Jairo Viola

Jairo Viola, YangQuan Chen, Jing Wang

Failure detection is employed in the industry to improve system performance and reduce costs due to unexpected malfunction events. So, a good dataset of the system is desirable for designing an automated failure detection system. However, industrial process datasets are unbalanced and contain little information about failure behavior due to the uniqueness of these events and the high cost for running the system just to get information about the undesired behaviors. For this reason, performing correct training and validation of automated failure detection methods is challenging. This paper proposes a methodology called FaultFace for failure detection on Ball-Bearing joints for rotational shafts using deep learning techniques to create balanced datasets. The FaultFace methodology uses 2D representations of vibration signals denominated faceportraits obtained by time-frequency transformation techniques. From the obtained faceportraits, a Deep Convolutional Generative Adversarial Network is employed to produce new faceportraits of the nominal and failure behaviors to get a balanced dataset. A Convolutional Neural Network is trained for fault detection employing the balanced dataset. The FaultFace methodology is compared with other deep learning techniques to evaluate its performance in for fault detection with unbalanced datasets. Obtained results show that FaultFace methodology has a good performance for failure detection for unbalanced datasets.

Jairo Viola, Alberto Radici, YangQuan Chen

This paper presents the design of multivariable temperature control for a refrigeration system based on vapor compression employing the internal model control technique. The refrigeration system is based on the PID18 benchmark, which is a $2\times 2$ MIMO system. The controlled output variables of the refrigeration system are the cooling power managed through the outlet temperature of the evaporator and the superheating degree at the condenser. The input variables of the system are the valve opening and the compressor speed. System identification is performed by applying stepped signals to the input variables, resulting in four transfer functions estimated with a Box-Jenkins model. From the MIMO system transfer functions, the relative gain array is calculated to determinate the best variables to be paired. After that, according to the variables to be paired, the corresponding transfer functions are reduced to order two to design a PID controller for each output variable employing the internal model control technique. Then, the controllers are contrasted employing a set of quantitative performance indexes with the control results achieved in the PID18 workshop. Obtained results show that the proposed Internal Model controllers have better performance than most of the proposed controllers at the PID18.