600 gas turbine has been employed for the training of an ANN model. This ANN model is subsequently used for the prediction of performance parameters of the gas turbine. Simulated anomalies are introduced on two different sets of operational data, acquired one year apart, whereupon this data is compared with corresponding ANN predictions. The cumulative sum (CUSUM) technique is used to improve and facilitate the detection of such anomalies in the gas turbine's performance. The results are promising, displaying fast detection of small changes and detection of changes even for a degraded gas turbine.
|
|
| 2. |
- De, S., et al.
(författare)
-
Development of an artificial neural network model for the steam process of a coal biomass cofired combined heat and power (CHP) plant in Sweden
- 2007
-
Ingår i: Energy. - : Elsevier BV. - 1873-6785 .- 0360-5442. ; 32:11, s. 2099-2109
-
Tidskriftsartikel (refereegranskat)abstract
- The development of a model for any energy system is required for proper design, operation or its monitoring. Models based on accurate mathematical expressions for physical processes are mostly useful to understand the actual operation of the plant. However, for large systems like combined heat and power (CHP) plants, such models are usually complex in nature. The estimation of output parameters using these physical models is generally time consuming, as these involve many iterative solutions. Moreover, the complete physical model for new equipment may not be available. However, artificial neural network (ANN) models, developed by training the network with data from an existing plant, may be very useful especially for systems for which the full physical model is yet to be developed. Also, such trained ANN models have a fast response with respect to corresponding physical models and are useful for realtime monitoring of the plant. In this paper, the development of an ANN model for the biomass and coal cofired CHP plant of Visthamnsverket at Helsingborg, Sweden has been reported. The feed forward with back propagation ANN model was trained with data from this plant. The developed model is found to quickly predict the performance of the plant with good accuracy. (C) 2007 Elsevier Ltd. All rights reserved.
|
|
| 3. |
- Fast, Carina, 1944-
(författare)
-
Sju barn lär sig läsa och skriva : Familjeliv och populärkultur i möte med förskola och skola
- 2007
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- This thesis is an ethnographic study in which seven Swedish children from different social and cultural environments were followed over three years. The study began when the children were four years old.The main aim of this thesis is to investigate through what social and cul-tural practices seven children meet literacy events in their families. A second aim is to track informal and everyday literacy events which the children take part in or practice on their own. A further aim is to study the transition be-tween home, preschool, preschool class and primary school in order to de-termine to what extent and in what ways the children are allowed to use their previous experience with and knowledge of literacy.The results show that the children are socialised in practices rich in liter-acy events via their culture, traditions, language and religion. The children practise reading and writing in a number of contexts. This occurs long before the pupils have been exposed to formal education.In the work on this study, it has also become clear that the seven children, regardless of their cultural, language or socioeconomic background, share experiences and knowledge relating to popular culture and the media. The children have a common understanding of texts in the form of words, names, images and icons. The value of this knowledge and experience is often as-signed by the children themselves in their contact with other children.The seven children, in this study, come to preschool and primary school with a wealth of experience in literacies from their family lives. Some chil-dren are allowed to bring their experience to the classroom. For them, there is continuity between literacy practices at home and at school. Others are forced to leave their experience outside the classroom. What is common to all children is that their knowledge about literacy related to popular culture and the media has a low cultural value in instructional settings.
|
|
| 4. |
- Fast, Jonatan, et al.
(författare)
-
Hot-carrier separation in heterostructure nanowires observed by electron-beam induced current
- 2020
-
Ingår i: Nanotechnology. - : IOP Publishing. - 1361-6528 .- 0957-4484. ; 31:39
-
Tidskriftsartikel (refereegranskat)abstract
- The separation of hot carriers in semiconductors is of interest for applications such asthermovoltaic photodetection and third-generation photovoltaics. Semiconductor nanowiresoffer several potential advantages for effective hot-carrier separation such as: a high degree ofcontrol and flexibility in heterostructure-based band engineering, increased hot-carriertemperatures compared to bulk, and a geometry well suited for local control of light absorption.Indeed, InAs nanowires with a short InP energy barrier have been observed to produce electricpower under global illumination, with an open-circuit voltage exceeding the Shockley-Queisserlimit. To understand this behaviour in more detail, it is necessary to establish control over theprecise location of electron-hole pair-generation in the nanowire. In this work we performelectron-beam induced current measurements with high spatial resolution, and demonstrate therole of the InP barrier in extracting energetic electrons.We interprete the results in terms ofhot-carrier separation, and extract estimates of the hot carriers’ mean free path.
|
|
| 5. |
|
|
| 6. |
- Fast, Magnus, et al.
