This work is aimed towards the development of an artificially intelligent search algorithm used in conjunction with an Auto Associative Neural Network (AANN) to help locate and reconstruct faulty sensor inputs in control systems. The AANN can be trained to detect when sensors go faulty but the problem of locating the faulty sensor still remains. The search algorithm aids the AANN to help locate the faulty sensors and reconstruct their actual values. The algorithm uses domain specific heuristics based on the inherent behavior of the AANN to achieve its task. Common sensor errors such as drift, shift and random errors and the algorithms response to them have been studied. The issue of noise has also been investigated. These areas cover the first part of this work. The second part focuses on the development of a web interface that implements and displays the working of the algorithm. The interface allows any client on the World Wide Web to connect to the engineering software called MATLAB. The client can then simulate a drift, shift or random error using the graphical user interface and observe the response of the algorithm.

Among the renewable energies, wind energy presents the highest growth in installed capacity and penetration in modern power systems. This is why reliability of wind turbines becomes an important topic in research and industry. To this end, condition monitoring (or health monitoring) systems are needed for wind turbines. The core of any condition monitoring system (CMS) are fault diagnosis algorithms whose task is to provide early warnings upon the occurrence of incipient (small magnitude) faults. Thanks to the use of CMS we can avoid premature breakdowns and reduce significatively maintenance costs. The present thesis deals with fault diagnosis in sensors of a doubly-fed induction generator (DFIG) for wind turbine (WT) applications. In particular we are interested in performing fault detection and isolation (FDI) of incipient faults affecting the measurements of the three-phase signals (currents and voltages) in a controlled DFIG. Although different authors have dealt with FDI for sensors in induction machines and in DFIGs, most of them rely on the machine model with constant parameters. However, the parameter uncertainties due to changes in the operating conditions will produce degradation in the performance of such FDI systems. In this work we propose a systematic methodology for the design of sensor FDI systems with the following characteristics: i) capable of detecting and isolating incipient additive (bias, drifts) and multiplicative (changes in the sensor gain) faults, ii) robust against changes in the references/disturbances affecting the controlled DFIG as well as modelling/parametric uncertainties, iii) residual generation system based on a multi-observer strategy to enhance the isolation process, iv) decision system based on statistical-change detection algorithms to treat the entire residual and perform fault detection and isolation at once. Three novel sensor FDI approaches are proposed. The first is a signal-based approach, that uses the model of the balanced three-phase signals (currents or voltages) for residual generation purposes. The second is a model-based approach that accounts for variation in the parameters. Finally, a third approach that combines the benefits of both the signal- and the model-based approaches is proposed. The designed sensor FDI systems have been validated using measured voltages, as well as simulated data from a controlled DFIG and a speed-controlled induction motor. In addition, in this work we propose a discrete-time multiple input multiple output (MIMO) regulator for each power converter, namely for the rotor side converter (RSC) and for the grid side converter (GSC). In particular, for RSC control, we propose a modified feedback linearization technique to obtain a linear time invariant (LTI) model dynamics for the compensated DFIG. The novelty of this approach is that the compensation does not depend on highly uncertain parameters such as the rotor resistance. For GSC control, a LTI model dynamics is derived using the ideas behind feedback linearization. The obtained LTI model dynamics are used to design Linear Quadratic Gaussian (LQG) regulators. A single design is needed for all the possible operating conditions.

This work is aimed towards the development of an artificially intelligent search algorithm used in conjunction with an Auto Associative Neural Network (AANN) to help locate and reconstruct faulty sensor inputs in control systems. The AANN can be trained to detect when sensors go faulty but the problem of locating the faulty sensor still remains. The search algorithm aids the AANN to help locate the faulty sensors and reconstruct their actual values. The algorithm uses domain specific heuristics based on the inherent behavior of the AANN to achieve its task. Common sensor errors such as drift, shift and random errors and the algorithms response to them have been studied. The issue of noise has also been investigated. These areas cover the first part of this work. The second part focuses on the development of a web interface that implements and displays the working of the algorithm. The interface allows any client on the World Wide Web to connect to the engineering software called MATLAB. The client can then simulate a drift, shift or random error using the graphical user interface and observe the response of the algorithm.

The substantial growth in the use of automated in-process sensing technologies creates great opportunities for manufacturers to detect abnormal manufacturing processes and identify the root causes quickly. It is critical to locate and distinguish two types of faults - process faults and sensor faults. The procedures to monitor and diagnose process and sensor mean shift faults are presented with the assumption that the manufacturing processes can be modeled by a linear fault-quality model. A W control chart is developed to monitor the manufacturing process and quickly detect the occurrence of the sensor faults. Since the W chart is insensitive to process faults, when it is combined with U chart, both process faults and sensor faults can be detected and distinguished. A unit-free index referred to as the sensitivity ratio (SR) is defined to measure the sensitivity of the W chart. It shows that the sensitivity of the W chart is affected by the potential influence of the sensor measurement. A Bayesian variable selection based fault diagnosis approach is presented to locate the root causes of the abnormal processes. A Minimal Coupled Pattern (MCP) and its degree are defined to denote the coupled structure of a system. When less than half of the faults within an MCP occur, which is defined as sparse faults, the proposed fault diagnosis procedure can identify the correct root causes with high probability. Guidelines are provided for the hyperparameters selection in the Bayesian hierarchical model. An alternative CML method for hyperparameters selection is also discussed. With the large number of potential process faults and sensor faults, an MCMC method, e.g. Metropolis-Hastings algorithm can be applied to approximate the posterior probabilities of candidate models. The monitor and diagnosis procedures are demonstrated and evaluate through an autobody assembly example.

