SELF-CALIATION OF A VENTILATION SYSTEM

Description

Technical field

[0001] The present invention relates generally to the field of ventilation, and more specifically to a method for self-calibration of a VMC (Controlled Mechanical Ventilation) type ventilation installation, comprising: a central box comprising a variable-speed air extraction or supply fan, N air extraction or supply ducts, a first end of each air extraction or supply duct being connected to an air inlet or outlet connection of said central box, at least one air discharge or supply duct, and a flow regulator with adjustable passage opening for each air extraction or supply duct. Technological background

[0002] As schematized on the figure 1, a non-limiting example of a known residential ventilation installation comprises a ventilation system comprising a fan 1 for extracting an air flow at fixed or variable speed, preferably placed in a volute of a central box 2 comprising a number N of air inlet connections such as the three air inlet connections 3 1 , 3 2 , 3 3 visible on the figure 1, each connected to a first end of an air extraction duct 4 1 , 4 2 , 4 3 , a plurality of air extraction vents such as the three extraction vents 5 1 , 5 2 , 5 3 each connected to the other end of an extraction duct, and at least one discharge device 6 such as a roof cap, connected to an outlet 7 of the central box 2 via a discharge duct 8. The extraction ducts equipped with extraction vents open for example into wet rooms (bathroom, toilet, kitchen) and are made up of mechanical parts which can adapt the opening (or passage section) of the air according to one or more parameters such as, but not limited to, the humidity level of the room, the detection of the presence of a person, or by any mechanical action (kitchen pull cord etc.). The air extraction fan 1 and the extraction vents are generally independent of each other.The air extraction fan 1 extracts an overall air flow by providing the vents with an overall pressure allowing them to operate at the correct extraction flow rate. Each extraction vent constitutes a flow regulator with adjustable passage opening which operates by modifying the passage section of the air entering the corresponding duct according to the humidity or customer needs without taking into account the state of the centralized extraction fan 1. In a so-called speed jump system depending on the flow rate, the system can detect a variation in the flow rate and adapt the speed of the fan 1 in order to modify the pressure available to the vents.

[0003] Alternatively, as shown diagrammatically in the figure 2 , extraction vents 5 1 , 5 2 , 5 3 of the figure 1 are replaced by mouths 5 1 ′ , 5 2 ′ And 5 3 ′ whose function is purely aesthetic, and the flow regulators with adjustable passage opening are made up of registers 5 1 " , 5 2 " And 5 3 " each comprising a movable flap, and placed inside the air inlet tappings of the central box 2. Each register will adjust the position of the flap in order to meet the extraction flow rate requirement of the room to which it is connected via the extraction duct, based on information from sensors such as, but not limited to, humidity sensors, VOCs, etc. Document EP 2 363 656 A2 shows a method according to the preamble of claim 1.

[0004] In some known systems, each tapping is additionally equipped with a flow sensor. Each damper can thus regulate the position of the associated flap to match the flow measured by the flow sensor to a set flow rate. However, flow sensors are expensive devices.

[0005] However, to control the extraction flow rates of each room without resorting to expensive flow sensors, it is necessary to know precisely, on the one hand, the pressure made available by the extraction fan 1, and on the other hand, the pressure losses for each duct of the installation, namely the pressure differences in each duct at a given flow rate. When first commissioning an extraction fan ventilation system such as that shown in any of the diagrams figures 1 or 2 previous, the load losses induced by the installation's own air network are generally unknown.

[0006] Furthermore, after a certain period of use of the installation, the pressure losses may have changed, for example due to clogging of the extraction ducts. Summary of the invention

[0007] The present invention aims to overcome the limitations of the prior art by proposing a simple solution for precisely determining the pressure losses of a ventilation installation, consisting of determining in situ the pressure losses of the installation, whether at the time of the first commissioning of the installation, or at any time during the use of the installation.

[0008] More specifically, the present invention relates to a method for self-calibration of a controlled mechanical ventilation installation, VMC, said VMC installation comprising: a central box comprising an air extraction or supply fan; N air extraction or supply ducts, a first end of each air extraction or supply duct being connected to an air inlet or outlet connection of said central box; at least one air discharge or supply duct, and a flow regulator with adjustable passage opening for each air extraction or supply duct, said self-calibration method being according to claim 1.

