• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • 2
  • 2
  • Tagged with
  • 6
  • 6
  • 6
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Proposta de um controlador automático de trens utilizando lógica nebulosa preditiva. / Proposal for an automatic trains control using predictive fuzzy logic.

Rachel, Flávio Monteiro 26 July 2006 (has links)
Esta dissertação apresenta um estudo da viabilidade de aplicação de ferramentas de Inteligência Artificial, em especial da aplicação da lógica nebulosa preditiva (lógica fuzzy), em sistemas de controle automático de trens. Para isto, um panorama do contexto de aplicação do estudo é apresentado, incluindo a problemática dos transportes nas grandes cidades e a utilização de técnicas de computação em sistemas metroferroviários. Também são apresentados os princípios básicos de controle automático de trens, de lógica nebulosa preditiva e de sistemas de controle, particularmente os sistemas especialistas e as aplicações críticas de controle. Um modelo de um sistema de controle utilizando lógica nebulosa preditiva é então proposto e detalhado. Em seguida, estudos de caso para os trens do metropolitano de Sendai (Japão) e de São Paulo são efetuados. Simulações, testes comparativos e a análise dos resultados são realizados, levando às considerações finais sobre a viabilidade da aplicação. A conclusão é que, mesmo se tratando de uma aplicação crítica de controle, envolvendo a segurança dos passageiros, a utilização de ferramentas de Inteligência Artificial, especialmente a lógica nebulosa preditiva, é perfeitamente viável, desde que uma atenção especial com a segurança do sistema seja tomado. Neste caso, esta atenção é tomada por meio da implantação de um módulo supervisor (módulo ATP) que realiza uma monitoração constante das entradas e saídas do sistema de controle. O ATP atua, preferencialmente, ao sistema de controle para garantir a sua segurança. Embora as normas européias (IEE e CENELEC) não recomendem a utilização de ferramentas de Inteligência Artificial em aplicações críticas de controle, este estudo demonstra que é possível efetuar a aplicação destas ferramentas, aproveitando os benefícios por elas proporcionados, sem comprometer a segurança do sistema. / This dissertation presents a study on the viability of applying Artificial Intelligence tools, especially predictive fuzzy logic application, in automatic train control systems. For this purpose, an overview of the study application context is presented, including transportation troubling in big cities and computational techniques used in subway and railway systems. The basic principles of automatic train control, predictive fuzzy logic and control systems, particularly specialist and control critical systems applications are also presented. A model for a control system using predictive fuzzy logic is then proposed and detailed. Next, study cases for Sendai (Japan) and São Paulo subway trains are conducted. Simulations, comparative tests and results analysis are carried out, leading to the final considerations about the application viability. The conclusion is that, even being a critical control application, involving passenger safety, the use of Artificial Intelligence tools, especially predictive fuzzy logic, is perfectible viable, since special attention is given to the safety of the system. In this case, this attention is given by means of the implementation of a supervisory module (ATP module) that constantly monitors control system inputs and outputs. The ATP acts preferably on the control system in order to guarantee its safety. Although European standards (IEE and CENELEC) do not recommend the use of Artificial Intelligence tools in critical system applications, this study shows that it is possible to apply these tools, enjoying the benefits provided by them, without compromising the system safety.
2

Proposta de um controlador automático de trens utilizando lógica nebulosa preditiva. / Proposal for an automatic trains control using predictive fuzzy logic.

