FR3001068A1 - Method for modeling thermal behavior of human being in meeting room of enterprise building, involves modeling living being present in area of building, and cutting model to consider influence of sub-models according to periods of day - Google Patents

Abstract

The method involves modeling a human being present in an area of a building as a thermal source (alpha 2R). A model (63) of thermal behavior of the area is obtained, where the model includes a first sub-model characterizing thermal inertia of the area, a second sub-model characterizing the thermal contribution of a heating or air-conditioning device present in the area, and a third sub-model characterizing the thermal contribution of a human being present in the area. The model is temporally cut to consider influence of the sub-models according to periods of a day. An independent claim is also included for a thermal control system for an area of a building.

Description

FIELD OF THE INVENTION The present application relates generally to thermal regulation and, more particularly, to a thermal control method and system based on modeling the thermal behavior of a building. BACKGROUND OF THE PRIOR ART Many heating control systems are known which take into account various control data including the indoor temperature, the outdoor temperature, the individual room temperature of the building. For example, EP-A-0249531 discloses a central heating control method and apparatus which combines a model characterizing the heating distribution network as a heat exchanger between the coolant stream and the building to be heated and a model characterizing the building as space. This method further provides self-learning characteristics of these models. Existing systems remain imperfect. In particular, they fail to effectively predict the thermal needs of building areas in order to modulate heating or air conditioning systems in advance.

B12156 2 Abstract There is a need for a thermal regulation system that takes into account different predictive factors in order to optimize energy consumption.

There is also a need for a model of thermal behavior of a building exploitable by such a control system. Thus, an embodiment provides a method for modeling the thermal behavior of at least one area of a building, wherein a living being present in said area is modeled as a thermal source. According to one embodiment, a model of thermal behavior of said zone comprises several submodels among which at least: a first sub-model characterizing the thermal inertia of the zone; a second sub-model characterizing the heat supply of at least one heating or cooling device present in the zone; and a third sub-model characterizing the heat input of a human being present in the area. According to one embodiment, a fourth sub-model characterizes the thermal contribution by solar radiation. According to one embodiment, at least the first sub-model is also a function of the presence of a living being in the zone, in that it temporarily modifies the thermal inertia of the zone. According to one embodiment, the model is temporally divided to take into account the influence of the different sub-models according to the periods of the day. According to one embodiment, the model is obtained in a learning phase during which measurements from sensors among which at least one indoor temperature sensor, an outdoor temperature sensor, a human presence detector and a sensor. Energy consumption sensor are performed periodically. According to one embodiment, the periodicity of the measurements is between 5 minutes and 2 hours.