[0009] In a first embodiment not covered by claim 1, the in situ determination of a pressure loss coefficient k¡ associated with the i th air extraction or insufflation duct comprises the following successive steps: a) the flow regulator associated with the i th< air extraction or air supply duct is controlled so that it is in an open position and all other flow regulators are in a closed position; b) the air extraction or air supply fan is activated at a given speed; c) the flow rate QT of the air extraction or air supply fan is determined and the difference ΔP between the atmospheric pressure P 0 and the pressure P -< provided by the air extraction or air supply fan is measured; d) the pressure loss coefficient ki is calculated from the determined flow rate QT and the measured difference ΔP.

[0010] In this case, the pressure loss coefficient ki can be calculated in step d) by applying the relation in which kreg i is a known pressure loss coefficient of the flow regulator associated with the i th air extraction or insufflation duct.

[0011] In a second embodiment not covered by claim 1, the in situ determination of a pressure loss coefficient ki associated with the i th air extraction or insufflation duct comprises the following successive steps: a) the flow regulator associated with the i th air extraction or supply duct is controlled so that it is in an open position and all the other flow regulators are in a closed position; b) the air extraction or supply fan is activated at a first given speed v1; c) the flow rate Q Tv1 of the air extraction or supply fan is determined for this first given speed and the difference ΔP v1 between the atmospheric pressure P 0 and the pressure P -< made available by the air extraction or supply fan is measured; d) the air extraction or supply fan is activated at a second given speed v2; e) for this second given speed, the flow rate Q Tv2 of the air extraction or supply fan is determined and the difference ΔP v2 between the atmospheric pressure P 0 and the pressure P -< made available by the air extraction or supply fan is measured;f) the pressure drop coefficient ki is calculated from the flow rates Q Tv1 and Q Tv2 determined and the differences ΔP v1 and ΔP v2 measured in steps c) and e). ;

[0012] In this case, the pressure loss coefficient ki can be calculated in step f) by solving the system of equations: in which kreg i is a known pressure loss coefficient of the flow regulator associated with the i th air extraction or insufflation duct.

[0013] In possible embodiments, said open position preferably corresponds to a maximum position for which the flow regulator associated with the i th air extraction or insufflation duct allows the passage of a maximum flow rate.

[0014] In the case of the first embodiment mentioned above, the in situ determination of the pressure loss coefficient k R is carried out at step d), by applying the following relationship: where ΔP v is the known fan pressure.

[0015] In this case, the in situ determination of the pressure loss coefficient k R can be carried out by the following additional steps: e) all flow regulators are controlled so that they are in an open position; f) the air extraction or supply fan is activated at a given speed; g) the flow rate QT of the air extraction or supply fan is determined and the difference ΔP between the atmospheric pressure P 0 and the pressure P -< provided by the air extraction or supply fan is measured; h) the pressure loss coefficient k R is calculated by applying the following relationship: where ΔP v is the known fan pressure.

[0016] The present invention also relates to a ventilation installation comprising: a central box comprising an air extraction or supply fan; N air extraction or supply ducts, a first end of each air extraction or supply duct being connected to an air inlet or outlet connection of said central box; at least one air discharge or supply duct, and a flow regulator with adjustable passage opening for each air extraction or supply duct, said installation being characterized in that the central box comprises self-calibration means capable of determining and storing in situ a pressure loss coefficient associated with each air extraction or insufflation duct by implementing the steps of the method according to claim 1.

[0017] Each flow regulator may be an air extraction or insufflation vent arranged at a second end of the corresponding air extraction or insufflation duct. Alternatively, each flow regulator is a register arranged inside the corresponding air inlet or outlet connection of said central box. Brief description of the figures

[0018] The following description with reference to the attached drawings, given as non-limiting examples, will make it clear what the invention consists of and how it can be implemented. In the attached figures: There figure 1 , already described above, schematically illustrates a ventilation installation with adjustable extraction vents not forming part of the invention; The figure 2 , already described above, schematically illustrates a ventilation installation with adjustable extraction registers not forming part of the invention; The figure 3represents a modeling of the pressure losses corresponding to a ventilation installation with adjustable extraction dampers of the figure 2 ; There figure 4 illustrates possible steps for a first self-calibration method not forming part of the invention; The Figure 5 illustrates possible steps for a second self-calibration method in accordance with the present invention. Description of embodiment(s)

[0019] In the figures, identical or equivalent elements will bear the same reference signs. The various diagrams are not to scale.