(författare)
-
Application of artificial neural network to the condition monitoring and diagnosis of a CHP plant
- 2008
-
Ingår i: [Host publication title missing]. - 9788392238140 ; , s. 981-988
-
Konferensbidrag (refereegranskat)abstract
- The objective of this study has been to create an online system for condition monitoring and diagnosis for a combined heat- and power plant in Sweden. This system consists of artificial neural network models, representing each main component of the combined heat- and power plant, accompanied with a graphical user interface. The artificial neural network models are integrated on a power generation information manager server in the combined heat- and power plant computer system and the graphical user interface is made available on workstations connected to this server. The artificial neural network models have been constructed with the multi-layer feed-forward network type and trained with operational data from the combined heat- and power plant using back-propagation. The plant consists of a Siemens gas turbine with a heat recovery steam generator and a bio fuelled boiler and its steam cycle. Steam from the heat recovery steam generator and the bio fuelled boiler expands together in one common steam turbine, producing both electricity and heat. Each component, i.e. gas turbine, heat recovery steam generator, bio fuelled boiler and steam turbine, are modelled separately with artificial neural network. To ensure accurate predictions from the models a baseline is established after which all training data is collected.
|
|
| 7. |
- Fast, Magnus, et al.
(författare)
-
Application of artificial neural networks to the condition monitoring and diagnosis of a combined heat and power plant
- 2010
-
Ingår i: Energy. - : Elsevier BV. - 1873-6785 .- 0360-5442. ; 35:2, s. 1114-1120
-
Konferensbidrag (refereegranskat)abstract
- The objective of this study has been to create an online system for condition monitoring and diagnosis of a combined heat and power plant in Sweden. The system in question consisted of artificial neural network models, representing each main component of the combined heat and power plant, connected to a graphical user interface. The artificial neural network models were integrated on a power generation information manager server in the computer system of the combined heat and power plant, and the graphical user interface was made available on workstations connected to this server. The plant comprised a Siemens SGT800 gas turbine with a heat recovery steam generator as well as a bio-fueled boiler and its steam cycle. Steam from the heat recovery steam generator and the bio-fueled boiler expanded together in a common steam turbine, producing both electricity and heat. The artificial neural network models were trained with operational data from the components of the combined heat and power plant. Accurate predictions from the ANN (Artificial neural network) models in combination with an undemanding integration in the power plant's computer system were some of the main conclusions from this study. (C) 2009 Elsevier Ltd. All rights reserved.
|
|
| 8. |
- Fast, Magnus
(författare)
-
Artificial Neural Networks for Gas Turbine Monitoring
- 2010
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Due to the deregulation of the electricity market the power producers are forced to continuously investigate various means of maintaining/increasing their profits. Improving the electrical efficiency through hardware upgrades is probably the most commonly employed measure, although the interest for enhancements with regard to plant operation is on the rise. Plant operation improvement is often measured in RAM (reliability, availability and maintainability) which acts as an indication of how well a plant can be utilized. The availability can be increased by employing various monitoring tools allowing the maintenance to be based on the condition rather than equivalent operating hours, thereby extending the periods between overhauls. The reliability can also be increased by employing a combination of monitoring tools alerting the plant operators before faults are fully developed. Modern power plants are equipped with distributed control systems delivering data to the control room through a considerable number of sensors. This data enables the development of data-driven methods for tasks such as condition monitoring, diagnosis and sensor validation. Artificial neural networks have proven suitable for the non-linear modeling of power plants and its components, and represent the data modeling tools used in this research. Some of the results of the case studies are very accurate ANN models for different types of gas turbines. Furthermore, the integration of these models and the development of user interfaces for online condition monitoring have been demonstrated.
|
|
| 9. |
- Fast, Magnus, et al.
(författare)
-
Condition based maintenance of gas turbines using simulation data and artificial neural network: A demonstration of feasibility
- 2008
-
Ingår i: Proceedings of the ASME Turbo Expo 2008. - 0791838242 ; , s. 153-161
-
Konferensbidrag (refereegranskat)abstract
- Gas turbine maintenance is crucial due to high cost for the replacement of its components and associated loss of power during shutdown period. Conventional scheduled maintenance, based on equivalent operating hours, is not the best alternative as it can require unnecessary shut downs. Condition based maintenance is an attractive alternative as it decreases unnecessary shut downs and has other advantages for both the manufacturers and the plant owners. However, this has shown to be a complex/difficult task. A number of methods and approaches have been presented to develop condition monitoring tools during the past decade. Condition monitoring tools can e.g. be developed by means of training artificial neural networks (ANN) with historical operational data. Such tools can be used for online gas turbine performance prediction where input data from the plant is fed directly to the trained ANN models. The predicted outputs from the models are compared with corresponding measurements and possible deviations are evaluated. With this method both recoverable degradation, caused by fouling, and irrecoverable degradation, caused by wear, can be detected and hence both compressor wash and overhaul periods optimized. However, non-availability of operational data at the beginning of the gas turbine operation may cause problems for the development of ANN based condition monitoring tools. Simulation data, on the other hand, may be generated by using a manufacturer’s engine design program. This data can be used for training artificial neural networks to overcome the problem of non-availability of operational data. ANN models trained with simulation data could be used to monitor the engine from the very beginning of its operation. A demonstration case using a Siemens gas turbine has been shown for this proposed method by comparing two ANN models, one trained with operational data and the other with simulation data. For the comparison an arbitrary section of operational data was used to produce predictions from both models, whereupon these were plotted with corresponding measured data. The comparison shows that the trends are very similar but the parameter values for the measured and the simulated data are shifted by a constant. Using this knowledge, one can provide an ANN based engine monitoring tool that could be adjusted to a certain engine using engine performance test data. The study shows promising results and motivates further investigations in this field.
|
|
| 10. |
|
|