Freeway work zones with lane closures can adversely affect mobility, safety, and sustainability. Capacity drop phenomena near work zone areas can further decrease work zone capacity and exacerbate traffic congestion. To mitigate the negative impacts caused by freeway work zones, many variable speed limits (VSL) control methods have been proposed to proactively regulate the traffic flow. However, a simple yet robust VSL controller that considers the nonlinearity induced by the associated capacity drop is still needed. Also, most existing studies of VSL control neglected the impacts of traffic sensor failures that commonly occur in transportation systems. Large deviations of traffic measurements caused by sensor faults can greatly affect the reliability of VSL controllers. To address the aforementioned challenges, this research proposes a fault-tolerant VSL controller for a freeway work zone with consideration of sensor faults. A traffic flow model was developed to understand and describe the traffic dynamics near work zone areas. Then a VSL controller based on sliding mode control was designed to generate dynamic speed limits in real time using traffic measurements. To achieve VSL control fault tolerance, analytical redundancy was exploited to develop an observer-based method and an interacting multiple model with a pseudo-model set (IMMP) based method for permanent and recurrent sensor faults respectively. The proposed system was evaluated under realistic freeway work zone conditions using the traffic simulator SUMO. This research contributes to the body of knowledge by developing fault-tolerant VSL control for freeway work zones with reliable performance under permanent and recurrent sensor faults. With reliable sensor fault diagnosis, the fault-tolerant VSL controller can consistently reduce travel time, safety risks, emissions, and fuel consumption. Therefore, with a growing number of work zones due to aging road infrastructure and increasing demand, the proposed system offers broader impacts through congestion mitigation and consistent improvements in mobility, safety, and sustainability near work zones. / Thesis / Doctor of Philosophy (PhD) / Freeway work zones can increase congestion with higher travel time, safety risk, emissions and fuel consumption. This research aims to improve traffic conditions near work zones using a variable speed limits control system. By exploiting redundant traffic information, a variable speed limit control system that is insensitive to traffic sensor failures is presented. The proposed system was evaluated under realistic freeway work zone conditions in a simulation environment. The results show that the proposed system can reliably detect sensor failures and consistently provide improvements in mobility, safety and sustainability despite the presence of traffic sensor failures.

No description available.

Among the renewable energies, wind energy presents the highest growth in installed capacity and penetration in modern power systems. This is why reliability of wind turbines becomes an important topic in research and industry. To this end, condition monitoring (or health monitoring) systems are needed for wind turbines. The core of any condition monitoring system (CMS) are fault diagnosis algorithms whose task is to provide early warnings upon the occurrence of incipient (small magnitude) faults. Thanks to the use of CMS we can avoid premature breakdowns and reduce significatively maintenance costs.<p><p>The present thesis deals with fault diagnosis in sensors of a doubly-fed induction generator (DFIG) for wind turbine (WT) applications. In particular we are interested in performing fault detection and isolation (FDI) of incipient faults affecting the measurements of the three-phase signals (currents and voltages) in a controlled DFIG. Although different authors have dealt with FDI for sensors in induction machines and in DFIGs, most of them rely on the machine model with<p>constant parameters. However, the parameter uncertainties due to changes in the operating conditions will produce degradation in the performance of such FDI systems.<p><p>In this work we propose a systematic methodology for the design of sensor FDI systems with the following characteristics: i) capable of detecting and isolating incipient additive (bias, drifts) and multiplicative (changes in the sensor<p>gain) faults, ii) robust against changes in the references/disturbances affecting the controlled DFIG as well as modelling/parametric uncertainties, iii) residual generation system based on a multi-observer strategy to enhance the isolation process, iv) decision system based on statistical-change detection algorithms to treat the entire residual and perform fault detection and isolation at once.<p><p>Three novel sensor FDI approaches are proposed. The first is a signal-based approach, that uses the model of the balanced three-phase signals (currents or voltages) for residual generation purposes. The second is a model-based approach<p>that accounts for variation in the parameters. Finally, a third approach that combines the benefits of both the signal- and the model-based approaches is proposed. The designed sensor FDI systems have been validated using measured voltages, as well as simulated data from a controlled DFIG and a speed-controlled induction<p>motor. <p><p>In addition, in this work we propose a discrete-time multiple input multiple output (MIMO) regulator for each power converter, namely for the rotor side converter (RSC) and for the grid side converter (GSC). In particular, for RSC<p>control, we propose a modified feedback linearization technique to obtain a linear time invariant (LTI) model dynamics for the compensated DFIG. The novelty of this approach is that the compensation does not depend on highly uncertain parameters such as the rotor resistance. For GSC control, a LTI model dynamics<p>is derived using the ideas behind feedback linearization. The obtained LTI model dynamics are used to design Linear Quadratic Gaussian (LQG) regulators. A single design is needed for all the possible operating conditions. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished

Effektiviteten av automatiserade reningsprocesser inom avloppsreningsverk beror ytterst på kvaliteten av de mätdata som fås från installerade instrument. Givarfel påverkar verkens styrning och är ofta anledningen till att olika reglerstrategier fallerar. Idag saknas standardiserade riktlinjer för hur instrumenteringsarbetet på svenska reningsverk bör organiseras vilket ger begränsade förutsättningar för reningsverken att resurseffektivt nå sina utsläppskrav. Mycket forskning har gjorts på att optimera olika reglerstrategier men instrumentens roll i verkens effektivitet har inte givits samma uppmärksamhet. Syftet med detta examensarbete har varit att undersöka hur instrumentering på reningsverk kan organiseras och struktureras för att säkerställa mätdata av god kvalitet och att undersöka effekter av givarfel på reningsprocessen. Inom arbetet genomfördes en litteraturstudie där instrumentering på reningsverk under-söktes. Effekter av givarfel på reningsprocessen undersöktes genom att simulera en fördenitrifikationsprocess i Benchmark Simulation Model no. 2 där bias och drift implementerades i olika givare. Simuleringar visade att positiva bias (0,10–0,50 mg/l) i en ammoniumgivare inom en kaskadreglering bidrar till att öka luftförbrukningen med cirka 4–25 %. Vidare resulterade alla typer av fel i DO-givare i den sista aeroba bassängen i en markant större påverkan på reningsprocessen än samma fel i DO-givare i någon av de tidigare aeroba bassängerna. Om den sista aeroba bassängen är designad för att hålla lägre syrehalter är DO-givaren i den bassängen den viktigaste DO-givaren att underhålla. Positiva bias (200–1 000 mg/l) i TSS-givare som används för att styra uttaget av överskottsslam bidrog till kraftiga ökningar av mängden ammonium med cirka 29–464 % i utgående vatten. Negativ drift i DO-givare visade att stora besparingar i luftningsenergi, cirka 4 %, var möjliga genom ett mer frekvent underhåll av DO-givarna. Huruvida ett instrument lider av ett positivt eller negativt givarfel, bias eller drift, kommer att påverka hur mycket och i vilken mån reningsprocessen påverkas. Studien av givarfel visade att effekten av ett positivt eller ett negativt fel varierade och att effekten på reningsprocessen inte var linjär. Effekten av givarfel på reningsprocessen kommer i slutändan att bero på den implementerade reglerstrategin, inställningar i regulatorerna och på den styrda processen. / The effectiveness of automated treatment processes within wastewater treatment plants ultimately depend on the quality of the measurement data that is given from the installed sensors. Sensor faults affect the control of the treatment plants and are often the reason different control strategies fail. Today there is a lack of standardized guidelines for how to organize and work with online sensors at Swedish wastewater treatment plants which limits the opportunities for treatment plants to reach their effluent criteria in a resource efficient manner. Much research has been done on ways to optimize control strategies but the role of sensors in the efficiency of the treatment plants has not been given the same level of attention. The purpose of this thesis has been to examine how instrumentation at wastewater treatment plants can be organized and structured to ensure good quality measurement data and to examine how sensor faults affect the treatment process. Within the thesis a literature study was conducted where instrumentation at wastewater treatment plants was examined. The effects of sensor faults were examined by simulating a pre-denitrification process in Benchmark Simulation Model no. 2 where off-sets (biases) and drift where added to measurements from different implemented sensors. The simulations showed that positive off-sets (0.10–0.50 mg/l) in an ammonium sensor within a cascaded feedback-loop adds to the energy consumption used for aeration by roughly 4-25%. It could further be shown that all types of faults in a DO sensor in the last aerated basin had significantly larger effect on the treatment process than the same fault in any of the other DO sensors in the preceding basins. If the last aerated basin is designed to have low DO concentrations the DO sensor in that basin is the most important DO sensor to maintain. Positive off-sets (200–1 000 mg TSS/l) in suspended solids sensors used for control of waste activated sludge flow contributed to large increases of ammonia, by 29-464%, in effluent waters. Negative drift in DO sensors showed that significant savings in aeration energy, roughly 4%, was possible to achieve with more frequent maintenance. Whether a sensor is affected by a positive or a negative fault, be it off-set or drift, will affect how much and in what way the treatment process will be affected. The study of sensor faults showed that the effect of a positive or a negative fault varied and that the effect on the treatment process was not linear. The effect of a sensor fault on the treatment process will ultimately depend on the implemented control strategy, settings in the controllers and on the controlled process.