Search Results

Early fault detection is vital in maintaining system stability and to decrease the cost associated with maintenance. This paper presents an approach to identify the fuel line failure for a diesel engine based on vibration signals and machine learning. Vibration measurements are performed on the fuel line of the engine for both normal and faulty conditions for engine ramp up condition. After acquiring the time domain vibration signals, various features were extracted and have been analyzed in time and time-frequency domains. Based on the most effective feature, a machine learning model (i.e., support vector machine (SVM)) for fault diagnosis is developed. Results showed that the proposed SVM based model can detect the fuel line fault correctly. This study can be useful for early detection of this critical fault in diesel engine and take useful decision before any catastrophic failure happens because of this fault.

A typical cab used on agriculture machines is made up of a metal frame structure with large enclosing panels of glass, plastic, and metal. Acoustic treatments such as coatings, textiles and foams are used within the cab for aesthetics but also to mediate undesired noise. To develop effective designs for the cab to combat noise, accurate tools for measurement, and predictive methods for sound transmission loss are needed. This paper focuses on Sound Transmission Loss (STL) of the rear upper panel of a cab used in agriculture machines. Results from CAE based tools such as Statistical Energy Analysis (SEA), Finite Element Analysis (FEA) and Hybrid FE-SEA methods are compared to measurements. The panel studied included features such as curvature, deep drawn beads with a glass window and a damping coating. The simulation results are refined by incorporating methods for accurate modeling of ribs stiffness, curvature effect and radiation efficiency by synthetic modal approach.