SmarTHERM Thermostats (HVAC)


“Thanks to its unique Artificial Intelligence architecture,
the SmarTHERM sets apart from the competition mainly in 4 domains”


What is the SmarTHERM?

The SmarTHERM is an Intelligent Heating and A/C Control system for residential homes. The system is a PC based system and is founded on Artificial Intelligence where its name comes from (intelligent (“Smart“) and thermostat (“Therm“) ). Thanks to its Artificial Intelligence infrastructure, the system adapts itself by learning experience to each house and environment where it is installed. It is not driven by classic hard coded pre-programmed rules set in a general manner for any houses and any conditions. Differently, the system is coded to learn and adapt. To this end, the system is equipped among other things with an integrated Weather Station. 

The project came alive at the beginning of the nineties and ran in lab and in test homes for more than 25 years. Its design and tuning process has been rather long and laborious since it has to be tested in a real live context, especially during the few months where the rough winter conditions prevail in winter at north of the 49th parallel (Canada).

The SmarTHERM pursues 2 specific objectives which are generally opposite in matter of indoor temperature control:

In cold temp, many complex phenomena combine to contribute to the heat loss of the house. By the same token, these phenomena make it difficult for any heating system to reach and maintain a given degree of heat inside the house. The most contributing factors that we can think of are:

Pre-Heating Curve Graph

Pre-Heating Curve Graph


  • outdoor temperature (heat loss increases exponentially with the lowering of the outdoor temperature);
  • the speed and direction of winds (the wind chill factor imposes on living bodies an effective temperature many degrees lower than the nominal temperature);
  • the degree or the lack of sunshine (the sunshine has for consequence to increase significantly the temperature of bodies and objects that are exposed to, particularly if these ones are in a closed space);
  • the wanted indoor temperature (the difficulty to reach and maintain every additional degree of indoor temperature inside the house increases in an exponential way) (see graph on the right);
  • the degree of insulation of the house (the quality of the insulating material, the total amount of window’s surfaces, the orientation and location of the house, etc.);
  • the living patterns or habits of the occupants (their degree of tolerance to cold, the frequency of going in and out, openings and closing of windows and/or doors, opening and closing of curtains or blinds, etc.).

The SmarTHERM is an “auto-adaptive“system which, thanks to its various sensors and to the artificial intelligence at the heart of the design of its software, strives to take into account all the major climatic and environmental factors that affect the thermal effort required to reach and maintain the level of comfort expected inside a house at cold temperatures, and this at the most economical way.
read more / less ...^

In its basic version, it comes with an indoor temperature sensor and an outdoor temperature sensor.
In its most advanced version, it comes also with a wind sensor (anemometer) and a sunlight sensor.

The SmarTHERM is made up of 4 basic elements:

  • a Weather Station which gathers all the outdoors sensors (temperature, wind, sunshine)
  • an indoor electronic thermometer (sensor)
  • a PC electronic card (first level interface with the various sensors and the heating hardware)
  • a PC software to perform the thermostatic control functions and display the meteorological data (user interface)

The system continuously ‘monitors’ the climatic and environmental conditions through the electronic card which interacts with the various sensors. The climatic and environmental data observed are recorded regularly at any hour of the day and night in the computer database.

In the normal course of its operation, the software makes use of its accumulated knowledge base to predict the calorimetric behavior of the residence in which it is installed and this, at any time where a calorific effort is required (maintenance of thermal comfort) or about to be required (planning of an anticipated calorific effort to provide in order to achieve an imminent programmed temperature increase). At time deemed appropriate (following the analysis of the various data), the software issues the appropriate commands to the heating system, once again through the electronic card which serves also as an interface with the electric or oil or gas furnace, a heat pump or others systems.


At the end of each given effort (heating cycle), the system evaluates the performance reached compared to its previous forecast of the required effort and if necessary, automatically adjusts its forcast criteria according to the variation (deviation) or of the absence of variation (optimal performance) observed, and this in order to improve its learning process (auto-training). Thus, no matter in which home it is installed, the system adapts continuously to its environment and improves its performance continually. Even if important modifications are made to the environment (expansion of the house, improvement of the insulation, upgrade of the heating system, etc), the system will re-adapt itself with time. No need to reprogram the system, nor to change any parameters or settings. It is what we mean by a “self-adapting” system.

Fundamental characteristics of the SmarTHERM?

How does the SmarTHERM succeed in maximizing the heating energy savings while maximizing the home comfort?