Flávio Monteiro Rachel 26 July 2006 (has links)
Esta dissertação apresenta um estudo da viabilidade de aplicação de ferramentas de Inteligência Artificial, em especial da aplicação da lógica nebulosa preditiva (lógica fuzzy), em sistemas de controle automático de trens. Para isto, um panorama do contexto de aplicação do estudo é apresentado, incluindo a problemática dos transportes nas grandes cidades e a utilização de técnicas de computação em sistemas metroferroviários. Também são apresentados os princípios básicos de controle automático de trens, de lógica nebulosa preditiva e de sistemas de controle, particularmente os sistemas especialistas e as aplicações críticas de controle. Um modelo de um sistema de controle utilizando lógica nebulosa preditiva é então proposto e detalhado. Em seguida, estudos de caso para os trens do metropolitano de Sendai (Japão) e de São Paulo são efetuados. Simulações, testes comparativos e a análise dos resultados são realizados, levando às considerações finais sobre a viabilidade da aplicação. A conclusão é que, mesmo se tratando de uma aplicação crítica de controle, envolvendo a segurança dos passageiros, a utilização de ferramentas de Inteligência Artificial, especialmente a lógica nebulosa preditiva, é perfeitamente viável, desde que uma atenção especial com a segurança do sistema seja tomado. Neste caso, esta atenção é tomada por meio da implantação de um módulo supervisor (módulo ATP) que realiza uma monitoração constante das entradas e saídas do sistema de controle. O ATP atua, preferencialmente, ao sistema de controle para garantir a sua segurança. Embora as normas européias (IEE e CENELEC) não recomendem a utilização de ferramentas de Inteligência Artificial em aplicações críticas de controle, este estudo demonstra que é possível efetuar a aplicação destas ferramentas, aproveitando os benefícios por elas proporcionados, sem comprometer a segurança do sistema. / This dissertation presents a study on the viability of applying Artificial Intelligence tools, especially predictive fuzzy logic application, in automatic train control systems. For this purpose, an overview of the study application context is presented, including transportation troubling in big cities and computational techniques used in subway and railway systems. The basic principles of automatic train control, predictive fuzzy logic and control systems, particularly specialist and control critical systems applications are also presented. A model for a control system using predictive fuzzy logic is then proposed and detailed. Next, study cases for Sendai (Japan) and São Paulo subway trains are conducted. Simulations, comparative tests and results analysis are carried out, leading to the final considerations about the application viability. The conclusion is that, even being a critical control application, involving passenger safety, the use of Artificial Intelligence tools, especially predictive fuzzy logic, is perfectible viable, since special attention is given to the safety of the system. In this case, this attention is given by means of the implementation of a supervisory module (ATP module) that constantly monitors control system inputs and outputs. The ATP acts preferably on the control system in order to guarantee its safety. Although European standards (IEE and CENELEC) do not recommend the use of Artificial Intelligence tools in critical system applications, this study shows that it is possible to apply these tools, enjoying the benefits provided by them, without compromising the system safety.
3

Sampled-data control of high-speed trains,

21 June 2002 (has links)
by Alexander H. Levis and Michael Athans. / Bibliography: p. 27. / PB 177 669. / Prpared for the U.S. Dept. of Transportation under Contract C-85-65, DSR Project no. 76105.
4

Periodic operation of a diesel locomotive for fuel optimization

Pendegrass, Barry L. January 1985 (has links)
An attempt was made to reduce fuel costs of a diesel-electric locomotive by operating the locomotive in a periodic manner, as opposed to operating at a constant velocity. The periodic operation consisted of accelerating the locomotive in a high throttle position and then deccelerating at a low throttle position. An SD40 Locomotive was modeled to test the periodic operation. The periodic operation was actually found to offer no improvement in fuel performance over that of constant velocity operation. A modification to the diesel engine that takes advantage of the periodic operation is suggested that will give a better fuel performance. The modification to the engine would not be possible with constant velocity operation. With the modification, the periodic operation was found to save fuel over the constant velocity operation. A controller was then designed to implement the periodic operation. The key component of the controller is an observer to determine unknown hill forces. The controller was found to work successfully. / M.S.
5

Kunimodellen : En utbildningsmodell för signaltekniker på Roslagsbanan / The Kuni model : An education model for signal maintenance technicians at Roslagsbanan

