Open-loop control system for improved actuated damper air flow control in HVAC, VAV and heat exchange systems

Abstract

This invention relates to an air flow open-loop control system (20) with the ability to correct for non-linearity in a non-linear actuated damper, resulting in improved precision and less pressure variations when controlling multiple face-dampers (31) and bypass-damper (52) The controller includes a processor that receives data on the damper's position and determines the non-linearity of the damper. The correction module then corrects for this non-linearity so that the damper can be used to steer a linearly proportional air flow based on a linear control signal. The correction module generates a corrected signal for the damper that is indicative of the desired position for the damper that will result in the desired air flow. In operation mode, the processor uses this corrected signal to control the damper and produce a linearly proportional air flow. This invention enables more precise control of HVAC air flow and ensures that non-linearity in the damper does not affect the system's performance.

Description

[0001] OPEN-LOOP CONTROL SYSTEM FOR IMPROVED ACTUATED DAMPER AIR FLOW CONTROL IN HVAC, VAV AND HEAT EXCHANGE SYSTEMS

[0002] TECHNICAL FIELD

[0003] Control systems for air flow control using dampers in Heating, Ventilation, and Air Conditioning (HVAC), Variable Air Volume (VAV) including baffles, chilled beams, plenum boxes and heat exchange systems.

[0004] BACKGROUND

[0005] In HVAC, VAV and heat exchange systems, air flow can be controlled using dampers. Air flow control systems may use a computer, processor, control logic or software to decide how and when the air flow should be controlled based on input parameters. Outdoor air temperature (OAT) or outdoor air relative humidity are sensor data that may be observed by an air flow control system to determine control actions. Actuators are used to open, close, and adjust dampers to maintain a desired air flow, air flow distribution, temperature, relative humidity, and pressure.

[0006] The control system controls one or several actuators that are in mechanical or pneumatical connection with adjustable air dampers. Adjustable dampers may be of a variety of models having different air flow control characteristics. Actuators are typically controlled using a voltage control signal between 0 and 10 [Volt], As a result, the damper is set to a certain opening angle, or percentage of open area, allowing a corresponding air flow to pass.

[0007] The relationship between actuator control signal between 0 to 10 [V] to the relative airflow percentage is rarely linear, due to the behavior of actuator and damper functions, as well as damper design, dynamics of the air ventilation system, and interaction between different dampers. HVAC, VAV and heat exchange systems are complex systems where all communicating cavities may interact with each other, along with air flow. The mentioned systems are complex and challenging to control and follow the laws of compressible fluid dynamics.

[0008] Also, multiple air-dampers can be combined into air-damper units comprising multiple air-dampers of different, and or similar control characteristics. An example of such a controllable air-damper unit may comprise one adjustable by-pass air-dampers and several controllable face-dampers, where a first face-damper and an at least second face-damper may be controlled independently, or in coordination by the air-flow control system.

[0009] An earlier invention EP0321515 B1 , Procedure and means for controlling volumetric flow rate in air-conditioning installations, describes a procedure where the volumetric flow passing through an air-conditioning duct is controlled by a damper disposed in the air duct by opening or closing the control damper. EP0321515 B1 is characterized in that the damper control signal is inversed relative to the effective characteristic curve of the control damper, such that a change of the control signal corresponds to a given change of the volumetric flow so that a 0 to 100 % control value change linearly corresponds to a 0 to 100 % volumetric flow rate change. A procedure to achieve a linearized control function for a single damper is taught in EP0321515 B1 , with: a first aim to alter the effective characteristic curve of the control damper in such a manner that to a given change of control signal corresponds to a given change of volumetric flow rate, and a second aim to linearize the opening of the control damper.

[0010] In EP0321515 B1 , the user will have to enter as quoted from its description:

[0011] “As taught by the invention, one may particularly in the electrical procedure apply different non-linearizing functions. To carry out a given, and desired, transformation, the user may select the non-linearizing function curve he desires from among various, pre-defined such curves stored in the computer memory. The user may also exert an influence with the aid of the connection associated with the control damper, directly on the non-linearity of the control signal. The user may tune the non-linearity curve to conform to the system that is being controlled by means of such parameters which have been measured on the system that is being controlled and which have been supplied to the non-linearizing unit.”