According to one embodiment, the method is reproduced for several zones of the building. There is also provided a method for planning a thermal regulation of a zone of a building, using a model obtained by implementing the modeling method 10 to simulate the thermal behavior of the zone. According to one embodiment, the planning takes into account a provisional schedule of occupation of the area by living beings. According to one embodiment, the planning takes into account weather forecasts. According to one embodiment, the model is updated over time. According to one embodiment, the planning results in a heating plan operated by a heating system. There is also provided a heat regulation system of at least one zone of a building, comprising a computer tool programmed for implementing the planning method. BRIEF DESCRIPTION OF THE DRAWINGS These and other features and advantages will be set forth in detail in the following description of particular embodiments in a non-limiting manner with reference to the accompanying figures in which: FIG. schematic of areas of a building equipped with an embodiment of a control system; Figure 2 is a block diagram illustrating an embodiment of a thermal control system; FIG. 3 is a simplified block diagram illustrating an embodiment of the method for learning and modeling the thermal behavior of an area of a building; Figs. 4A and 4B are equivalent diagrams of submodels of one embodiment of the modeling method; and Figure 5 illustrates in more detail one embodiment of the thermal behavior modeling method. DETAILED DESCRIPTION For the sake of clarity, the same elements have been designated with the same references in the various figures. In addition, only the steps and elements useful for understanding the embodiments that will be described have been detailed. In particular, the heating, air conditioning, and energy recovery devices have not been detailed, the embodiments described being compatible with the use of existing devices. The inventors have found that one of the origins of the inaccuracy of the existing systems stems from the fact that the occupation of the premises was not adequately taken into account. In particular, it sometimes happens that a room occupancy parameter is taken into account to provide that this room 25 will be occupied for a certain period and heat it in advance, then stop the heating when its occupation ceases. but it does not go much beyond that. However, the thermal behavior of a room is sensitive to the presence of living beings. Moreover, the human presence can temporarily modify the physical parameters of the building (for example, an open window reduces the insulation of a building), and therefore its thermal inertia. Thus, the inventors plan to consider the human presence as a thermal source and to take it into account in the model of thermal behavior used for the B12156 regulation. To account for this thermal source, it is also expected a temporal division of the thermal behavior of the building area and a prediction of the human presence in terms of time and importance. Figure 1 shows a simplified example of a building equipped with an embodiment of a control system. In this arbitrary example, it assumes a building 1 with 4 pieces 11, 12, 13, 14 and a corridor 15, which represents 5 zones. Each zone 11 to 15 comprises at least one door 111, 112, 121, 131, 141, 142 and 151. All or part of the zones comprises one or more windows 115, 125, 135, 145. To illustrate the operation of the control method thermal, it is assumed that each zone is equipped with one or more heaters 21 to 25, for example individual radiators or connected to a central heating. For simplicity, one will take for example a thermal regulation based on a heating system, but all that will be described also applies to an air conditioning system. More generally, an area may include one or more rooms. Zone division depends, among other things, on the position of the heaters and the accuracy required for control. Each zone 11 to 15 is furthermore equipped with several sensors. Among these sensors, there is provided for each zone at least one temperature sensor 31 (T) and at least one presence detector 32 (P). Preferably, other sensors equip the different areas, for example, radiation sensors measuring the solar radiation on the windows, unrepresented moisture sensors, etc. The system also takes into account the energy consumption of the building and, preferably, each zone or heating source equipping each zone, for example, by means of sensors 33 (C). Furthermore, the system receives other information such as the outside temperature, either from a weather station, or from one or more sensors 34 (ET) outside the building. It is planned to assign to each zone a model of thermal behavior. This model is then used to plan the regulation of the temperature in the area concerned. Preferably, the model is obtained by learning the thermal behavior of the zone. It is further adapted over time. Figure 2 is a schematic representation of the elements of a control system. Such a system, symbolized by a block 5 (SYSTEM), comprises one or more computer devices for operating the measurements made by the sensors and for interpreting these measurements for the purpose of establishing and adapting the model or models. of thermal behavior of the building zones by learning, and planning of the necessary heating (heating plan). For example, the computer tools constituting the system 5 comprise different input / output ports wired or wirelessly connected to the various temperature sensors 31, presence 32, consumption 33, outside temperature 34 and radiation 35. Other input / output ports of the system 5 serve to control the heat generating elements or control elements that drive heat generating elements, symbolized by blocks 21 to 25 in FIG. of the human activity, the system 5 also receives (link 51) information on the expected occupation of the different zones. In addition, the system preferably receives weather forecast information. This information comes, for example, from a computer server 52 to which different equipment is connected (for example, desktop computers 53, laptops 54, smart phones or digital tablets 55, etc.). The server 52 is also preferably connected to the internet 56 (WEB) so as to obtain information on the expected weather conditions. Typically in an electronic calendar management system, different users enter their electronic diaries or their room occupancy requirements (for example, to plan meeting room occupations in a company). This information is therefore centrally available at the server 52. The representation of Figure 2 is schematic in that the system 5 will, in most cases, in practice integrated into a computer server. However, it will be noted that it may also be an application accessible via the Internet and hosted by a remote server of the building whose zones are managed, the information relating to the measurements taken by the sensors then passing through a computer terminal present at the site. building level and connected to this remote server. FIG. 3 schematically illustrates, in block form, the overall operating phases of the system of FIG. 2. The various global phases described with reference to FIG. 3 will be detailed individually in the following figures. A first phase 61 (LP - Learning Phase) represents a learning phase during which the system 5 collects the information provided by the different sensors.