Kuniholm, Johanna January 2019 (has links)
För att kunna hitta och åtgärda signalfel på Roslagsbanan i Stockholm är det viktigt att de tekniker som arbetar med signalsystemet har rätt kunskaper och känner sig trygga i hur systemet fungerar. En del av säkerhetssystemet på Roslagsbanan kallas ATC (automatic train control) och ger tågföraren information om var tåget framför befinner sig, vilket besked nästa signal ger och hur växlar och vägbommar står. ATC ingriper även genom att bromsa tåget automatiskt om föraren inte stannar vid en stoppsignal. För att effektivisera felsökning i ATCsystemet vill företaget som sköter drift och underhåll av signalsystemet, banan och elförsörjningen på Roslagsbanan, Strukton Rail AB, genomföra utbildningar för sina signaltekniker. De vill ha en mall för hur utbildningar i felsökning kan se ut som kan användas med övningsutrustning, och som kan appliceras på andra tekniska system. För att skapa den här mallen var mina frågeställningar ”Hur kan en pedagogisk modell för utbildning inom felsökning i säkerhetssystemet ATC på Roslagsbanan se ut? Hur påverkar modellen deltagarnas aktivitet jämfört med andra utbildningar inom järnvägsbranschen?”. Det var viktigt från Struktons sida att deltagarna skulle vara aktiva och samarbeta med varandra under utbildningen. Efter att ha läst om det sociokulturella perspektivet på lärande, accelererat lärande och lärande på arbetsplatsen och även hållit intervjuer med två tekniker på Strukton och två utbildare i andra sammanhang i järnvägsbranschen för att höra vad de tycker är bra utbildning har jag skapat en utbildningsmodell för internutbildning i signalsystemet ATC, döpt till Kunimodellen. Modellen baseras på att deltagarna tillsammans löser olika felsökningsfall. Utbildningen börjar med en genomgång av systemet som kan vara olika djupgående beroende på deltagarnas förkunskaper. Sedan löser deltagarna fall genom att sitta i grupper och diskutera fram en plan för hur de ska felsöka och sedan testa sin plan praktiskt i testutrustningen där felet är planterat. Beroende på utbildningsbehov och tillgänglig tid kan de få lösa olika många och olika svåra fall, men fallen bör täcka in så många komponenter i systemet som möjligt. I slutet av utbildningen finns ett reflektionsmoment där deltagarna får berätta för varandra vad de har lärt sig under dagen. För att få en mätbar indikation på om deltagarna har lärt sig något under utbildningen får de fylla i en enkät precis innan och precis efter utbildningen med frågor om hur trygga de känner sig på en femgradig skala att felsöka i de olika komponenterna. Det genomfördes två utbildningar med Kunimodellen, där vardera innehöll två fall att lösa, för sammanlagt sju deltagare. För att mäta deltagaraktiviteten under utbildningarna fyllde jag i ett observationsschema under 20 minuter per timme där jag registrerade om varje deltagare hade talat eller inte under 40 intervall om 30 sekunder var, vilket jag även gjorde på två andra utbildningar som signaltekniker går för att få en kontrollgrupp. Dessutom hölls gruppintervjuer med deltagarna efter utbildningarna för att få höra vad de tyckte om modellen och vad som kan förbättras. Resultaten visar att deltagarna har varit nöjda i intervjuerna och tyckt att det varit en bra modell att arbeta efter, men att det saknades viss teknisk information kring testutrustningen som finns i verkligheten. De ville även ha mer djupgående information i genomgången i början av utbildningen med fokus på bland annat varför ATC-systemet finns. Deltagarnas upplevda trygghet i felsökning ökade i enkäten efter utbildningen, men de var inte entydigt mer aktiva i utbildningar med Kunimodellen än i de utbildningar som jämförts med. Kunimodellen förändrades inte särskilt mycket mellan utbildningarna. Resultaten ska dock bara ses som indikationer på grund av det lilla antalet deltagare, att deltagarna inte nödvändigtvis var representativa för hela gruppen och att alla personer som blev intervjuade kände mig som intervjuade. / To be able to find and solve signal failures in Roslagsbanan in Stockholm, it is important that the maintenance technicians working with the signalling system has the right knowledge and feel confident with how the system works. One part of the safety system at Roslagsbanan is called ATC (automatic train control) which sends information to the train driver about where adjacent trains are located and how upcoming signals, switches and road crossing barriers are set. The ATC also interferes by applying brakes if the driver does not stop at a stop signal. To make troubleshooting of the ATC system more efficient, the company maintaining the signalling system, tracks and power supply at Roslagsbanan, Strukton Rail AB, would like to further educate their maintenance technicians. They want a model for training sessions in troubleshooting that can be used with training equipment and can be applied to other technical systems. To create this model my research questions were ”What could a pedagogical model for troubleshooting education on the safety system ATC at Roslagsbanan look like? How does the model affect the participants’ activity compared to other education sessions in the railway business?”. It was important to Strukton that participants should be active and cooperate with each other during the session. After reading about the sociocultural perspective on learning, accelerated learning and workplace learning and also conducting interviews with two technicians at Strukton and two teachers in other parts of the railway business to hear what they value as good education, I created an educational model for inservice training of the signalling system ATC, named the Kuni model. The model is based on participants solving troubleshooting cases together. An education session starts with an overview of the system that can vary in detail depending on the participants’ previous knowledge. Then the participants solve cases by discussing in groups and agreeing on a plan on how they should troubleshoot, and then try their plan on the test equipment where the error is planted. Depending on their need for education and the available time, they can get a different number of cases on different levels, but the cases should try to cover as many of the system’s components as possible. At the end of the session there is time for reflection where the participants tell each other what they have learnt. To get a measurable indication of if the participants have learned anything, they fill out a questionnaire just before and just after the education with questions on how confident they feel troubleshooting the different components. Two such sessions were held with the Kuni model, each with two cases to solve, for a total of seven participants. To measure participant activity, I filled out an observation form for 20 minutes per hour where I registered whether each participant had spoken or not for 40 intervals of 30 seconds each, which I also did at two other training sessions that signalling technicians attend to get a control group. Additionally, I conducted group interviews with the participants after the sessions to see what they thought of the model and what could be improved. The results from the interviews show that the participants were content and thought it was a good model to work with, but that some technical information about the test equipment that exists in the real world was missing. They also wanted more detailed information in the initial overview with focus on, among other things, why the ATC system exists. The participants’ selfestimated confidence in troubleshooting increased after the session, but they were not significantly more active during training with the Kuni model than the control groups. The Kuni model itself did not change very much between sessions. The results should however only be seen as indications due to the small number of participants, the participants not necessarily being representative for the whole group and since all people interviewed knew me when I interviewed them.
6

