In this report we investigate whether an embedded system can be used to determine materials quality by analysing the acoustic resonance frequencies. Through experiments the necessary specifications are established and suitable circuits constructed. Signal analysis is performed by a FFT-based algorithm. We have verified that the system is succesful in detecting the five strongest resonance frequencies and list these in order of amplitude. By using embedded devices, it's possible to lower the cost of purchase as well as power consumption dramatically compared to alternative solutions. / I denna rapport undersöker vi hurvida ett inbyggt system kan användas för att bestämma materialkvalité genom analys av akustiska resonansfrekvenser. Genom experiment fastställs de specifikationer systemet måste ha, varefter lämpliga kretsar konstrueras. Med hjälp av en FFT-baserad algoritm utförs viss signalanalys. Vi har verifierat att systemet framgångsrikt kan detektera de fem starkaste resonansfrekvenserna och lista dessa efter amplitud. Genom användandet av inbyggda system kan energiförbrukningen och inköpskostnaden bli mycket lägre än alternativa lösningar.

Cette thèse propose une méthode d'analyse des signaux triphasés pour la surveillance d'état des systèmes électromécaniques. La méthode proposée repose sur l'utilisation de la transformée en composantes symétriques instantanées et d'outils simples de traitement du signal pour détecter les défauts électriques et mécaniques dans de tels systèmes. Les avantages de cette approche triphasée par rapport à une approche monophasée pour la surveillance d'état sont étudiés en détail. Tout d'abord, pour les défauts électriques, l'utilisation de la transformée triphasée permet de séparer les composantes symétriques et asymétriques, et facilite ainsi la détection d'un déséquilibre électrique. Ensuite, pour les défauts mécaniques, l'approche par transformée en composantes symétriques permet de travailler dans des espaces avec un meilleur rapport signal à bruit. En effet, en appliquant le même traitement à la fois en monophasé et en triphasé sur les composantes symétriques, on observe que certains défauts mécaniques ne sont détectables qu’en utilisant la séquence positive des composantes symétriques. La méthodologie complète et les algorithmes pour calculer les indicateurs de défaut pour les défauts électriques et mécaniques sont donnés et les résultats sont validés sur signaux synthétiques et expérimentaux. En termes d'application, l'accent est mis sur la surveillance d'état des composants de turbines éoliennes. Toutefois, le procédé proposé peut être appliqué à des systèmes électromécaniques en général et peut facilement être étendu à des systèmes polyphasés. / This thesis proposes a three-phase electrical signals analysis method for condition monitoring of electromechanical systems. The proposed method relies on the use of instantaneous symmetrical components (ISCs) transform and simple signal processing tools to detect both electrical and mechanical faults in such systems. The advantages of using this three-phase approach for condition monitoring instead of single-phase ones are thoroughly detailed. Firstly, for electrical faults the use of the three-phase transform separates the balanced and unbalanced components thus making electrical unbalance detection easier. Secondly, for mechanical faults the ISCs approach has better signal-to-noise ratio (SNR). Indeed, by applying the same processing to both single-phase and ISCs, some mechanical faults are only detectable using the positive-sequence ISC. The complete methodology and algorithms to compute fault indicators for both electrical and mechanical faults are given and the results are validated using synthetic and experimental signals. In terms of application, the focus was on condition monitoring of wind turbine components. However, the proposed method can be applied on electromechanical systems in general and can easily be extended to poly-phase systems.

The aim of this thesis is to define a perceptual scale for the ‘Time-Frequency’ analysis of music signals. The equal tempered ‘Bach ’ scale is a suitable scale, since it covers most of the genres of music and the error is equally distributed for each semi-tone. However, it may be necessary to allow a tolerance of around 50 cents or half the interval of the Bach scale, so that the interval can accommodate other common intonation schemes. The thesis covers the formulation of the Bach scale filter-bank as a time-varying model. It makes a comparative study with other commonly used perceptual scales. Two applications for the Bach scale filter-bank are also proposed, namely automated segmentation of speech signals and transcription of singing voice for query-by-humming applications. Even though this filter-bank is suggested with a motivation from music, it could also be applied to speech. A method for automatically segmenting continuous speech into phonetic units is proposed. The results, obtained from the proposed method, show around 82% accuracy for the English and 85% accuracy for the Hindi databases. This is an improvement of around 2 -3% when the performance is compared with other popular methods in the literature. Interestingly, the Bach scale filters perform better than the filters designed for other common perceptual scales, such as Mel and Bark scales. ‘Musical transcription’ refers to the process of converting a musical rendering or performance into a set of symbols or notations. A query in a ‘query-by-humming system’ can be made in several ways, some of which are singing with words, or with arbitrary syllables, or whistling. Two algorithms are suggested to annotate a query. The algorithms are designed to be fairly robust for these various forms of queries. The first algorithm is a frequency selection based method. It works on the basis of selecting the most likely frequency components at any given time instant. The second algorithm works on the basis of finding time-connected contours of high energy in the ‘Time-Frequency’ plane of the input signal. The time domain algorithm works better in terms of instantaneous pitch estimates. It results in an error of around 10 -15%, while the frequency domain method results in an error of around 12 -20%. A song rendered by two different people will have quite a few different properties. Their absolute pitches, rates of rendering, timbres based on voice quality and inaccuracies, may be different. The thesis discusses a method to quantify the distance between two different renderings of musical pieces. The distance function has been evaluated by attempting a search for a particular song from a database of a size of 315, made up of songs sung by both male and female singers and whistled queries. Around 90 % of the time, the correct song is found among the top five best choices picked. Thus, the Bach scale has been proposed as a suitable scale for representing the perception of music. It has been explored in two applications, namely automated segmentation of speech and transcription of singing voices. Using the transcription obtained, a measure of the distance between renderings of musical pieces has also been suggested.