[0020] In a ventilation installation as shown schematically in the figure 2, the pressure losses induced by the air network depend on the one hand, on each extraction branch (each branch comprising an extraction duct and, at both ends of the duct, an extraction vent and an extraction register), and on the other hand, on the discharge branch(es) (each discharge branch comprising a discharge duct and a roof cap).

[0021] There figure 3 represents a model of the air network of an installation similar to that shown schematically on the figure 2 , comprising N branches referenced B 1 to BN . The following notations and references are used for this diagram: P -< is the pressure made available by the air extraction fan 1 at the central box 2; P 0 is the atmospheric pressure (considered hereinafter as zero for the sake of simplification); ΔP is the difference between the pressure P 0 and the pressure P -< made available by the air extraction fan 1; PR is the pressure at the outlet 7 of the central box 2, connected to the discharge duct 8; k R is a pressure loss coefficient of the discharge branch composed of the discharge duct 8 and the discharge means 6; ΔP v is the pressure generated by the fan; ΔP ii is the pressure loss of the discharge branch; B i is an i th< branch of the air network (i being an integer varying from 1 to N); P i is the pressure at the inlet of the 5" i register of branch B i; ΔP i is the pressure drop created by the 5" i register of branch B i;ΔPg i is the pressure loss created by branch B i , outside register 5" i , in other words, the pressure loss associated with the extraction duct 4 i ; kreg i is the pressure loss coefficient created by register 5" i of branch B i , for a given position of register 5" i ; ki is the pressure loss coefficient associated with branch B i , outside register 5" i , in other words, the pressure loss coefficient associated with the extraction duct 4 i ; Q i is the flow rate at each branch B i ; QT is the flow rate of air extraction fan 1. ;

[0022] The air extraction fan 1 may be a fan whose impeller rotates at a fixed speed. Alternatively, the fan may be variable speed. In this case, the rotational speed of the impeller may be controlled (for example by a constant speed or constant pressure or constant flow rate, or constant torque control).

[0023] It is known that the pressure loss ΔPg i can be calculated classically by the following relation (1): [Math. 1] Δ Pg i = k i * Q i a = Δ P − ΔP i in which a is a coefficient allowing to take into account the turbulent effects due to friction inside the ducts which modify the behavior of the pressure drop.

[0024] In possible implementations of the method not forming part of the invention, it can be considered that this coefficient a is equal to 2, so that relation (1) above becomes the following relation (2): [Math. 2] Δ Pg i = k i * Q i 2 = Δ P − ΔP i

[0025] A self-calibration method according to the invention aims to allow self-calibration means included in the central box 2 (typically a controller not shown) to determine in situ at least the pressure drop coefficient ki associated with each duct of each branch B i of the installation. The coefficients ki thus determined are stored in a memory (not shown) associated with the controller. At the end of the self-calibration, the controller is then able to calculate, at any time during operation of the ventilation installation, the actual pressure drop ΔPg i of a duct 4i of the branch B i for any flow rate Q i , by applying the first equality of the relation (1) or (2).

[0026] Several self-calibration methods will now be described. All these methods consist of determining in situ at least the pressure loss coefficients ki for all the ducts of the installation, by controlling the dampers so that they occupy different opening positions, according to a particular sequencing differing from one method to another, and by measuring or estimating in situ the flow rates QT obtained at different stages of the sequencing. These methods differ from each other depending on whether or not there is, in the ventilation system (in this case in the central box 2), a sensor to measure the difference ΔP between the pressure P 0 and the pressure P -< made available by the air extraction fan 1. Case 1: ΔP measurement available

[0027] In this case, it is assumed that the ventilation system is equipped, inside the central box 2, with a pressure sensor (not shown) which measures the value ΔP. This pressure sensor has a measuring point placed between the fan 1 for extracting the air flow and the air inlet connections 3 1 , ...3 i , ...3 N , and another measuring point outside to measure the atmospheric pressure P 0 serving as a reference.