Kunimodellen : En utbildningsmodell för signaltekniker på Roslagsbanan / The Kuni model : An education model for signal maintenance technicians at Roslagsbanan

Kuniholm, Johanna January 2019 (has links)
För att kunna hitta och åtgärda signalfel på Roslagsbanan i Stockholm är det viktigt att de tekniker som arbetar med signalsystemet har rätt kunskaper och känner sig trygga i hur systemet fungerar. En del av säkerhetssystemet på Roslagsbanan kallas ATC (automatic train control) och ger tågföraren information om var tåget framför befinner sig, vilket besked nästa signal ger och hur växlar och vägbommar står. ATC ingriper även genom att bromsa tåget automatiskt om föraren inte stannar vid en stoppsignal. För att effektivisera felsökning i ATC-systemet vill företaget som sköter drift och underhåll av signalsystemet, banan och elförsörjningen på Roslagsbanan, Strukton Rail AB, genomföra utbildningar för sina signaltekniker. De vill ha en mall för hur utbildningar i felsökning kan se ut som kan användas med övningsutrustning, och som kan appliceras på andra tekniska system. För att skapa den här mallen var mina frågeställningar ”Hur kan en pedagogisk modell för utbildning inom felsökning i säkerhetssystemet ATC på Roslagsbanan se ut? Hur påverkar modellen deltagarnas aktivitet jämfört med andra utbildningar inom järnvägsbranschen?”. Det var viktigt från Struktons sida att deltagarna skulle vara aktiva och samarbeta med varandra under utbildningen. Efter att ha läst om det sociokulturella perspektivet på lärande, accelererat lärande och lärande på arbetsplatsen och även hållit intervjuer med två tekniker på Strukton och två utbildare i andra sammanhang i järnvägsbranschen för att höra vad de tycker är bra utbildning har jag skapat en utbildningsmodell för internutbildning i signalsystemet ATC, döpt till Kunimodellen. Modellen baseras på att deltagarna tillsammans löser olika felsökningsfall. Utbildningen börjar med en genomgång av systemet som kan vara olika djupgående beroende på deltagarnas förkunskaper. Sedan löser deltagarna fall genom att sitta i grupper och diskutera fram en plan för hur de ska felsöka och sedan testa sin plan praktiskt i testutrustningen där felet är planterat. Beroende på utbildningsbehov och tillgänglig tid kan de få lösa olika många och olika svåra fall, men fallen bör täcka in så många komponenter i systemet som möjligt. I slutet av utbildningen finns ett reflektionsmoment där deltagarna får berätta för varandra vad de har lärt sig under dagen. För att få en mätbar indikation på om deltagarna har lärt sig något under utbildningen får de fylla i en enkät precis innan och precis efter utbildningen med frågor om hur trygga de känner sig på en femgradig skala att felsöka i de olika komponenterna. Det genomfördes två utbildningar med Kunimodellen, där vardera innehöll två fall att lösa, för sammanlagt sju deltagare. För att mäta deltagaraktiviteten under utbildningarna fyllde jag i ett observationsschema under 20 minuter per timme där jag registrerade om varje deltagare hade talat eller inte under 40 intervall om 30 sekunder var, vilket jag även gjorde på två andra utbildningar som signaltekniker går för att få en kontrollgrupp. Dessutom hölls gruppintervjuer med deltagarna efter utbildningarna för att få höra vad de tyckte om modellen och vad som kan förbättras. Resultaten visar att deltagarna har varit nöjda i intervjuerna och tyckt att det varit en bra modell att arbeta efter, men att det saknades viss teknisk information kring testutrustningen som finns i verkligheten. De ville även ha mer djupgående information i genomgången i början av utbildningen med fokus på bland annat varför ATC-systemet finns. Deltagarnas upplevda trygghet i felsökning ökade i enkäten efter utbildningen, men de var inte entydigt mer aktiva i utbildningar med Kunimodellen än i de utbildningar som jämförts med. Kunimodellen förändrades inte särskilt mycket mellan utbildningarna. Resultaten ska dock bara ses som indikationer på grund av det lilla antalet deltagare, att deltagarna inte nödvändigtvis var representativa för hela gruppen och att alla personer som blev