This information is collected periodically (for example every one to two hours, every half hour or every five to ten minutes) and is used to determine the thermal behavior of the building zones. The system 5 then establishes a model (block 63, TEMPLATE) of the thermal behavior of each of the zones. Alternatively, the same model is assigned to several areas or to the entire building. However, the finer the granularity of the model, the more efficient the planning and the optimization of the regulation.

B12156 8 The thermal behavior model is then used, again by the system 5, to predict the thermal requirements of the building zones (block 65, PP - Prediction Phase). This prediction phase is used to provide a heating plan (block 67, PLAN) to the model considered. This heating plan is provided according to a temporal division chosen according to the desired accuracy. According to a preferred embodiment, the granularity of the measurements and the heating plane is at least two hours, preferably between two hours and five minutes. The choice of granularity depends in particular on the desired accuracy and the calculation means available. The heating plan provides individual or global instructions to the heaters 21.

Moreover, during the operation of the system, the measurements of the various sensors are further exploited to adapt the heating plan in real time and / or to adjust the thermal behavior model of the zone, that is to say continue learning.

The initial learning phase consists, preferably for each zone, in defining an initial model of thermal behavior consisting of a superposition of elementary modules (sub-models) each defining an aspect of this thermal behavior.

A first module characterizes the "empty" behavior of the zone, that is to say without heating and without occupation (without human presence). This module corresponds in fact to the thermal inertia of the zone and takes into account in particular the insulation, thermal leakage to the outside and to the neighboring zones, and ventilation systems. This module takes into account the data of the indoor temperature sensors of the zone considered, of the neighboring zones and of the outside. A second module characterizes the contribution due to heating. It takes into account the indoor and outdoor temperatures as well as the instantaneous energy consumption of the zone's heaters. A third module characterizes the thermal behavior related to human activity. In other words, the human presence is considered as a source of thermal energy that is taken into account for the modeling of the thermal behavior of the building zones. This allows, among other things, to improve the thermal regulation by cutting the heating means in advance of the time of release of a room to take into account that this room is occupied by a number of people. Such anticipation over a simple real-time correction based on the current room temperature optimizes energy consumption. Preferably, the human behavior is also modeled by the potential modification (ie according to the learning data) of parameters learned in the other modules (building insulation, heating efficiency, etc.). In other words, the evolution of the different modules or sub-models by learning over time preferably takes into account the data of the presence sensors. Other optional elementary modules may be provided. For example, we can take into account a module characterizing the orientation of the area relative to the sun and the exposure of its glazed surfaces.

All these elementary modules or sub-models, and therefore the resulting model of thermal behavior, are subject to a time division with a periodicity that depends on the desired accuracy. FIGS. 4A and 4B show, in a schematic manner, modeling in the form of RC resistive and capacitive circuits of the different modules. The first module can be modeled in the form of a series RC circuit (FIG. 4A), with R representing the inverse of the conductance of the zone and C its inertia. The input voltage (across the series connection of the resistor R and the capacitor C) represents the outside temperature oe (t) and the output voltage (across the capacitance C) represents the internal temperature el (t). The other modules, which in fact represent thermal sources (heating, human presence, solar radiation, etc.), can be modeled in the form of parallel RC circuits (FIG. 4B), with the input current (sum of the currents in FIG. the resistance R and in the capacitance C) representing the thermal power Q (t) of the source 10 in question. Each elementary module is symbolized by an RC circuit symbolizing the thermal source, its inertia defining, in the manner of an RC filter, an impedance and a time constant. FIG. 5 very schematically shows in the form of blocks an example of steps of the learning phase 61. The first elementary module 611, characterizing the thermal inertia of the zone, is expressed in the form a time constant T (representing the product RC). One can write: eloe (t) = (dei (t) / dt) + el (t), or 'z (t) = (00e (t) - ei (t)) / (dei (t) / dt ). In fact, the thermal conductivity R is difficult to determine. It is easier to directly determine the thermal inertia. It suffices to periodically measure the indoor and outdoor temperatures. The successive values of T are thus measured in order to determine the shape of the function T (t). These measurements are preferably carried out for at least an initial period in the absence of a controllable heat source, i.e. at least in the absence of human presence and in the absence of heating. Likewise, night measurements (without solar radiation) are preferred. The elementary modules characterizing the thermal sources can each be expressed as the product of the thermal efficiency of the energy source by the thermal conductivity R. We can write: (011 (t) + T.dei (t) / dt), or aR (t) = (1 / Q (t)). (011 (t) + T.deli (t) / dt).

