With advances in IT technologies even complex machines with several gears and bearings, running at varying rotational speed can be effciently monitored with an algorithm consisting of a combination of signal envelope and synchronous resampling. For implementation, robust advanced software development technologies are applied, based on an object-oriented approach, supported by formal methodologies, like UML. Another concept important for distributed systems is middleware, like RPC or DCOM. The paper presents a proposal of the architecture of the distributed vibration monitoring and diagnostic system, based on the approaches mentioned above. Principal components are described and their functionality is discussed. Special interest was put on flexibility of architecture and optimization of remote access. XML was proposed as universal and effcient means of configuration storage. The structure of the monitored machine should also be stored in XML. A proposal of such a data structure is described for a gas compressor. Another key component is the database server. It needs to handle multiple databases and multiple users, delivering fast and reliable data. Vibration signal handling can be more effcient when signals are processed on the server side. Effciency is essential when remote access to vibration data is required and data link is limited. Such a mechanism, based on plug-in approach, is proposed and described. Finally, the paper presents a case study which describes the installation of the system on a gas compressor. The measured data are presented after different processing algorithms. High resolution envelope order spectrum was able to detect the bearing failure.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Vibrodiagnosis helps in detecting incipient faults in rotating machines like pumps and generators. Early detection prevents from undesired breakdown of the machine and allows to schedule maintenance times. The application of neural networks in classification of the rotating machine condition has been described in this work. Different types of networks and methods on feature extraction was described and compared. Additionally it was proposed a novelty feature set consisted of harmonics from vibration spectrum. The set were combined with using of probabilistic neural networks which has been modified that it could recognize defects that did not occur in the training set. Such architecture was tested in detection of two defects, shaft misalignment and mass unbalance. It was found that such network works better than a multi layered perceptron with statistical features.
W wyniku prac badawczych prowadzonych w Katedrze Górnictwa Odkrywkowego opracowano oryginalną metodę dokumentowania robót strzałowych w otoczeniu kopalń odkrywkowych. Zbudowano również komputerowe systemy pomiarowo-analityczne, które znalazły zastosowanie w kopalniach jako Kopalniana Stacja Monitoringu Drgań KSMD i Mała Stacja Monitoringu Drgań Explo 504. W artykule zwrócono uwagę na możliwości analityczne systemów oraz dokumentowanie i archiwizowanie poziomu ich oddziaływania w sposób ciągły. Artykuł przedstawia również propozycję powołania w AGH Laboratorium dokumentowania oddziaływania robót strzałowych w otoczeniu.
EN
There is worked out original method of shooting documentation in the surroundings of opencasts as a result of research work carried out in the Chair of Open-pit Mining. There are built analytical and measuring computer systems, as swell, which found application in mines as the Mine Vibration Monitoring Station KSMD and Smali Vibration Monitoring Station Explo 504. Attention to be paid in this article to analytical preferences of systems as well as level documentation and archiving of their impact in a continuous manner. The article sets a proposal for creation in the Academy of Mining and Metallurgy in Kraków of Laboratory for shooting impact documentation in the surroundings.
Symptomy diagnostyczne wielu diagnozowanych obiektów mają dużą wrażliwość na zakłócenia. Zatem symptomowa macierz obserwacji (SMO) ma składową przypadkową. Może to zachodzić z tytułu niekontrolowanej zmiany obciążenia. Ma to istotny wpływ na końcową ocenę stanu obiektu. Jako remedium w pracy zaproponowano uśrednianie początkowej wartości symptomów, a dopiero następnie zastosowanie rozkładu względem wartości szczególnych (SVD). Jak się okazuje to proste pociągnięcie może przynieść liczące się rezultaty w podwyższeniu stabilności i diagnostycznej informatywności wyników przetwarzania SMO.
EN
Machine symptoms of condition of many diagnosed object have some sensitivity to disturbances, either of stochastic nature or as the load sensitivity, if machine works in nonstationary regime. As a remedy to this inconvenience we can apply the averaging of initial symptom value before classical application of singular value decomposition (SVD). As it is shown in the paper, this simple preprocessing can bring us the good result in the stability and diagnostic information value of applied symptoms.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.