intervjuade kände mig som intervjuade. / To be able to find and solve signal failures in Roslagsbanan in Stockholm, it is important that the maintenance technicians working with the signalling system has the right knowledge and feel confident with how the system works. One part of the safety system at Roslagsbanan is called ATC (automatic train control) which sends information to the train driver about where adjacent trains are located and how upcoming signals, switches and road crossing barriers are set. The ATC also interferes by applying brakes if the driver does not stop at a stop signal. To make troubleshooting of the ATC system more efficient, the company maintaining the signalling system, tracks and power supply at Roslagsbanan, Strukton Rail AB, would like to further educate their maintenance technicians. They want a model for training sessions in troubleshooting that can be used with training equipment and can be applied to other technical systems. To create this model my research questions were ”What could a pedagogical model for troubleshooting education on the safety system ATC at Roslagsbanan look like? How does the model affect the participants’ activity compared to other education sessions in the railway business?”. It was important to Strukton that participants should be active and cooperate with each other during the session. After reading about the sociocultural perspective on learning, accelerated learning and workplace learning and also conducting interviews with two technicians at Strukton and two teachers in other parts of the railway business to hear what they value as good education, I created an educational model for in-service training of the signalling system ATC, named the Kuni model. The model is based on participants solving troubleshooting cases together. An education session starts with an overview of the system that can vary in detail depending on the participants’ previous knowledge. Then the participants solve cases by discussing in groups and agreeing on a plan on how they should troubleshoot, and then try their plan on the test equipment where the error is planted. Depending on their need for education and the available time, they can get a different number of cases on different levels, but the cases should try to cover as many of the system’s components as possible. At the end of the session there is time for reflection where the participants tell each other what they have learnt. To get a measurable indication of if the participants have learned anything, they fill out a questionnaire just before and just after the education with questions on how confident they feel troubleshooting the different components. Two such sessions were held with the Kuni model, each with two cases to solve, for a total of seven participants. To measure participant activity, I filled out an observation form for 20 minutes per hour where I registered whether each participant had spoken or not for 40 intervals of 30 seconds each, which I also did at two other training sessions that signalling technicians attend to get a control group. Additionally, I conducted group interviews with the participants after the sessions to see what they thought of the model and what could be improved. The results from the interviews show that the participants were content and thought it was a good model to work with, but that some technical information about the test equipment that exists in the real world was missing. They also wanted more detailed information in the initial overview with focus on, among other things, why the ATC system exists. The participants’ self-estimated confidence in troubleshooting increased after the session, but they were not significantly more active during training with the Kuni model than the control groups. The Kuni model itself did not change very much between sessions. The results should however only be seen as indications due to the small number of participants, the participants not necessarily being representative for the whole group and since all people interviewed knew me when I interviewed them.

Page generated in 0.1185 seconds