[0028] There figure 4 represents the possible steps of implementing a self-calibration method not in accordance with the present invention:

[0029] The method preferably begins with a preliminary phase 100 aimed at defining the reference (zeroing) of the position of the flaps of the various 5" i registers and at estimating the internal air leaks of the installation. Knowledge by the controller of the position of the registers may in fact be made necessary depending on the technology used for these registers, for example in the non-limiting case where these registers use a stepper motor without a position sensor to change the position of the register, that is to say to move a flap between an open position and a closed position corresponding to mechanical stops.

[0030] The preliminary phase 100 begins with a step 101 during which the controller, preferably positioned inside the central box 2, sends to each 5" i damper (i varying from 1 to N) a stop search order. When all the dampers are at the stop, the controller commands the closing of all the dampers (step 102) so that each 5" i damper is in a closed position in which it allows the passage of a minimum leakage flow.

[0031] The controller then activates the air extraction fan 1 at at least one given speed and triggers the measurement or estimation of the air flow rate QT of the fan (step 103). This measurement or estimation can be implemented in various ways.

[0032] A first solution consists of using a flow sensor at the fan outlet, the flow sensor being for example an ultrasonic sensor or a propeller / iris flow meter with pressure drop measurement.

[0033] Another solution consists of using a fan map previously stored in the memory associated with the controller, this map giving on the one hand, the value of the flow rate QT as a function of the power absorbed by the fan, its rotation speed and its control, and on the other hand, the pressure P -< made available by the air extraction fan 1 as a function of the flow rate QT . Thus, by measuring the power consumed and the speed of the fan, it is possible to obtain an estimate of the flow rate QT , and of the pressure P -< .

[0034] Another solution can be used if fan 1 is a jet fan. Indeed, in this case, the curves connecting the pressure P -< to the flow rate QT for different fan speeds are bijective. The measurement of the pressure P -< by the pressure sensor thus makes it possible to deduce the flow rate QT

[0035] Other solutions could be to measure the pressure difference in the roof and the air intake in the central box 2.

[0036] Whatever the solution chosen to measure or estimate the QT flow rate, this measured or estimated flow rate must be very low and correspond only to the internal leaks of the box since all the registers are closed.

[0037] If the measured or estimated flow rate QT corresponding to internal leaks is too high, for example higher than a threshold stored in the memory associated with the controller, this may mean that an air inlet is not closed (tapping not connected, missing cap on the box, etc.). In this case, the controller preferably generates a visual and / or audible alarm (step 104) and the self-calibration process ends, until the problem is resolved.

[0038] Otherwise, the preliminary phase 100 ends with a step 105 during which the controller stores in its associated memory the value of the measured or estimated flow rate QT, divided by the number N of registers, so as to be able to subsequently take into account these leak values ​​per register.

[0039] The self-calibration process can be continued by the steps making it possible to determine at least the pressure loss coefficient ki of each sheath 4 i in each branch B i , and preferably also the pressure loss coefficient k R of the rejection sheath. Subcase 1: coefficient a is equal to 2 :

[0040] As seen previously, the pressure loss ΔPg i can be calculated in this case by the relation (2) given above. Thus, each pressure loss coefficient ki can be determined by the following relation (3): [Math. 3] k i = ΔP Q i 2 − ΔP i Q i 2 = ΔP Q i 2 − kreg i in which ΔP i is a function translating the behavior of the 5" i register depending on the flow rate Q i and the opening position α of the register and kreg i is the coefficient of pressure loss created by the 5" i register of the branch B i , for a given position of the 5" i register. Each register having been characterized beforehand, the values ​​ΔP i and kreg i are known for each possible flow rate value Q i and for each possible position α of the register, and stored in the memory associated with the controller.

[0041] The following steps are performed to calculate each pressure drop coefficient ki and the pressure drop coefficient k R of the discharge branch:

[0042] The controller commands the opening, preferably maximum, of a single 5" damper i , for example the 5" damper 1 (i = 1) and the closing of all the other dampers (step 110). The controller then commands the start of fan 1 at a given speed (any), which allows the system to deliver an available pressure P -< (step 120).