Basically, in 4 ways:


A) Optimization of the time at which the heating system will be restarted following a stop or a lowering of the programmed temperature (‘SRAS (Smart Restart After a STOP)’ system)

Contrary to the majority of programmable thermostats for which the user must determine himself the time at which the heating system will be restarted following a programmed stop (for example in the morning following a lowering of the programmed temperature during the night or at the end of the day, before the return to the home), the SmarTHERM, thanks to its ‘SRAS‘ system, determines itself the optimal time of restarting the heating system, based on a set of surrounding conditions (indoor temperature, outdoor temperature, wind speed, sunshine percentage, living habits of the occupants of the house, the power of the heating system, the quality of the thermal insulation of the house, etc).

With the SmarTHERM, the user does not enter his programming schedule according to the time at which he wants the system to restart following a programmed stop or a lowering of the temperature. He enters the time at which the desired room temperature must be in force (“target time”) (for example 72°F at 8:00 AM). The ‘SRAS‘ system will determine itself, following the analysis of the climatic and environmental conditions, the exact time at which it must restart the heating system to make sure that the room temperature of the house is at the wanted temperature, at the time fixed by the user (72°F at 8:00 AM). For the same target time, the heating system restarting time computed by the SmarTHERM will vary greatly from one day to the next, especially in winter when the variations of temperatures and weather conditions (wind, sun, etc) vary greatly and frequently. Thus, for a programming of 72°F at 8:00 AM for example, the restarting time of the heating system could just be 10 or 15 minutes before the target time (7h45 AM) in autumn or spring, whereas it could be 4:00 AM and even 3:00 AM in winter when outdoor temperatures are at -5°F or -15°F.

The sudden changes of the weather conditions are often impossible to predict and even more difficult to ‘monitor’ by the occupant of the house in such a way that he can continuously manually modify its programming, especially when this one is in its sleep or outside of his home, at work or elsewhere. This way, the programming of conventional programmable thermostat is rarely optimal and rarely allows the greatest energy savings and the greatest level of comfort possible.

Pre-Heating Curve (-3.2'C, 6.7'C deg., 3h34m)

Pre-Heating Curve (-3.2’C, 6.7’C deg., 3h34m)

read more / less ...^
Indeed, in the example used above, assuming that in winter the occupant sets his conventional programmable thermostat at 72°F at 6:00 AM with the aim of having a comfortable temperature (72°F) at the time of his awakening at 8:00 AM, its goal will rarely be achieved throughout the course of the year. Indeed, either the desired temperature of 72°F will have been reached several minutes before 8:00 AM, in which case he will have spent energy needlessly, or it will have been reached several minutes and even hours later, in which case his comfort will suffer from it. The same phenomenon is likely to be repeated at the end of the day if the house is unoccupied during the day and that a lowering of the temperature (setback) was programmed.

It is that exact problematic that the SmarTHERM is built to solve. Achieve the maximum savings while assuring the highest level of comfort for the occupants. With the SmarTHERM, if the user sets it to have 72°F at 8h00 AM, he can almost be certain that at his awakening at 8h00 AM, the temp will be at 72°F in the house, not before, not after(1)

(The graph above shows a real time prediction of the pre-heating curve and time estimated by the ‘SRAS‘ system on an extreme cold day of -22°F with a ‘setback‘ of 12°F and a pre-heating time of 3h34min. to reach the requested indoor temp of 72°F at 8h00 AM. The GREEN curve is the prior curve estimated by the ‘SRAS‘ system, while the RED curve is the true curve traced by the heating system in real life. As one can see by enlarging the image, the 2 curves almost perfectly lie one on the other, which is a clear demonstration of the precision of the ‘SRAS‘ system in this case).


The ability of the ‘SRAS‘ system to predict with very great precision the pre-heating curve and the required pre-heating time in any case is largely attributable to its on-site integrated Weather Station. Weather StationNo so-called “intelligent” (Smart) thermostat can reach the precision of the SmarTHERM in this field without having a Weather Station attached to the house to heat.

The SmarTHERM continuously monitors in real time (many times by second) the weather conditions to which the house is exposed (cold, wind, sun) and continuously recalculates the heating system restarting time accordingly. It does not rely on imprecise weather forecasts or data coming from regional weather stations updated hourly (or more…), often located many miles away and applying to a very wide area. As a matter of fact, 2 similar houses located in the same neighbourhood will require very different pre-heating time based on their spatial positioning (facing north, facing south, etc.), their immediate environment (isolated, surrounded by trees, by large buildings, etc…), their thermal efficiency (insulation, fenestration, etc…). The SmarTHERM takes all these factors and many more into account to determine the optimal heating system restarting time after a stop or after a programmed indoor temp lowering (setback).


