Optimizing Power System Monitoring and Control

Bhavesh Patel is Director of Marketing and Customer Support at ASCO Power Technologies, Florham Park, NJ, a business of Emerson Network Power.

Optimizing power system monitoring and control at a data center can go a long way to helping ensure power reliability – a universal goal of data center management.

There are four prevalent sophisticated technologies that can provide key information about the status of onsite power: two legacy systems – Building Management Systems (BMS), and Supervisory Control and Data Acquisition (SCADA) systems, and two newer systems – Data Center Infrastructure Management (DCIM) and Critical Power Management Systems (CPMS).

The first three aim to monitor and control an entire facility or campus, including critical power. The fourth is designed to control only critical power generation and distribution systems. Each system has its particular capabilities, strengths, and limitations which should be considered when evaluating options.

A BMS is a computer based control system that provides integrated management of the control and monitoring of a building’s core electrical and mechanical equipment. Installed in new buildings or in renovations, a BMS typically covers heating, ventilation, and HVAC systems, and often includes lighting, security, fire alarm systems, plumbing and water monitoring. A BMS also tracks and schedules building maintenance. A system can use proprietary controls or, as is increasingly more common, open standard controls. While it does not have the scope of capabilities of more sophisticated systems and does not specifically address power reliability, a BMS can provide early detection of problems with electrical power via basic alarm and control notification and may include remote as well as onsite alarm monitoring.

Because of the narrow bandwidth at which it operates, a BMS has limited capabilities with respect to high speed monitoring and control. The speed and bandwidth at which data transfer occurs between critical power equipment components could incapacitate most BMSs. Power quality data such as transient harmonic displays or wave form capture are possible examples of that.

Pro: Popular at standalone, single function buildings, including data centers.

Con: May not distinguish between critical and non-critical monitoring and does not necessarily include software to manage mission critical operations and processes.

A SCADA system is designed to monitor and control business operations and processes via sensors placed at multiple sites at various locations which are monitored from a single centralized location, utilizing coded signals over communications channels. Most are PLC-based. Aiming to improve efficiency and operational reliability, lower costs, and enhance worker safety, these systems are particularly suited for enterprises across large distances or occupying multiple facilities under one management, such as a data center with multiple sites.

Today’s sophisticated SCADAs include a computer and open (off-the-shelf) system architecture that acquire data from, and send commands to, monitored equipment, a human-machine interface, usually a computer monitor screen, a networked communication infrastructure, sensors and control relays, remote terminals units (RTU) and programmable logic controllers (PLC). Functions include: alarm handling, trending, diagnostics, maintenance scheduling, logistics management, detailed schematics for a particular sensor or machine, and expert-system troubleshooting guides.

Pro: Can provide equipment status to remote Internet-connected mobile devices including tablets and smartphones, which can expedite alarm notification to key personnel, and improve alarm handling and response time.

Con: For alarm handling, a cascade of quick alarm events may ‘hide’ the underlying cause. Standard protocols and Internet accessibility of networked SCADA systems make them susceptible to remote natural or human-made electromagnetic pulse (EMP) attack. Not the best choice when reliable power is critical.

The newer systems are DCIMs and CPMSs.

A DCIM, which by design is focused on data centers, can provide more of a holistic view of a data center’s IT and facilities infrastructure. DCIMs, which can handle the lightning speed of data generated for analytics, have specialized capabilities to monitor, measure, and manage both facility infrastructure components and IT equipment at a data center, especially larger ones, using data derived from SNMP, Modbus, or BACnet. A DCIM can not only monitor the facility infrastructure but can also use powerful analytics to provide “intelligence” and reporting for decision making that can improve efficiency of operation. That said, it cannot do everything a BMS does and may be utilized as complementary to a BMS.

Pro: A sophisticated system can improve uptime and efficient capacity planning and management, and provide valuable business analytics and deeper process and change management.

Con: Like with a BMS, DCIMs need to be sophisticated enough to import volumes of operational data from power controls in order to effectively monitor and control critical power systems. However, the majority of that data transfer (such as transient harmonic displays or wave form capture) occurs at speeds and bandwidths that may incapacitate many DCIM systems. Given that there are no standardized platforms or protocols like Modbus or BACnet at that level of technology, to benefit such sophisticated analysis, IT would need to rely on vendor proprietary software.

A CPMS, which is focused on any type of facility, is designed to monitor, control and analyze equipment for power generation and distribution, both for normal power and for emergency/back up power. It is an excellent choice when power reliability is crucial, 24/7, as is the case in a data center serving a varied clientele. Usually, a CPMS is set up to monitor all data from the point electricity enters the facility from the utility main throughout the entire facility. And for the emergency/backup power system, a CPMS generally oversees gen-sets, transfer switches, paralleling control switchgear, uninterruptible power systems, circuit breakers, bus bar, and other critical power distribution equipment.

Today’s full featured CPMSs have wide bandwidth and operate at extremely high speed and can cache or share large amounts of data from one device to another without disrupting building functions.
A CPMS will monitor normal and emergency voltages and frequency, indicate transfer switch position, source availability, normal and emergency voltage and frequency, current, power, and power factor; and display transfer switch event logs, time-delay settings, rating and identification. It will also facilitate critical power system load management, bus bar optimization, testing, maintenance, reporting, trending and analytics, all with an aim of ensuring power reliability during surges, sags and outages.

In addition, CPMSs often have functions and alarms integrated into the data center’s building management system (BMS). For example, CPMS could send automatic alerts on system operation via email, text, or selected system alarms to the BMS. High-end CPMSs feature integrated devices communicating on a dedicated network.

Compared to monitoring and control capabilities of BMSs, SCADAs, and DCIMs which can address much more than electrical power components, CPMS monitoring and control capabilities are more narrowly and sharply focused and are dedicated to managing critical power generation and distribution power. While they typically work in concert with a BMS, SCADA, or DCIM, CPMSs provide the necessary sophistication, speed, and analytics specific to power generation and distribution.

The ability of a CPMS to operate at very extremely high speed and share or cache tremendous amounts of data between devices without disrupting building functions is advantageous when doing post-event troubleshooting or forensics, which benefits from fast and accurate time marks to track down where and when things went wrong. The analytics benefit from a recording scale fast enough to identify, with time marks, precisely what started the event within a very short time frame (often milliseconds). For example, the analytics can look into why the data center lost a particular breaker that tripped the PDU and caused a chain of events that caused a switchover to the UPS and help determine whether the precipitating event was an electrical spike, a floating ground or a short.

Typically, CPMSs have the scalability to accommodate expansions and upgrades of a data center facility or campus as the enterprise grows and if/when the business model changes.

Pro: Can take advantage of continuous monitoring, which utilizes intelligent controls and sensors along with testing and retesting to make sure facility systems operate as designed and constructed, not only initially but also over time as equipment ages, with the aim of optimizing owner cost and occupant comfort.

Con: Does not have some of the IT operational details that a DCIM can offer a data center. Furthermore, a CPMS generates a lot of information which, without good data visualization capability built into the system, could become overwhelming and even unproductive.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.