[0012] The procedure in EP0321515 B1 requires the user to select the non-linearizing function or tune the non-linearity curve using parameters that have been measured on the system, that is being controlled, that is a damper.

[0013] But what if the system needs to control a group of actuated dampers, or a damper unit having multiple individually actuated dampers, sharing the same volumetric pressure chamber or air duct? In a damper group or damper unit, opening one damper section may influence the air flow of the remaining sections, even in a constant pressure chamber. For example, opening a first damper section completely may influence the flow through a nearby second damper section which previously was 3% open.

[0014] A way to operate complex non-linear actuated damper configurations by an open-loop control system, would need to consider more aspects than controlling just a single non-linear actuated damper, namely, how to control complex non-linear actuated damper units comprising multiple dampers, having for example multiple actuated face-dampers for multiple sections, as well as actuated bypass-dampers. Also, being able to control any type of non-linear as well as linear actuated damper, alone or in combinations could shorten commissioning and time to adjust the control system for precision operation.

[0015] As a result, there is a need for an improved air flow control system for complex nonlinear actuated dampers in HVAC, VAV and heat exchange systems, as presented in the following invention.

[0016] SUMMARY OF THE INVENTION

[0017] The invention is an improved method for controlling air flow, actually here the distribution of air flow as a percentage of a maximal air flow, using dampers with actuators having non-linear or linear proportional control scheme between the control actuator signal, typically a voltage between 0 and 10 [Volt] but may be another control signal such as a digital, and the actuated air flow as a percentage from 0% to 100% air flow passage, where 100% would represent a fully opened damper and 0% a fully perfectly closed damper.

[0018] The actuator may be an electrical-, pneumatic-, thermos-electrical motor, hydraulic, or another electrically controllable motor function in connection with the variable motorized damper.

[0019] Actuator to damper control usually results in non-linear and non-proportional control of the actual air flow in the range as a percentage from 0% as closed to 100% as fully opened. Due to mechanical properties of the actuator to damper control, geometrical shape of damper, and shape of air distribution system, pressure variations because of actuator operation, it is assumed that most actuator to damper control will not result in a linear proportional control of the air flow.

[0020] In a non-linear air flow control system, the magnitude of air flow changes because of a certain change in damper position, depending on where in the damper operational range the change occurs. Such bias errors in prediction of actual air flow may lead to an unbalanced HVAC, VAV or heat exchanged system if many dampers are acting together with the aim to distribute, mix or divert an air flow. Hence, the invention as a method or control system may also provide feedback to a manual input as well as a superordinate system when restrictions in linear proportional controllability is restricted. For example, a superordinate control system or manual input may want to request a closed 20% air flow, due to non-linearity, this is typically not achieved at 20 % damper opening but in another position which is provided by the invention. Such feedback information further enables and eases integration of units and groups of controlled dampers, as multi-damper control units. Likewise, the same function also enables predictable integration of clusters of communicating HVAC, VAV and heat exchange systems.

[0021] Similar results as proposed by the invention could be achieved by adding an air flow measurement device providing feedback to the damper control. However, with the invention a predictable air flow can be achieved without the air flow sensor function. This is especially suitable in systems with dampers where individual flow measurements are hard to achieve or become too costly. Further, due to the non-linear behavior of air flow control, a feedback controller with air flow measurement would need regulate slowly, to avoid unstable control, which is avoided by the proposed invention.

[0022] The invention includes a control system configured to perform the improved method, where the method may be implemented together with control logic, circuits, common linear controllers (P-, PI-, or PID regulators), Filter mechanisms, Field Programmable Arrays (FPGA), software, or at least partially trained artificial intelligence control functions.

[0023] To increase the predictability of a damper with actuator adjustment alone or co-working in a group of inter-dependent damper with actuators, the inventors have identified that it is highly beneficial to translate the non-linear control signal to a linear control signal for controllability, that is precision in control without having to use an air flow sensors and having the problem of slow or unstable control for dampers operating as interconnected units sharing air flow, communicating pressure volumes, and with shared air volumes.

[0024] Pre-linearization of non-linear actuator to damper signals may be carried out on individual actuator-controlled dampers, as well as for complex groups, or units having multiple actuators-controlled dampers. Linearization of complex groups, or units may require several control signals to be coordinated for linearization of complex units with non-linear actuators to damper control.

