Thermal Gradient Sensor and Algorithm

How Automated Shading Reduces HVAC Energy Consumption

One of the most valuable potential benefits of automated shading is energy savings via reduced HVAC loads.  This is possible because adjustable window coverings can vary both the thermal transmittance (i.e. the U-value) and the Solar Heat Gain Coefficient (SHGC) of  a host window.

Minimizing HVAC loads with an adjustable shading device involves either fully opening or fully closing the device, depending on two variables: the direction of heat flow across the window, and the function being performed by the HVAC system (either heating or cooling):

Of course, these HVAC-driven shading settings can be inconsistent with a user’s preferred setting (e.g. to main privacy, provide natural illumination, or block glare), so HVAC-driven operation is typically appropriate only in unoccupied spaces or at nighttime with the lights off.  An especially apt application for HVAC-driven automated shading is in partially occupied office buildings.

The Problem with Conventional Automated Shading Technology

The conventional approach to implementing the shading-setting algorithm summarized in Figure 1 is to infer the direction of heat flow through the window on the basis of the outside temperature, and to determine the HVAC operating mode by “talking” to the HVAC system:

This conventional approach has significant limitations.

First, the actual exterior temperature outside the window can differ substantially from the local temperature obtained from a weather service (or from a building management system), which means that the inferred direction of heat flow across the window can be incorrect.

Second, the HVAC system might not have a communications interface that’s compatible with the shading system, or it might not be capable of proving mode information at all.  Further, even if the HVAC and shading systems are interoperable, the HVAC operating mode will not necessarily be consistent with actual heat load in the area in which the window is located.  For example, the HVAC system can be operating in the heating mode even though the windowed area is experiencing a net heat gain over time.

Finally, the need for connectivity to the internet and HVAC system can present cost and security issues in some applications, such as retrofit automated shading in office buildings.

Our Solution

Our patented solution is based on the fact that the actual direction of heat flow across a window is indicated by the sign of the temperature gradient normal to the window surface, and that the actual HVAC load in the windowed area is indicated by the sign of the integrated (time-averaged) temperature gradient.

Our approach involves obtaining the difference between two temperature values:

• A window-ish temperature is a temperature closer to the window temperature than to the room’s core temperature (at the same distance from the ceiling).  It can be obtained via an actual sensor at or near the window, or (less preferably) it can be assumed to be the same as the external temperature obtained from a weather server.
• A room-ish temperature is a temperature closer to the room’s core temperature than to window temperature (at the same distance from the ceiling).  It can be obtained via a dedicated sensor, or by applying a fixed offset to the actual room temperature (to account for thermal stratification) obtained via an interface to the HVAC system or a Smart Thermostat.  Alternatively, in many applications, an assumed room temperature can be used with minimum impact on performance.

The magnitude of the temperature gradient varies with proximity to the window surface.  Relatively close to the window surface, the gradient is large enough that a useful indication of the heat flow can be obtained with a thermal gradient sensor consisting of two precision temperature-sensing chips separated by as little as 2 cm.  Such a thermal gradient sensor can be integrated into a variety of products intended to be mounted at or near the window, such as our IntelliLux™ smart sensors:

Not only does such a sensor provide the most reliable indication of the actual direction of heat flow across the window, it also completely eliminates the need for temperature information from external sources.

Regardless of how the thermal gradient signal is obtained, the algorithm used to process the signal requires only a tiny fraction of the processing capacity of even the least expensive microcontrollers.  It can therefore be implemented by virtually any smart shading product, as well as by a voice assistant, home-automation hub, building-management system, or cloud server.

Thus, in addition to saving more HVAC energy than the conventional approach of Figure 2, our approach is potentially far simpler to implement.

Applications

Our thermal-gradient sensing technology is potentially applicable to any automated shading system intended to reduce HVAC energy consumption, whether in residential or commercial buildings.  For example, one potential application is to reduce HVAC loads in temporarily unoccupied spaces in partially occupied office buildings.

Our technology can be implemented using a combination of hardware and algorithms, or purely via algorithms in conjunction with existing hardware.  This enables it to be inserted into a wide variety of product types.  Three exemplar applications are shown in Figure 5:

Smart Shade

The first application shown in Figure 5 is a window-mounted motorized shading product.  Many such products (like our IntelliBlind™ reference design) will be amenable to integration of a thermal gradient sensor like the one previously described.  However, for illustrative purposes, Figure 5 assumes a Smart Shade which lacks a gradient sensor.

Instead, the product’s microcontroller calculates the thermal gradient  by obtaining the room temperature wirelessly from a Smart Thermostat, adding an offset to account for temperature stratification, and subtracting the result from the output of a single temperature sensor embedded in the product.  The microcontroller time-integrates the gradient over a multiple of 24 hours to obtain the daily average gradient, and actuates the shading (per Figure 3) based on the signs (not the magnitudes) of the instantaneous and integrated gradients.

While this example assumes the presence of an interoperable Smart Thermostat, Matter-compatible Smart Thermostats are still relatively rare.  Alternatively, an assumed room temperature could be used instead of an actual room temperature (again adjusted for stratification).  This eliminates the need for connectivity to the thermostat (and the attendant interoperability issues) with some loss in the achievable energy savings.

Smart Sensor

The second application shown in Figure 5 is a wireless smart sensor. Such a sensor can be mounted directly on a window (like our IntelliLux™ Window Sensor previously shown in Figure 4), on the headrail of a window covering (like our IntelliLux™ Headrail Sensor previously shown in Figure 4), or on any surface near a window.  Due to proximity to the window, a useful thermal gradient signal can be obtained with relatively closely-spaced sensors, so such a product can integrate a thermal-gradient sensor of the type described above.  Like our IntelliLux™ sensors, such a product can also include a wireless microcontroller which implements a low-power RF protocol (e.g. Zigbee, Thread, or BLE), implements the algorithm of Figure 3, and communicates HVAC-optimized shading settings to a conventional Smart Shade product.

Ideally, like our IntelliLux™ sensors, such a wireless sensor would also incorporate our multi-spectral glare sensor for advanced control, as well as a small photovoltaic panel, supercapacitor, and energy-harvesting controller to eliminate the need for batteries.  Optionally, the equivalent functionality could be built into a larger window-mounted photovoltaic panel intended to power a motorized window covering.

Software Upgrade to Conventional Automated Shading System

The third application shown in Figure 5 is a software-only implementation of the approach of Figure 5; it obtains the thermal gradient as the difference between the room temperature (as obtained from a connected smart thermostat) and the outside temperature (as obtained from a connected weather server), implements in a control app running on a local or remote server, and sends shading commands to a conventional Smart Shade product.