Gearbox Failure Diagnosis Using a Multisensor Data-Fusion Machine-Learning-Based Approach

Abstract

Failure detection and diagnosis are of crucial importance for the reliable and safe operation of industrial equipment and systems, while gearbox failures are one of the main factors leading to long-term downtime. Condition-based maintenance addresses this issue using several expert systems for early failure diagnosis to avoid unplanned shutdowns. In this context, this paper provides a comparative study of two machine-learning-based approaches for gearbox failure diagnosis. The first uses linear predictive coefficients for signal processing and long short-term memory for learning, while the second is based on mel-frequency cepstral coefficients for signal processing, a convolutional neural network for feature extraction, and long short-term memory for classification. This comparative study proposes an improved predictive method using the early fusion technique of multisource sensing data. Using an experimental dataset, the proposals were tested, and their effectiveness was evaluated considering predictions based on statistical metrics.

Fault Diagnosis Method for Industrial Robots Based on DBN Joint Information Fusion Technology

“…Information fusion technology is a multi-level and multi-faceted statistical process in its essence by detecting and combining the estimation of multiple sources of data to obtain information that information fusion can use [ 14 ]. Unlike simple signal processing techniques, multi-source information fusion techniques are suitable for handling multi-modal and conflicting forms of data and can achieve different levels and conditions of information fusion [ 15 ]. Thus, it performs well in improving the real time and reliability of mechanical systems, increasing the detectability of mechanical systems, and reducing the uncertainty of mechanical equipment.…”

Internal Combustion Engine in Agricultural Machinery Field Relying on Artificial Fish Swarm Algorithm

“…During the implementation of the algorithm, the behavior evaluation is used to select which behavior the AF performs, the bulletin board is set to record the global historical optimal position of the AF, and finally the termination condition of the algorithm is set to terminate the algorithm. Behavior evaluation: according to the nature of the problem to be optimized and whether the problem is a maximum problem or a minimum problem, set the food concentration function, calculate the food concentration value at the initial position of the AF in this iteration, then execute the above three behaviors respectively, calculate the fitness value corresponding to the position after each behavior moves, compare the fitness change values of different behaviors, and select the behavior leading to the best direction in the alternative rows [14][15].…”

Global Optimal Time Trajectory Planning of Construction Machinery Robot Considering Ant Colony Algorithm

“…When the environment information changes (with unknown dynamic obstacles added), the improved ACA is called again for secondary obstacle avoidance optimization. At this time, the optimization path obtained in the global environment is the initial path of the secondary optimization process, and finally a short and smooth path is found, at the same time, it can avoid the optimal feasible path of dynamic and static obstacles [12]. The overall process of path planning is shown in Figure 1.…”