B) Optimization of the degree at which the indoor temperature can be lowered at night or during the day when the house is unoccupied (‘SMART SetBack’ system)

One of the basic problems with home temperature lowering at night or during the day when the house is unoccupied, it is to determine the optimal threshold at which the temperature can be lowered to realize the maximum energy saving, while not compromising the comfort of the occupants.

Basically for the same reasons that those set out previously, it proves extremely difficult to determine this optimal threshold. For this reason most owners of conventional programmable thermostats are satisfied with a fixed temperature setback of a few degrees (typically 5°F or 10°F, rarely more), without knowing if this lowering is the most economic for the circumstances or if it will allow an effective and economic increase of the temperature at wake up time or at home coming time at the end of the day.

Temp Setback Graph

Temp Setback Graph

The ‘SMART SetBack‘ system of the SmarTHERM has the required intelligence to determine this optimal and most economic temperature lowering threshold. To benefit from this unique feature, suffice to program a really low temperature (for example 50°F or 40°F), and to let the SmarTHERM manage itself the optimal lowering threshold. In practice, the temperature will probably never reach the ‘floor‘ programmed threshold (unless in the case of a particularly prolonged absence). Still there, depending on the climatic and environmental conditions, the optimal temperature lowering threshold could happen to be at 65°F, 60°F or even 55°F for a given day and be 3°F or 5°F higher or lower the day after. In spite of the optimal threshold determined by the system, the user has always the option to fix himself a ‘floor‘ threshold (for instance 66.5°F) that the system will respect, even if it is not the most economic.

For those who would doubt of the economic efficiency of the inherent temperature lowering logic of the SmarTHERM, this one is equipped with a monitor of the effectiveness of the heating cycles performance visible at any time. So, the user can see at any time the energy saving realized (in Kilowatts and in money($)).


C) Keeping of a constant temperature (‘SMART COMFORT’ system)

Regarding the room temperature keeping, the basic problem for most thermostats is to maintain the temperature as constant and as stable as possible.

It is known that depending on their quality level, all thermostats have a certain gap between their point of contact and their point of release. This variation commonly called the ‘swing’ varies greatly from one thermostat to another. It can easily be of 2°F or more for mechanical thermostats and of 0.2°F or more for the best electronic thermostats.

Moreover, it is also known that each time a heating system is started, a certain inertia makes so that the temperature continues to go down for several minutes (3 to 5 min.) before starting to raise. In the case of a forced-air heating system, this phenomenon of temperature fall is amplified by the feeling of cold caused by the circulation of cold air pushed by the ventilation system before this pushed air reaches a temperature slightly higher than the temp of the room (not to mention that most furnaces delay the starting of their fan until each and every one of their heating elements (sometimes 3, often 5 or more), started sequentially and with a delay (many seconds), has reached their maximum heat). A similar but reversed phenomenon of inertia makes so that the temperature continues to go up during several minutes after the thermostat stopped the heating system. The bottom line of these phenomena of inertia added to the lack of precision of the majority of thermostats can easily result in effective variations of 2°F to 4°F and sometimes even more.

For its part, the SmarTHERM has a precision of 0.01°C (0.02°F). Moreover, to counter the problem of temperature variations inherent to conventional thermostats as well as electronic ones, thanks to the Artificial Intelligence on which it is build, the SmarTHERM has a totally unique technology called SMART COMFORT. By the means of this technology, the system continually analyzes a set of variables with the aim of determining the optimal starting and stopping points of the heating system causing the less ambient temperature variations possible. Thus, instead of operating the heating at fixed points like 72.5°F (starting and stopping) for example, the SmarTHERM does it at points which vary constantly according to the circumstances and which are reversed compared to the way of functioning of conventional thermostats (starting at 72.62°F for example and stopping at 72.38°F). By doing so, the system makes sure that the temperature variations (swing) will remain inside these borders, taking into account downwards inertia following starting of the heating system and of upwards inertia following its stopping. If the surrounding conditions make so that the room temperature gets out of one or the other of these two borders or even both, the lower or higher threshold or both will automatically be readjusted dynamically in order to maintain the stability of the room temperature at its best. This process is controlled by a constant ‘monitoring‘ and ‘adjusting‘ to surrounding conditions from the SMART COMFORT system.

So, thanks to the SMART COMFORT technology, the ambient temperature variations are maintained to a practically imperceptible level by most of the occupants. Moreover, by eliminating the constant variations of the room temperature, one can set the temperature some tenths of degrees lower, which result once again in more energy savings.