[0043] In a step 130, the controller triggers the measurement of ΔP (by the pressure sensor) and the measurement of the air flow rate QT according to any of the methods described above. As only the 5" damper i is in the open position, the measured air flow rate QT corresponds to the flow rate Q i in the branch B i , excluding the tapping leaks, so that the controller can proceed (step 140) to the calculation of the pressure drop coefficient by applying the following relationship (4): [Math. 4] k i = ΔP Q T 2 − kreg i

[0044] During this step 140, the controller can also calculate the pressure drop coefficient k R of the discharge branch by applying the following relation (5): [Math. 5] k R = ΔP V Q T 2 − ΔP Q T 2 in which ΔP v is the fan pressure obtained by mapping, as described above.

[0045] At the end of step 140, the memory associated with the controller has the load loss coefficient ki for the branch B i tested and the load loss coefficient k R .

[0046] Steps 110 to 140 are repeated until the pressure loss coefficients associated with the N branches have been determined.

[0047] It can be seen that according to this method, the calculation of the pressure drop coefficient k R of the discharge branch is carried out at each iteration of steps 110 to 140. In practice, the values ​​of k R obtained at each iteration are never equal. A choice must then be made by a method allowing the value of k R to be obtained closest to reality, for example by taking an average of the values ​​of k R obtained at each iteration. Subcase 2: coefficient a is not equal to 2 :

[0048] As seen previously, the pressure loss ΔPg i can be calculated in this case by the relation (1) given above. Thus, each pressure loss coefficient ki and the pressure loss coefficient k R can be determined by the following relations (6): [Math. 6] k i = ΔP Q i a − ΔP i Q i a = ΔP Q i a − kreg i k R = ΔP V Q T a − ΔP Q T a with kreg i corresponding to the pressure loss coefficient created by the 5" i register of the branch B i , for a given position of the 5" i register.

[0049] Steps similar to steps 110 to 140 and described above will in this case have to be carried out to calculate each pressure drop coefficient ki and the pressure drop coefficient k R of the discharge branch. However, to be able to overcome the lack of knowledge of the coefficient a, steps 120 and 130 seen previously must be carried out not for a single speed of the air extraction fan 1, but for at least two different speeds v1 and v2 of the fan 1, so as to have two pairs of measurements [ΔP v1 , Q Tv1 ] and [ΔP v2 , Q Tv2 ], and solve the following system of equations (7): [Math. 7] k i = ΔP v1 Q T <none / > v1 a − kreg i = ΔP v2 Q T <none / > v2 a − kreg i 7 a k R = ΔP V_v1 Q T <none / > v1 a − ΔP v1 Q T <none / > v1 a = ΔP V_v2 Q T <none / > v2 a − ΔP v2 Q T <none / > v2 a 7 b in which ΔP v1 is the difference ΔP for speed v1; ΔP v2 is the difference ΔP for speed v2; Q Tv1 is the flow rate QT for speed v1; Q Tv2 is the flow rate QT for speed v2; ΔP V_v1 is the fan pressure, obtained by mapping, for speed v1; and ΔP V_v2 is the fan pressure, obtained by mapping, for speed v2. Variant for determining the coefficient k R :

[0050] In the method described with reference to the figure 4 , we have seen that the calculation of the pressure drop coefficient k R of the discharge branch was carried out at each iteration of steps 110 to 140. Alternatively, however, this calculation can be carried out independently, for example before or after all the pressure drop coefficients ki have been calculated by applying the relations (4) or (7) indicated above and stored in the memory associated with the controller. To do this, we can proceed for example in the following manner, in the case where the coefficient a is equal to 2: the controller controls the opening, preferably maximum, of all the N registers, then controls the start of the fan 1 at a given speed (any), which allows the system to deliver an available pressure P -<; the controller triggers the measurement of ΔP (by the pressure sensor), and the measurement or estimation of the air flow QT and the available pressure P -< according to any one of the methods described above. As all the registers are in the open position, the measured air flow QT corresponds to the sum of the flow rates Q i in the branch B i , excluding the tap leaks; the controller then determines the pressure loss coefficient k R by applying the relationship (5).

