Ensuring Network Uptime through Reliable Power Management

Electrical Energy Management

Network uptime is critically dependent on a reliable power plant. When a rectifier trips in the middle of a power cycle, it starts to choke output power, impacts battery charging, and drives the site towards a precipitous network blackout. Additionally, a partially charged battery causes more frequent running on expensive diesel generator (DG) power.

Ranked among the top 3 mobile service providers globally, our client sought to improve the power cycle at its telecom sites through continuous visibility of energy parameters, battery, and power plant condition monitoring. The client wanted us to monitor the power plant infrastructure, detect anomalies, generate new hypotheses, and identify root causes of infrastructure failures.

Our Solution

Key rectifier parameters such as battery charge/discharge currents, bus bar voltage, and device temperatures are monitored continuously through the Remote IoT diagnostic appliance. This pre-configured device can be easily installed by client technicians in a few minutes. The appliance polls real-time input energy parameters over a MODBUS interface. This cloud-aggregated data provides a bird's-eye view of performance analytics and anomalies across thousands of sites in real-time.

While the grid power is ON, an anomalous and sudden discharge of the battery often indicates rectifier failure. This failure causes the bus bar voltage to drop as the system transitions into discharge mode. These events, which appear random and are resolved without clear cause a few minutes later, are notoriously difficult to detect and anticipate. With thousands of sites spread over vast areas, such occurrences pose a significant challenge for operations teams. Without proper detection, planning spares becomes nearly impossible, leading to inefficient resource allocation.

Impact

Faulty rectifiers start to overwork the good ones, resulting in asymmetric ageing. Moreover, a power plant running on a partial bank of rectifiers tends to clip output current and DC power, leading to incomplete battery charging. An incomplete battery charging cycle causes the site to rely on expensive DG power, increasing energy bills. At 8KW per site, a 10% DG overuse (non-urban sites tend to run DG for an average of 5 hours/day) can lead to an excessive energy bill of Rs. 23,000 per site annually. For a tower company running 100,000 assets, even a 10% rectifier fault rate can lead to an energy cost overrun of Rs. 23 Crores annually. This is besides the SLA penalties and goodwill loss when one of these sites goes down unexpectedly due to an undetected fault.

Our Analysis

Time series analysis and correlation between rectifier current and environmental temperature helped detect the root cause. High shelter temperature was causing rapid deterioration of rectifier performance, resulting in temporary module isolation.

Mindshift’s algorithms detect these anomalies at scale. They create an intelligent pattern detection framework to enable proactive maintenance and minimize network blackouts. By leveraging advanced analytics and real-time monitoring, we ensure optimal performance of the power infrastructure, reducing operational costs and enhancing network reliability.

Conclusion

Mindshift Analytics offers a comprehensive solution for electrical energy management in telecom tower infrastructure. By providing continuous visibility, detecting anomalies, and enabling proactive maintenance, we help telecom operators maintain network uptime, reduce energy costs, and avoid SLA penalties. Our scalable machine learning algorithms ensure that even the most minute anomalies are detected, ensuring efficient and reliable network operations.