Al, a2, OE3 will denote the respective thermal efficiencies of the heating devices, the heat sources represented by the human presence, and external thermal sources such as solar radiation. There is a problem similar to the first module, namely that the thermal conductivity R is difficult to determine. To simplify the process, it is assumed that this conductivity is constant, which in first approximation is perfectly realistic for a given building. In addition, it can be seen that, for the purposes of modeling the thermal behavior, it is in practice not necessary to determine the thermal conductivity. In particular, for heating, knowing the product alR of the yield al by the thermal conductivity R is sufficient. In fact, it is assumed that this product varies over time and, by measuring the amount of energy supplied, i.e., the consumption, the evolution over time of the product aR is determined, therefore the function alR (t). For this learning, after step 611, during which the thermal inertia without heating is determined, for example over a few days, a second step 613 is performed during which the heating is activated in order to determine the shape of the function (t). This amounts to introducing the heat source representing the heating of the zone considered. For the heat source related to human presence, it is considered that the thermal efficiency of a human being can be likened to a constant. That is to say that, in a given time interval, the parameter a2R by human being is constant, that is to say that it is the same for each occupant. What varies over time is the human presence.

The determination of this parameter (step 615) is carried out by evaluating the temperature variations (with or without heating) in the presence (N data) of human beings in the zone and by estimating the human activity according to the learning data. To optionally take into account the influence of solar radiation (step 617), the energy of the radiation is known by measuring it with a sensor, thus the value a3R (t). As with other energy sources, the process is simplified by not attempting to evaluate either thermal conductivity or efficiency, but merely measuring the energy Q3 (t) provided by the radiation sensor. It is considered that by performing initial training of the elementary modules, that is to say by adding the variables one after the other, the accuracy of the model 63 of thermal behavior obtained is considerably improved. Preferably, at least the first sub-model (thermal inertia) is also adapted according to the presence of a living being in the area. Indeed, this presence 20 is capable of temporarily modifying the thermal inertia of the zone (opening / closing of exit). This allows, for example, to take into account the opening of a window without it being necessary for this window to be provided with an opening sensor. Indeed, its opening results in a variation of the temperature and can be correlated with the human presence. We can also correlate the behavior of the fourth sub-model (solar radiation) to the human presence, in that it can intervene on blinds or shutters. The fact of correlating the evolution of the various modules with the human presence considerably improves the heating plan provided, and therefore the efficiency of the regulation. The initial learning phase lasts preferably from one day to ten days. The longer this phase is, the better the initial model will be.