[0051] Note that the previous steps could be performed for any register positions. However, maximum register opening is preferred because the higher the total flow rate QT, the greater the accuracy of calculating KR.

[0052] In the case where the coefficient a is not equal to 2, the first two previous steps are carried out for two different speeds v1 and v2 of fan 1, so that the controller can then determine the pressure drop coefficient k R by solving the relation (7b) above.

[0053] It should be noted that the determination of the coefficient k R is not mandatory here in the sense that knowledge of this coefficient is not used to determine the flow rate in each Q i branch. Case 2: ΔP measurement not available

[0054] Unlike the previous case, it is assumed here that the ventilation system does not have any pressure sensor and is therefore not capable of measuring the ΔP value.

[0055] There Figure 5 represents the possible steps of implementing a self-calibration method in accordance with the present invention and particularly adapted to this scenario, and for which the fan 1 is here a variable speed fan:

[0056] The method preferably begins with the preliminary phase 100 aimed at defining the reference (zeroing) of the position of the flaps of the different registers 5" i and at estimating the internal air leaks of the central box 2, described above with reference to the figure 4 .

[0057] The controller must then proceed by selecting pairs of registers, as will be described below:

[0058] The controller commands the opening, preferably maximum, of a first 5" damper i of a first branch B i , for example the 5" damper 1 (i = 1) and the closing of all the other dampers (step 210). The controller then commands the start of the fan 1 at a given speed (any), which allows the system to deliver an available pressure P -< (step 220).

[0059] In a step 230, the controller triggers the measurement or estimation of the air flow rate QT according to any of the methods described above. As only the first damper 5" i is in the open position, the measured air flow rate QT corresponds to the flow rate Q i in the first branch B i , excluding tapping leaks. In step 230, the controller also triggers the measurement of the value ΔP v_i which corresponds to the pressure of the fan 1 when only the first damper of the first branch B i is open. This measurement is obtained using the fan map, as stored in the memory associated with the controller.

[0060] The controller then commands the closing of the first damper 5" i , and the opening, preferably maximum, of a second damper 5" i+1 of a second branch B i+1 , for example the damper 5" 2 and (step 240). Steps similar to steps 220 to 230 are then applied for this second branch B i+1 . Thus, the controller commands the starting of the fan 1 at a given speed (any), which allows the system to deliver a new available pressure P -< (step 250). Then, during a step 260, the controller triggers the measurement of the air flow rate QT according to any one of the methods described above. As only the second damper 5" i+1 is in the open position, the measured air flow rate QT corresponds to the flow rate Q i+1 in the second branch B i+1 , apart from the tapping leaks.During step 260, the controller also triggers the measurement of the value ΔP v_i which corresponds to the pressure of fan 1 when only the second register of the second branch B i+1 is open.

[0061] The controller then commands the simultaneous opening, preferably maximum, of the dampers of the first branch B i and of the second branch B i+1 (step 270), then commands the start of the fan 1 at a given speed (any), and triggers the measurement of the air flow rate QT according to any one of the methods described above (step 290). As the first and second dampers of the branches B i and B i+1 are in the open position, the measured air flow rate QT corresponds to the sum of the flow rates Qi and Q i+1 in the first and second branches, excluding tap leaks. During step 290, the controller also triggers the measurement of the value ΔP v_i_i+1 which corresponds to the pressure of the fan 1 when only the first damper and the second damper of the first branch B i , respectively the second branch B i+1 , are open.