The SMART COMFORT system has 2 other basic functions which aim at maintaining an optimal comfort level for the occupants and to maximizing energy saving.
read more / less ...^

On the one hand, in the case of 2 “stages” systems, the SmarTHERM tries to maintain the desired temperature by the most economical way possible. Thus, no matter the outdoor temperature, the SmarTHERM will try to take advantage as much as it can of the energy saver mode (stage 1) to maintain the room temperature. If it feels that it will not be able to succeed in maintaining the temperature this way, it will call on the second stage at the time that it will judge necessary to allow it to maintain or reach the desired room temperature, and then it will return to the basic stage as soon as it will have received the necessary boost. This call to the second stage may occur several times during the same heating cycle, if required.

On the other hand, we know that in particularly cold or windy weather, the occupants of the house feel an additional cold sensation for the same set degree of temperature.  Many feel the need to raise the thermostat temperature to negate this increased cold sensation.  To cope with this problem, the SMART COMFORT system has a mechanism enabling it to adjust the room temperature according to the prevailing weather conditions outside. Thus, in particularly cold or windy weather, no need to raise the desired temperature on the thermostat like with conventional thermostats, the system will automatically do it for you. Same thing in particularly mild or sunny weather, the system will lower the room temperature automatically in order to maintain an acceptable level of comfort and save energy at the same time. The increase or decrease of room temperature due to this compensation mechanism can be of some tenths of degree only, but is sufficient to maintain a constant comfort level day by day, no matter the bad weather or the sudden warmings of the outdoor temperature. This way, no more need to play constantly with the thermostats and to undergo the variations of comfort which often result from it and the increased costs of energy consumption which also result from it almost inevitably.


D) Great flexibility regarding the programming and parameterization (settings)

Since the SmarTHERM is a computer based system, its programming flexibility is almost unlimited compared to the one of conventional programmable thermostats.

One can enter several programs per day, per week, per month and even per year. The user can create various program’s scenarios or profiles and call them by their names (e.g.: Holiday, Weekend, Night, Day 2, Day 4, etc). One can also program in advance all the planned absences or holidays of the year, including the repetitive annual leaves (e.g.: **/12/25, 14/03/**, etc).

In short, the user has the full capability to create his own programming and to give symbolic names to hours, days, degrees of temperature, programs and other objects in order to facilitate their update system wide by making the change at a single place (the symbol).

To adapt to the preferences of the users, the SmarTHERM has a very great flexibility of parameterization. For instance, it is possible to set a lowering (setback) of 1° if it is more than +45°F outside and the sunshine strength is more than 70%. By the same way, it is possible to set a rise (set forward) of 0.5° if it is less than -5°F and the wind blows at more than 30 mi/hr. One can also add or withdraw a correction factor to the indoor and outdoor temperature to compensate for a bad localization of the temperature sensors, for example exposed to sun rays in front of a window or on a hot roof, close to a heat source, a door or a cold wall. The user can also choose to display the temperature in °F or °C, just as to display the wind speed in Mi/hr or Km/hr.




From 15% to 33% of heating costs savings depending upon :

  • power and type of heating system used
  • characteristics of the house
  • geographical localization of the house
  • spatial orientation of the house
  • periods of absences – temperature lowerings
  • life habits of the occupants
  • selected temperature

Normally, the system should pay for itself over the first year of use or within a 2 year period in the worst case scenario.


  • uniformity of the room temperature
  • adaptability to the environment conditions
  • anticipation of the heating needs



  • PC able to run Windows 98 or later
  • 1 Meg. of random access memory (RAM)
  • 1 Hard Disk with at least 1 Meg. of available space
  • 1 8 bits ISA/EISA expansion slot or
    1 16 bits PCI expansion slot (recommended)


  • HVAC type heating systems (electric furnace, oil, gas, etc)(2)
  • Wall mounted Heat Pump or integrated into the heating systems(3)


(1) Taking account a margin of error of few minutes varying from an installation to another.
(2) Currently tested only on electric furnace systems.
(3) Actually heat pumps are controlled in heating mode only.  A/C control coming soon…

For any additional information or comment, do not hesitate to contact us at:

Blue Line


Like us on Facebook:


To make sure you don't miss any post from us,
Click on the Facebook button below to go to our FB Page,
once there, Click on the "Follow" arrow, and then,
Check the "See First" option and
the "All Posts" option (under the Notifications parameter).


For all comments regarding the administration of this site: