A study is presented to explore the performance of bearing fault diagnosis using three types of artificial neural networks (ANNs), namely, Multilayer Perceptron (MLP) with BP algorithm, Radial Basis Function (RBF) network, and Probabilistic Neural Network (PNN). The time domain vibration signals of a rotating machine with normal and defective bearings are preprocessed using Lapalce wavelet analysis technique for feature extraction. The extracted features are used as inputs to all three ANN classifiers: MLP, RBF, and PNN for four-class: Healthy, outer, inner and roller faults identification. The procedure is illustrated using the experimental vibration data of a rotating machine with different bearing faults. The results show the relative effectiveness of three classifiers in detection of the bearing condition with different learning speeds and success rates.

Introduction: Early recognition of stroke with its two types Ischemic and Hemorrhagic, is one of the most crucial research points, commonly used methods are CT- (computerized tomography), and MRI- (Magnetic resonance imaging). These techniques cause a delay in the detection of the condition, which causes permanent disability. The main reason behind the fatal consequences of stroke is the delay of detection. Therefore, this research paper aims to early detection of the type of stroke without delay until the appropriate diagnosis of each type is made, and then the appropriate treatment without delay. Method: Using a non-invasive and fast technique to determine the stroke type by wave, we simulate and design a vessel containing a liquid as a laminar flow with the same density and velocity of blood, and it was surrounded by a Homogenized multi-turn coil consisting of (n) turns to represent the magnetic field, using specific frequency (HZ) with Electrical field in coil current (A) to see the changing in magnetic flux density (MFD), Depending on the changes in MFD, the flow of blood in laminar flow can be affected by clotting (Ischemic) or Hemorrhagic (cutting) in our vessel designed. We have built three different scenarios to apply the technique which are: First: Normal Scenario (where the blood in vessel has no problem), second: clotting (ischemic, where the vessel blocked in specific three position) and Third: Cutting (Hemorrhagic, where the vessel cut in certain nine positions). Results: This paper presents-through our own design-the studying of applying the electromagnetic waves on blood inside the vessel to detect the stroke type in our three scenarios (normal, ischemic three positions or hemorrhagic nine positions), Studying the magnetic field and laminar flow. This study covered in three areas. First: coil geometry analysis, Second: stationary, and Third: frequency domain. through the changes in Magnetic Flux Density -MFD- waves. The results were promising and distinct for distinguishing between the three scenarios which are normal, ischemic (3 positions) and hemorrhagic (9 positions) the results of MFD are: 0.09 to 3.3*10^-3, 0.08 to 3.15*10^-4, 0.15 to 6.2*10^-3 respectively.