The prediction phase (65, FIG. 3) exploits the model of thermal behavior obtained and, more specifically, the functions T (t), aRR (t), a2R (t) and a3R (t), to determine, from preference from a schedule occupying the area and weather forecast, the heating needs of this area. In fact, the parameters T (t), AlR (t), a2R (t) and a3R (t) are taken into account to calculate a thermal behavior equation. This equation allows to provide a final internal temperature at a given moment. This temperature is deduced from the sum of the thermal energies provided by the elementary modules. According to an alternative embodiment, the thermal behavior model of a building or of different zones uses a database consisting of the learning results of other buildings as a priori for the values of the different constants. This model is then adapted (corrected) by self-learning once exploited by a given area. More generally, the model can be initialized with values T (t), al.R (t), a2.R (t) and a3.R (t) (a priori) which are adapted by learning. Taking into account the expected occupancy and the meteorology makes it possible to calculate a final temperature and to determine a heating plan (67) for the zone considered. By way of a simplified exemplary embodiment, assuming the external temperature ®e stable (constant) and knowing the internal temperature el (t0) at a time t0, the internal temperature ei (t) at time t is, assuming the empty building and the constant heating power P, given by the following relation: 30 = Oe + (0i (t0) - 00e) .e -t / T + al.RP (1-and / T). Assuming a human presence from time t1, the above relationship becomes, assuming in a simplified way that the human presence produces a heating power Q: ii (t) + (0i (tO) - eloe) .e -t / T + al.RP (1-and / T) + - (t-t1) / T). As a variant, the different modules are used individually to determine their respective influences on the temperature and these influences are summed to obtain the heating setpoint time function (the heating plane). The transmission of the heating plan to the control system can be done locally or from the outside.

Particular embodiments have been described. Various variations and modifications will be apparent to those skilled in the art. In particular, although the embodiments have been described in connection with an example taking into account solar radiation, other sources of energy may be considered. In addition, although reference has been made to human beings, the described embodiments also apply to animals (for example for regulating the heating of a barn or the like), so generally speaking to living beings . In addition, although a periodicity of measurement (and adjustment of the model and the heating plan) of between 5 minutes and 2 hours is a good compromise of efficiency, it may be possible to provide a shorter periodicity (for example, the second). Finally, the practical implementation of the embodiments that have been described is within the abilities of those skilled in the art from the functional indications given above using tools and calculation algorithms in themselves usual.

Claims (15)

REVENDICATIONS1. A method for modeling the thermal behavior of at least one zone (11, 12, 13, 14, 15) of a building (1), comprising a step of modeling living beings present in said zone in the form of a thermal source (a2R) of constant efficiency by being alive. 2. Method according to claim 1, wherein a model (63) of thermal behavior of said zone (11, 12, 13, 14, 15) comprises several sub-models among which at least: a first sub-model (T ) characterizing the thermal inertia of the zone; a second submodel (aiR) characterizing the thermal input of at least one heating device (21) or air conditioning present in the zone; and a third sub-model (a2R) characterizing the thermal input of a human being present in the area. 3. The method of claim 2, wherein a fourth sub-model (a3R) characterizes solar thermal input. 20 4. Method according to claim 2 or 3, wherein at least the first submodel (T) is also a function of the presence of a living being in the zone, in that it temporarily modifies the thermal inertia of the zoned. 5. A method according to any one of claims 2 to 4, wherein the model (63) is temporally sliced to take into account the influence of the different sub-models (T, alR, a2R, a3R) according to the periods of time. the day. 6. The method according to claim 5, wherein the model is obtained in a learning phase (61) during which measurements from sensors among which at least one indoor temperature sensor (31), a temperature sensor. (34), a human presence detector (32) and an energy consumption sensor (33) are periodically carried out.B12156 16 7. The method of claim 5, wherein the periodicity of the measurements is between 5 minutes and 2 hours. 8. A method according to any one of claims 2 to 7, reproduced for several areas (11, 12, 13, 14, 15) of the building. 9. A method for planning a thermal regulation of a zone (11, 12, 13, 14, 15) of a building, using a model obtained by carrying out the method according to any one of the preceding claims. to simulate the thermal behavior of the area. 10. The method of claim 9, in which the planning takes into account a provisional schedule of occupation of the area by living beings. 15 The method of claim 9 or 10, wherein the planning takes into account weather forecasts. The method of any one of claims 9 to 11, wherein the model is updated over time. 13. A method according to any one of claims 9 to 12, wherein the planning results in a heating scheme operated by a heating system. 14. System (5) for thermal regulation of at least one zone (11, 12, 13, 14, 15) of a building (1), comprising a computer tool programmed for carrying out the method according to any one of claims 9 to 13.