[0025] The invention further provides a method to identify a linearization map, called the “Scaling map”, that may receive control signals that are linear, converting these signals to nonlinear control signals that counter act the characteristics of the previously mentioned non-linear actuator to damper control signals to air flow control percentage.

[0026] As a result, a user oriented manual input, or a superordinate control system, may via a linear control signal, typically 0-10 V, or a digital air-flow-request-signal, request a true percentage of air flow which, will result in an actual linear proportion of the air flow thru the damper.

[0027] To prepare a non-linear damper air-flow control, the system, or a test engineer at production test, needs to identify the characteristics of the non-linear control curve, including the extremes, fully closed and fully open damper, where each control signal is associated with an actual airflow, typically expressed as a percentage from 0 to 100%. Linear damper movements will result in a non-linear relation between actuator control signal and actual air flow percentage.

[0028] Once the actual non-linear actuator to damper control to air flow is registered, a remapping of actuator control signals can be made by translating a given linear control signal to a corresponding actuator control signal that would set the damper control in a linear corresponding air flow value.

[0029] By mapping the non-linear control signal to actual air flow, from both closed to open, and from open to closed, any play between the mechanics may be detected.

[0030] It is even possible to implement two linearization functions, one for a rising signal and one when dropping the signal, of improved precision.

[0031] A first mapping of the non-linear control characteristics will normally be made during design for the type of model, and then during manufacturing to identify the non-linear characteristics of the damper unit to air flow. It may even be possible to investigate any remaining or new nonlinearities between actuator to damper to air flow control function, after installation to compensate for damages or alterations of the damper mechanism during installation and after a while as a result from wear and tear, as well as changes in the complete air damper ventilation system.

[0032] By first measuring the original non-linear relation between a control signal from 0 to 10 [V], or 0 to 100%, with a measured air flow as a percentage, one may find a translation function, stored in the Scaling Map function, that would translate a request for a 30% air flow, to a control signal to the actuator and damper that results in a true 30% air flow, and then send this new control signal to the actuator as a new linearized control signal.

[0033] Hence, a manual input, or a superordinate control system would view each damper as a perfectly linear controlled damper, providing improved controllability for an open-loop control system. The Scaling Map is calibrated to linearize the actuator to damper and air flow control, which reduces the need for feed-back control of air flow together with associated sensors and avoiding potential problems regarding speed or instabilities.

[0034] The Scaling Map calibration process is applicable to groups of dampers and damper units having more than one sections that each may be separately controlled by an individual damper actuator, alone or in combination with an at least one bypass-damper that may or may not be actuator controlled. As a damper unit having multiple dampers the scaling map can be used to exercise an experienced linear air flow control in an open loop control system, to improve controllability precision, predictability, and response in HVAC, VAV and heat exchange systems. For a complex damper unit, or similar actuated damper group, the linearization may use more than one control variable or control signal to operate the actuated dampers.

[0035] The Scaling map can be implemented in software as a mathematical function that will make it possible to control a non-linear damper actuator to air flow arrangement using a linear signal, that corresponds to a linear air flow setting. It may also be implemented as an adaptive filter, a P, PI, PID regulator function, in hardware, digital or analog electronics.

[0036] To update the Scaling map with linearization correction data, each individual, type of, unit, or group of dampers with valve actuators can be analyzed by using correction data: a) measured and calibrated in a test lab before production, b) collected during production tests after final assembly, c) synthesized by flow dynamics simulations based on proportions, air flow and pressure range, as well as geometries of damper and duct, or, d) collected during adaptive test, adapting or a learning procedure, after installation of a HVAC, VAV and heat exchanger system provided with air flow sensors, pressure sensors, or differential pressure sensors over each damper, a damper unit, or group of dampers.

[0037] The non-linear air flow characteristics of each actuator with damper, may be transferred to the new control system, as correction data in a separate data file, stored for retrieval from the linearization calibrated or measured actuator with damper unit, or stored in a central cloud information system, providing characteristics for all known actuator with damper unit types as well as individual actuator with damper.

[0038] Groups, and units of actuators with air dampers may be measured to update an actual Scaling map controlling multiple actuators with independent dampers, in a similar way, as for single actuator to damper and air flow control. This provides a method to linearize, and control complex actuator-damper-air flow units and groups dampers, using linear control signals.

[0039] Technical problem

[0040] How can we make an open-loop control system for HVAC, VAV, AC systems, chilled beams, plenum chambers with actuated face-dampers and with or without bypass-dampers more predictable with improved precision? A challenge is that without a feedback loop, it is challenging to predict errors that occur due to imperfections, individual irregularities, and nonlinearity of actuated damper control of air flow. Also, due to the non-linear behavior of damper air flow control, any feedback controller needs to be adjusted to avoid instabilities, which can result in low speed. Also, any feedback controller requires an air flow measurement device, which might be hard to implement and costly, especially in complex damper systems. The invention presented proves that it is possible to increase precision and predictability in air flow control even in an open loop control system, to achieve similar performance as a feedback controller, but also in a more direct way.

[0041] When controlling air-dampers, it is not the position of the damper that is of importance but its influence on the air flow.

[0042] Air-dampers can be controlled by open-loop or closed-loop controllers, and in both cases, the system characteristics, that is the relationship between damper opening and air flow, is a critical aspect for the ability to design a feasible control loop. Ideally, a completely linear system characteristics is desirable, since then, a certain controller action will result in the same response in air flow over the whole operational range of the damper.

[0043] In the open-loop case, a linear system characteristic means that the outcome of a control action can be predicted, which is desirable since an open-loop control has no surveillance on the controlled outcome and hence has no possibility to make corrective actions.

[0044] In the closed-loop case, a linear system characteristic means that a common linear controller (such as P, PI or PID) can perform equally well in the whole operational range of the air damper, since both steep regions where instability is imminent and flat regions where slow response is imminent are avoided.

[0045] In the event of incorrect air flow in a closed-loop case, there is a risk that the outcome of the controller action does not become as intended, with various problems to follow, such as thermal, fluid mechanical etc.

[0046] In complex damper units having multiple individually actuated face-dampers for each section, and at least one actuated bypass-damper, it is more challenging to determine the air flow when one or many actuators regulates each damper connected with a shared air volume. Being able to predictably control the air flow of complex damper units based on a linear predictable control signal, would enable open-loop air flow regulation.

[0047] In essence, being able to control and predict the air flow in an open-loop air flow regulation system in an HVAC, VAV and heat exchanger would increase the precision of the control system and enable cost savings.

[0048] Solution to problem

[0049] The proposed invention comprises of a software filter, here referred to as the ’’scaling map” in the following text, between controller output and actual damper position which results in a linear relationship between controller output and relative air flow through the damper, that is the system characteristics. The fact that the invention is designed as a software filter is to have the ability to update common controllers by manipulating their output before the control signal is sent to the damper actuator that eventually achieves the damper movement.

[0050] The Scaling Map has been developed after numerous measurement experiments set ups to determine the characteristics of common air-dampers under different operation circumstances, or conditions, such as using a varied number of dampers, different sets of static air pressures, and differential pressure variations to estimate pressure drop and air flow.

[0051] The inventors have found that the measurement results were close to generic and only changed marginally for the different conditions that were tested, which means that also the resulting scaling map becomes close to generic. The Scaling Map usually does not have to be compensated for static pressure differences, differential pressure differences, and for a variation of the number of dampers controlled. The same Scaling Map may then be used to linearize actuated dampers with non-linear proportions between the actuator control signal and the resulting change in relative air flow.

[0052] An easy way to linearize a non-linear function is to identify its inverse function. Another way is to measure the control signal with resulting air flow signals for each step from 0 to 10 [V] or 0 to 100% requested relative air flow, as a table. Then to operate the non-linear actuated damper to produce an air flow of a certain rate, can be made by reversely looking up the earlier recorder Scaling Map, by first receiving the control signal indicating a desired air flow, and then looking up which non-linear control signal in the Scaling Map that generated the desired air flow, and using the looked up non-linear control signal to operate the non-linear actuated damper, thus achieving a linear control of the actuated damper. The HVAC control system may then request a specific air flow to be delivered in a forward open loop control system with no feedback sensors. The system then controls the actuated damper in such a way that a linearly proportional air flow is generated over the previously non-linear actuated damper range.

[0053] Further, in open-loop systems, the controller output is directly dependent on the controller input. The input is typically some system disturbance that needs to be acted upon, and the output is the corresponding action. For example, the input can be OAT and the output can be the amount of air that passes through a bypass to avoid freezing on a heat exchanger which is caused by low OAT. As the OAT is decreased, more air should pass through the bypass. However, with a non-linear damper system, the controller action depends on where in the damper operational range the system is residing, which means that different actions can be expected for the same change in OAT. With a linear damper system, the same change in OAT is always met by the same change in air flow through the bypass.

