Reference no: EM133296096
Smart Industries & Digital Manufacturing
Digital Twin
Digital twin is employed in smart manufacturing/industry 4.0, which can be used to evaluate the current condition of the asset and, more importantly, predict future behaviour, refine the control, or optimize operation. Such a digital twin is an up-to-date representation of a real asset in operation e.g. centrifugal pumps shown in Fig.1 which are employed across 50 oil wells.
From the 50 well sites each with one centrifugal pump, the sensor measures the discharge flow, pump head, and pump torque data and transmits this to a central database site. There is a requirement of creating a Digital-Twin model to predict which of the pumps may be operating normally or are likely to develop a fault in near future. The centrifugal pump data is stored given in files in TABLE 2.
Task 1.1 Develop algorithms to plot confidence ellipsoids to detect anomaly detection of the 50 pumps. Assume parameter uncertainty as a 99% confidence region. List which of the 50 pumps do not require immediate maintenance to avoid failing in the near future. You do not need to plot the Mahalanobis distances, only measure them.
Task 1.2 Repeat your results for a parameter uncertainty of 64% confidence region. Can you briefly explain the changes in the results, and which of the two results would you prefer to make your final decision on?
(~200 words )
Programmable Logical Controller
Safety is a priority in the design of a PLC system. Emergency stop buttons and safety guard switches are hard-wired and must not depend on the PLC software for implementation. This is to ensure that when there is a failure of the stop switch or PLC, the system is automatically safe i.e. the system must be fail-safe.
Task 2.1 Explain your reasoning in a maximum of ~300 words with an example of how a fail-safe system can be implemented with PLCs for motor control.
Task 2.2 Implement PLC ladder design for the following Boolean expression in Simulink. Plot the input-output waveforms for all combinations of input signals for a time period of 30 seconds.