[0062] At the end of step 290, the memory associated with the controller therefore has three pairs of measurements, namely: the pair (Q i ; ΔP v_i ) obtained in step 230; the pair (Q i+1 ; ΔP v_i+1 ) obtained in step 260; the pair (QT ; ΔP v_i_i+1 obtained in step 290;

[0063] These three pairs of values ​​are sufficient to allow the controller to perform, during a step 300, the calculations of the pressure loss coefficients ki , k i+1 and k R . Indeed, according to the modeling represented on the figure 3 , when only the first register of the first branch B i is open (steps 220 and 230), we have the following relation (8): [Math. 8] ΔP v_i = k R + k i * Q i 2 + ΔP i or, by combining with relation (2) given above, the following relation (9): [Math. 9] ΔP v_i = k R + K i * Q i 2 with K i = k i + kreg i

[0064] Furthermore, when only the second register of the second branch B i+1 is open (steps 250 and 260), we have the following relation (10): [Math. 10] ΔP v_i + 1 = k R + k i + 1 ∗ Q i + 1 2 + ΔP i + 1 or, by combining with relation (2) given above, the following relation (11): [Math. 11] ΔP v_i + 1 = k R + K i + 1 * Q i + 1 2 with K i + 1 = k i + 1 + kreg i + 1

[0065] Finally, when the first register and the second register of branches B i and B i+1 are simultaneously open (steps 280 and 290), we have the following relation (12): [Math. 12] ΔP v_i_i + 1 = k R + K * Q T 2 with the variable K = K i * K i + 1 K i + K i + 1 2

[0066] With relations (9), (11) and (12) above, the controller has a system of three equations with three unknowns consisting of the three pressure loss coefficients ki, k i+1 and k R sought and can therefore calculate and then store these coefficients in step 300.

[0067] Steps 210 to 300 are repeated for another pair of branches, until all the pressure loss coefficients ki have been calculated. For example, as shown in the Figure 5, increment i by the value 2 if N is even or by the value 1 if N is odd. The process then ends when the incremented value of i is greater than N in the case where N is even, or is equal to N in the case where N is odd.

[0068] As in the case of the method described with reference to the figure 4 , we note that the calculation of the pressure drop coefficient k R of the discharge branch is carried out at each iteration of steps 210 to 300. In practice, the values ​​of k R obtained at each iteration are never equal. We must then make a choice by a method allowing us to obtain the value of k R closest to reality, for example by taking an average of the values ​​of k R obtained at each iteration.

[0069] In summary, when the system does not have a pressure sensor allowing it to measure or estimate the ΔP value, the process according to the Figure 5 consists of: a) selecting a pair consisting of a first damper and a second damper; b) controlling (step 210) the first damper of the selected pair to be in an open position and all other dampers to be in a closed position; c) activating (step 220) the air extraction fan 1 at a given speed; d) determining (step 230) the flow rate QT of the air extraction fan (1) and measuring the pressure ΔP v_i of the air extraction fan 1; e) controlling (step 240) the first damper of the selected pair to be in a closed position, and the second damper of the selected pair to be in an open position; f) repeating steps c) and d) above so as to determine (260) the flow rate QT of the air extraction fan 1 and measure the pressure ΔP v_i+1 of the air extraction fan 1;g) controlling (step 270) the first damper and the second damper of the selected pair so that they are simultaneously in an open position; h) repeating steps c) and d) above so as to determine (290) the flow rate QT of the air extraction fan 1 and measure the pressure ΔP v_i_i+1 of the air extraction fan 1; i) deducing (step 300) the values ​​of the coefficients ki , k i+1 associated with the two registers of the selected pair, and the coefficient k R .;

[0070] Steps a) to i) above are repeated for another pair of registers comprising at least one register different from those constituting the pair of the previous iteration, until the coefficient associated with each duct of the installation has been obtained.

[0071] The self-calibration methods according to the invention have been described in a non-limiting manner in the case of an air extraction ventilation installation. The principles of the invention are nevertheless applicable to the case where the ventilation installation is an air insufflation one. In this case, it is sufficient to replace the air extraction fan 1 with an air insufflation fan. The N air extraction ducts 4 1 , 4 2 , 4 3 become air insufflation ducts, and the discharge duct 8 becomes an air inlet duct, preferably equipped with an air filter. In this case, the pressure drop of this filter must be taken into account in the calculation of the coefficient k R . In addition, the methods have been described in the context of the installation shown schematically in the figure 2 for which the flow regulators with adjustable passage opening are registers 5 1 " , 5 2 " , 5 3 " installed in the tappings of the central box 2. Alternatively, the flow regulators can be the adjustable vents 5 1 , 5 2 , 5 3 shown diagrammatically on the installation of the figure 1 .

