Machine Health Monitoring Of HVAC System

 Background

In the Pharma Industry CGMP specified functions of the QU (Quality Unit) are aligned with current quality system techniques (211.22):

Ensuring that controls are properly established and performed throughout manufacturing activities. The HVAC system's dependability and availability are crucial for pharmaceutical operations

To avoid product cross-contamination To assure the dependability of the manufacturing process To comply with FDA production operating rules

Problem Statement

Maintenance of CGMP facilities is necessary to ensure that they operate in a qualified and validated state and are fit for their intended use. Traditional preventive maintenance has direct costs, may require downtime of equipment or entire production lines, and may increase the risk of contamination. Preventive maintenance can be effective, but it is insufficient to prevent unplanned downtime through unpredicted failure; it can also result in unnecessary maintenance.

Our Solution

Machine anomaly detection and diagnostics solution, VIBit, and the iEdge360 IoT platform are deployed to monitor the pharma factory in real time.

Deployment

VIBit Edge IoT sensors were installed at Motor DE & NDE and Fan DE & NDE bearings AHU Unit.

In order to track the health of the machine, VIBit gathers tri-axial vibration and temperature every 10 seconds and sends it to the iEdge 360 cloud utilizing the plant's Wi-Fi and 4G connections.

The iEdge 360 platform's machine anomaly detection & diagnostics application delivers real-time machine health status and early anomaly detection utilizing a variety of AI/ML-based models.

Subject matter experts and the customer success team establish alert limitations, and the VIBit application notifies users about alerts via emails.

Machine operators may also detect anomalies with various LED statuses on the VIBit sensor.

Early structural looseness defect identification in a timely manner prevented unplanned downtime saving $100,000 in production loss and replacement costs

Real-time Machine Monitoring

On the AHU unit, overall vibration on motor DE side, RMS velocity (3-direction) high alerts were detected.

AI/ML in Action

Fault Type: Structural Looseness

Diagnostic:

Motor Drive end Structural Looseness

Risk Level: Medium Pattern of Structural Looseness fault observed at lower amplitude. They synchronous energy at 1x and multiples of running speed increased. Likely the structural Looseness is diagnosed. 

Motor Drive end Structural Looseness:

Continue to monitor the fault vibration trend for any significant increase. Reinforce the mounting and improve stiffness. Check the looseness of mountina bolts. Failed grout or cracks in concrete bases. cracks or other forms weakness in machinery mount.

Auto-diagnostics & recommendation algorithm detected a rotating looseness of risk level high on Nov 9th and provided recommendations to inspect the machine base

foot location and tighten the bolts with uniform torque.

Inferences

Customer Success and Subject Matter Team Expertise

Before the machine condition reaches the critical stage, our subject matter experts (SME) started studying vibration patterns, FFT charts, and time waveforms to plan maintenance action.

The synchronous energy increased in the FFT spectrum at running speeds of 1x and multiples. SME determined that the motor foot's construction was loose.

Impact

Early detection of rotating looseness and bearing wear on the AHU motor was leveraged by the customer maintenance team to help them schedule maintenance work orders and complete tasks during scheduled maintenance time.

Benefits

Shift unplanned to planned maintenance

Avoided cost of failure

Increased production performance and reliability