[0054] Hence as an advantage, the previous more complicated-to-control non-linear actuated damper to air flow control is made more predictable to be controlled by making sure that the same action to a certain system disturbance can be expected. As the control system 20 may further verify its operation during linearized control, it may further determine its limitations in control range and any deviations from a linear actuated damper to relative air flow response. In case the deviation fluctuates or exceeds a reference value, the system may once more improve its Scaling Map to acquire a sufficient or improved precision.

[0055] In Fig. 1 , diagram in the Scaling Map 60, the dotted line, the NAF 61 curve represents the normal change in air flow as a common damper is controlled from fully open to fully closed position. As can be seen, there is a large ’’dead region” between 7 and 10 [V], that is 70 - 100 % open, where the control action does not influence the air flow at all. Then, there is a slow region between 5 and 7 [V], that is 50 - 70 % open, where the influence of damper position on the air flow is marginal. Then, there is an active region between 3.5 and 5 V, that is 35 % - 50 % open, where the response in air flow is feasible. Finally, in the highly responsive region between 0 and 3.5 [V], that is 0 % - 35 % open, where the change of air flow is rapid, which means that accurate control is difficult, as the control signal results in large increments in adjusted air flow. Having a linear control of the actuated damper where the air flow is predictable and linear means that the air flow can be adjusted with precise proportional steps.

[0056] In the figure Fig 1 , the influence of the scaling map can also be seen. The result of the scaling map is represented by the continuous line AFWSM 62 curve, where a proportional change in damper position results in the same proportional change in air flow over the entire operational range. Or at least in the operation range that can be controlled as a linear air flow control function, due to limitations such as air leakages and insufficient possibilities to fully open a damper.

[0057] In practice, this is achieved due to the non-linear characteristics of the scaling map, which results in large damper steps in the upper range of the damper positions, and smaller and smaller steps as the closed position is approaching.

[0058] Linearization of system characteristics is commonly used in the technology domain hydronic controls. Control valves for heating are often designed to have equal-percentage- modified (EQM) characteristics, which is usually an exponential characteristic. By anticipating a logarithmic characteristic of the unit that is intended to emit heat, the combined characteristics of an EQM valve and the heat emitting unit becomes linear.

[0059] Technical effect

[0060] The technical effect is to provide an improvement in air flow control precision, as result from the improved predictability of a linear control signal resulting in a linear proportional relative air flow over each face-damper (32, 34, 36, 38) and bypass-damper (79) in an open-loop control system.

[0061] The invention while also may support closed-loop control, improves precision in mainly open-loop air flow control systems comprising at least one non-linear damper, especially when non-linear dampers are arranged together, as damper units having multiple dampers, and groups of dampers. By transforming linear control signals using a Scaling Map into signals that will resolve the non-linear actuated damper characteristics into a linear proportional relative air flow, the precision of the open-loop control system can be increased.

[0062] Linearization is made possible through a measurement phase, where the non-linear actuated damper is analyzed and the control signal to actual non-linear air flow rate is mapped and stored in a Scaling Map. The Scaling Map then calculates a function that rectifies the nonlinear air flow control to a linear controlled and proportional relative air flow, independent of air pressure.

[0063] When air flows through a damper, the resistance created by the damper changes the pressure of the air. The pressure difference before and after the damper is known as the differential pressure. The magnitude of the differential pressure is directly proportional to the resistance of the damper, and inversely proportional to the air flow rate.

[0064] Differential pressure = Resistance * FlowA2

[0065] As the air flow rate increases, the differential pressure across the damper also increases. This is because the air molecules are moving faster and exerting more force on the damper, resulting in a greater resistance and pressure drop across the damper.

[0066] Conversely, as the air flow rate decreases, the differential pressure across the damper decreases as well. This is because the air molecules are moving slower and exerting less force on the damper, resulting in a lower resistance and pressure drop across the damper.

[0067] Therefore, the relationship between air flow and differential pressure across a damper is a direct relationship, meaning that as one variable increases, the other variable also increases, and as one variable decreases, the other variable also decreases, but not the relative flow which is the relevant quantity for the invention.

