How does sensor accuracy impact building energy management?

Buildings contribute to about 32 percent of global energy consumption, much higher than transportation or any other category, according to the International Energy Agency. In developed economies such as the United States, this figure is much higher, reaching up to 48 percent in 2013.

In an increasingly energy-deficient world, energy conservation in operating buildings has become a key focus area. Building automation and control systems help with this goal. However, their effectiveness relies heavily upon the accuracy of sensing devices to provide inputs to the control system. These control systems make decisions around energy optimization in addition to comfort, convenience and safety.

Over the years, the level of automation steadily has risen. However, the sophistication of software and controls hasn’t always achieved the desired energy management goals due to other limitations, such as the accuracy of sensors.

Control systems maintain a proper building environment in a cost-effective manner through four variables — temperature, humidity, pressure and ventilation. Usually the control of the built environment is executed by starting, stopping or regulating the heating ventilation air conditioning system through a combination of hardware and software actions. The control process typically comprises three steps:

1. Measuring an environmental variable such as temperature or humidity at various points and collect the data;

2. Processing the data with other environmental and contextual information; and

3. Causing or triggering a control action and, if desired, providing feedback about the impact to the control system.

These steps are realized through sensors, controllers and software systems. If the sensors are not accurate, other actions will carry through the flaws, potentially even amplify them.

Consider a typical commercial office building of 100,000 square feet. The HVAC system comprises many parts. The chillers in the central plant make up the high side of the system and the air handling units (AHU) on various floors of the building make up the low side of the system. Many other devices and components are associated, such as pumps and cooling towers, but let’s limit this example to the chillers and AHUs. Sensors usually monitor the temperature of chilled water entering and leaving the chillers.

If the two temperature sensors in this case are off by 0.5 degrees Celsius each, it could lead to an efficiency error of about 17 percent in the chiller's performance. In the case of this building, we might end up paying about $29,000 excess in electricity charges each year. Two years of such losses almost equals the cost of the chiller itself. On the low side, a similar 0.5-degree calibration difference at the various points supply air temperature is measured in the AHUs could lead up to a 72 percent error in reported efficiency performance.

For every type of sensor, if there is an accuracy issue, you will see similar deviations between actual and reported performance, which completely will mislead the control system and thereby cause significant energy efficiency loss and other operational challenges.

For building managers, many day-to-day issues reported by occupants can be traced to sensor faults. This table captures some popular faults and causes.

In recent times, technology advancements have made automated fault detection and diagnostics popular as a tool to optimize energy consumption and building performance. The sensor output information is very critical for FDD. Sensor malfunction is amplified through the feedback loop and may trigger unrealistic fault alarms.

In addition to energy and financial losses, faulty sensors leading to incorrect control parameters also lead to deviations in building performance indicators such as energy benchmarking, energy forecasting, measurement and verification functions of guaranteed performance contracts or other statutory reporting. Sustainability programs, corporations or nations adversely can be affected by something as simple as faulty sensors.

Failures in sensing devices are manifested in many ways:

• Permanent: Sensor output takes on an incorrect value permanently.
• Offset: Output consists of correct sensor values offset by a constant amount; this applies to both analog and digital sensors.
• Erratic: Output consists of correct sensor values offset by varying amounts in time.
• Intermittent: Occurs only occasionally, disrupting what is otherwise a valid sensor output.
• Transient: Usually due to failure of electronics associated with the sensor (too large current or voltage, interferences, such as lightning, radio-interference).

Sensor accuracy increasingly is becoming a topic of concern for various stakeholders in the sustainability ecosystem. Countries such as Singapore are addressing this issue through regulatory actions implemented through its Building Construction Authority’s Green Mark program, which specify limits for sensor accuracy in various building equipment and systems. Many progressive architects and designers also incorporate sensor accuracy considerations in their building design. We need more education and collective action among the other players in this sustainability ecosystem to make a discernible impact.

In an upcoming story, I’ll explore various methods to identify sensor accuracy issues.