[0072] In all cases, it is possible to launch a self-calibration procedure at any time, whether at the first installation of the central box 2 and the associated vents, or after this first installation (for example during a maintenance operation, or following an incident), and thus to know precisely the pressure loss coefficients of the environment in which the ventilation system operates. Thanks to the invention, it is possible in particular to regularly re-evaluate the coefficients ki, and thus estimate the state of fouling of the ducts. In the case of an air insufflation installation, a regular re-evaluation of the coefficient k R also makes it possible to evaluate the state of fouling of the filter of the air inlet duct, and to change it if necessary.

Claims

1. Method for autocalibration of a controlled mechanical ventilation (CMV) installation, said CMV installation comprising: - a central box (2) having an air extraction or insufflation fan (1); - N air extraction or insufflation ducts (41, 42, 43), a first end of each air extraction or insufflation duct (41, 42, 43) being connected to an air inlet or outlet tapping (31, 32, 33) of said central box (2); - at least one air discharge or inlet duct (8), and - a flow rate regulator (51, 52, 53; 5"1, 5"2, 5"3) with adjustable passage opening for each air extraction or insufflation duct (41, 42, 43), said autocalibration method being characterized in that it consists in determining and recording in situ a head loss coefficient associated with each air extraction or insufflation duct (41, 42, 43) and a head loss coefficient kR associated with said at least one air discharge or inlet duct (8), the determination in situ of a head loss coefficient associated with each air extraction or insufflation duct and of a head loss coefficient kR associated with said at least one air discharge or inlet duct (8) comprising the following steps: a) selecting a pair constituted of a first flow rate regulator and of a second flow rate regulator of the ventilation installation; b) controlling (210) the first flow rate regulator of the selected pair such that it is in an open position, and all the other flow rate regulators such that they are in a closed position; c) activating (220) the air extraction or insufflation fan (1) at a given speed; d) determining (230) the flow rate QT of the air extraction or insufflation fan (1) and measuring the pressure ΔPv_i of the air extraction or insufflation fan (1) ; e) controlling (240) the first flow rate regulator of the selected pair such that it is in a closed position, and the second flow rate regulator of the selected pair such that it is in an open position; f) reiterating steps c) and d) above so as to determine (260) the flow rate QT of the air extraction or insufflation fan (1) and to measure the pressure ΔPv_i+1 of the air extraction or insufflation fan (1); g) controlling (270) the first flow rate regulator and the second flow rate regulator of the selected pair such that they are simultaneously in an open position; h) reiterating steps c) and d) above so as to determine (290) the flow rate QT of the air extraction or insufflation fan (1) and to measure the pressure ΔPv_i_i+1 of the air extraction or insufflation fan (1); i) deducing (300) the values of the coefficients ki, ki+1 associated with the ducts of the two flow rate regulators of the selected pair, and the coefficient kR associated with said at least one air discharge or inlet duct (8); steps a) to i) being reiterated with another pair comprising at least one flow rate regulator that differs from the ones constituting the pair for the preceding iteration until the coefficient for each flow rate regulator of the installation has been obtained.

2. Ventilation installation comprising: - a central box (2) having an air extraction or insufflation fan (1); - N air extraction or insufflation ducts (41, 42, 43), a first end of each air extraction or insufflation duct (41, 42, 43) being connected to an air inlet or outlet tapping (31, 32, 33) of said central box (2); - at least one air discharge or inlet duct (8), and - a flow rate regulator (51, 52, 53; 5"1, 5"2, 5"3) with adjustable passage opening for each air extraction or insufflation duct (41, 42, 43), said installation being characterized in that the central box (2) has autocalibration means able to determine and record in situ a head loss coefficient associated with each air extraction or insufflation duct (41, 42, 43) through implementation of the method according to Claim 1.

3. Ventilation installation according to Claim 2, wherein each flow rate regulator (51, 52, 53) is an air extraction or insufflation aperture disposed at a second end of the corresponding air extraction or insufflation duct (41, 42, 43) .

4. Ventilation installation according to Claim 2, wherein each flow rate regulator (5"1, 5"2, 5"3) is a damper disposed inside the corresponding air inlet or outlet tapping (31, 32, 33) of said central box (2) .