|
|
Measurement of weak magnetic field in 3D space
Bár, Martin ; Klusáček, Stanislav (referee) ; Havránek, Zdeněk (advisor)
The aim of this thesis is to theoretically examine the magnetic field of miniature cylindrical NdFeB magnets, compare the simulation results to real-world measurements, and design a magnetic field probe using suitable sensors. A FEM simulation was conducted using Ansys AIM and FEMM 4.2. The simulation shows that the magnetic induction on the surface of the magnet depends on the diameter/height ratio. The simulation results also show that 21 µm thick protective layer of Ni-Cu-Ni metal on the surface of NdFeB magnets lowers flux density of the smallest magnet sample (1x1 mm) by up to 14 %. It was also concluded that a larger magnet edge radius results in a higher flux density on the surface of the magnet. The biggest differences between simulation data and data measured by the F.W. BELL gaussmeter occurred in the smallest magnet sample. Using the FEMM 4.2 simulation software, a three-axis magnetic field probe was designed. Potential problems with this probe design have been identified. A simple procedure for probe calibration was proposed.
|
|
|
Implementation and practical verification of methods for predictive identification of rolling bearings failures
Bár, Martin ; Havránek, Zdeněk (referee) ; Klusáček, Stanislav (advisor)
The aim of this thesis is to identify and classify rolling bearing failures. The first part of the thesis deals with the diagnosis of bearings using in-house measured data. Faults were introduced into two bearings. The first one was deformed and corroded. In the second bearing, the outer raceway was damaged. Vibration data was collected at regular intervals and processed in MATLAB. The values of the statistical features indicated faults in both bearings. Envelope analysis showed that both bearings had developed a fault on the outer raceway and a gradual fault on the cage. In the second part of this thesis, machine learning methods were used to classify defective bearings using the CWRU data set. The accelerometer data were divided into blocks in two ways. Vibration images were created from these blocks for convolutional neural networks (CNNs). The best prediction accuracy was achieved by 1D convolutional neural network (1DCNN) (99.2 %), followed by neural network (94.6 %) and SVM (94.4 %). Random Forest and SVM are the best methods when the training set is reduced, and among CNNs, MATLAB architecture and 1DCNN are the best. The most noise resistant method is Random Forest and neural network, and among CNNs, 1DCNN is the best. Methods using statistical features perform better than CNNs on extremely noisy data. Convolutional networks do not achieve good accuracy, which is probably due to the conversion of raw accelerometer data into vibration images.
|
|
|
Implementation and practical verification of methods for predictive identification of rolling bearings failures
Bár, Martin ; Havránek, Zdeněk (referee) ; Klusáček, Stanislav (advisor)
The aim of this thesis is to identify and classify rolling bearing failures. The first part of the thesis deals with the diagnosis of bearings using in-house measured data. Faults were introduced into two bearings. The first one was deformed and corroded. In the second bearing, the outer raceway was damaged. Vibration data was collected at regular intervals and processed in MATLAB. The values of the statistical features indicated faults in both bearings. Envelope analysis showed that both bearings had developed a fault on the outer raceway and a gradual fault on the cage. In the second part of this thesis, machine learning methods were used to classify defective bearings using the CWRU data set. The accelerometer data were divided into blocks in two ways. Vibration images were created from these blocks for convolutional neural networks (CNNs). The best prediction accuracy was achieved by 1D convolutional neural network (1DCNN) (99.2 %), followed by neural network (94.6 %) and SVM (94.4 %). Random Forest and SVM are the best methods when the training set is reduced, and among CNNs, MATLAB architecture and 1DCNN are the best. The most noise resistant method is Random Forest and neural network, and among CNNs, 1DCNN is the best. Methods using statistical features perform better than CNNs on extremely noisy data. Convolutional networks do not achieve good accuracy, which is probably due to the conversion of raw accelerometer data into vibration images.
|
|
|
Measurement of weak magnetic field in 3D space
Bár, Martin ; Klusáček, Stanislav (referee) ; Havránek, Zdeněk (advisor)
The aim of this thesis is to theoretically examine the magnetic field of miniature cylindrical NdFeB magnets, compare the simulation results to real-world measurements, and design a magnetic field probe using suitable sensors. A FEM simulation was conducted using Ansys AIM and FEMM 4.2. The simulation shows that the magnetic induction on the surface of the magnet depends on the diameter/height ratio. The simulation results also show that 21 µm thick protective layer of Ni-Cu-Ni metal on the surface of NdFeB magnets lowers flux density of the smallest magnet sample (1x1 mm) by up to 14 %. It was also concluded that a larger magnet edge radius results in a higher flux density on the surface of the magnet. The biggest differences between simulation data and data measured by the F.W. BELL gaussmeter occurred in the smallest magnet sample. Using the FEMM 4.2 simulation software, a three-axis magnetic field probe was designed. Potential problems with this probe design have been identified. A simple procedure for probe calibration was proposed.
|
guest ::
RSS feed.