[0068] Advantageous effects of invention

[0069] There are several differences between the proposed invention and the approach used for control valves. In the case of control valves for heating control, the previously mentioned EQM-characteristics are traditionally achieved by mechanical design of the flow controlling part of the valve. Further, in this case, the characteristics are designed according to an anticipated influence on system characteristics, and not the real measured influence. Finally, the technique is not known to be used on air dampers, likely since compressible air is considered less problematic and more forgiving to control as a flow compared to non-compressible liquid.

[0070] The invention differs from the patent document EP0321515B1 in that EP0321515B1 does not have a learning function. The learning function in the invention can use neural networks and more basic methods to learn how learning and linearization should be done. The most basic method is to measure air flow characteristics on an actuated damper in a test lab. One may also measure air flow characteristics for a specific individual actuated damper unit, for increased precision of the scaling map, where individual actuated damper units may demonstrate small variations, in sensitive regulation range such as for leakage when the damper is fully closed, as well as due to imperfections when fully open, and mechanical variations in mechanical proportions during assembly, or due to minor damages to the actuated damper. Also, wear and adjustments made during maintenance can be compensated according to the invention proposed.

[0071] The Scaling Map may make use of a relative humidity sensor, or weather information source for relative humidity RH%, for usage as a coefficient for selection of an alternative linearization curve, to further compensate for non-linearity of the actuated damper's air flow control for variations in RH%. EP0321515B1 does not teach that a sensor for humidity RH% can be used to improve precision of linearization of non-linear actuated dampers. Pressure sensors may also be used beyond a plenum chamber or bypass-damper pressure sensor, but the inventor has discovered that variations in air pressure has less influence on the linearization of the non-linear actuated damper's using for example an inverse function to the non-linear actuator control to air flow characteristics.

[0072] Extra pressure sensors sensing the pressure after each damper, especially facedampers and possibly also after by-pass damper at its downstream flow to for example a heating or air-conditioning system, can be used during measurement of the damper unit, at a test lab, or as a quality inspection test at production, to determine the product's non-linear actuator to damper to air flow regulation characteristics. Where these non-linear characteristics may then later be used to rectify non-linearity in an installation setting. Also, if individual distribution among damper units of the same type is identified, then production quality tests can measure each individual damper unit having a unique identity, and store the non-linearity characteristics, or inverse function rectifying or transforming a linear control signal to an inverse function to the non-linear actuated damper to air flow characteristic, to achieve a perfectly linear control curve for the actuated damper unit type, or even an individual actuated damper unit, but the inventor has discovered that variations in number of dampers in a damper unit has less influence on the scaling map, that is the relative flow through an individual damper.

[0073] Measured type, or individual non-linear damper characteristics, as well as transformation functions that rectifies any non-linearity of air flow control, as well as Scaling Map tables can be stored on and provided with the damper unit for installation, or transferred as data parameters describing the type of damper, as well as individual damper characteristics, to the control system 20, and hardware, software, data 80 operating the specific installation of the actuated damper unit. These characteristics and transformation data needed to linearize the air flow control for actuated damper units, can be provided via a cloud database 90, over a wired or wireless communication interface 91 , or as a persistent memory, a persistent memory that can be stored in the damper actuator interface for information retrieval at installation.

[0074] Furthermore, the invention can also handle complex dampers / dampers with multiple face-dampers, one per ventilation section, and a bypass-damper sharing the same air stream. Then, the main benefit is that the scaling map can be derived for on one damper in the set and then be implemented individually on the remaining in the same set. Then, the great benefit is that the how the air is divided over the dampers in the set can be predicted, which is information that is beneficial for the control and that otherwise would been hard to retain since individual flow sensor over each damper might be impossible or costly to implement.

[0075] A more complex model of linearization can be recreated by measurement, which considers each individual actuator. For example, the control system can achieve a 30% flow by opening a damper for one section to 80% and zero flow for remaining dampers. The same relative air flow can be achieved if each damper opens, for example 18% of fully opened, while the relative air flow for an individual damper remains unchanged. Why then should not every damper be opened to, say, 20% for a given situation? This is likely due to flow dynamics and turbulence, as well as how local pressure distributes in the damper structure. It is therefore reasonable that different damper sections can affect each other. Furthermore, the bypassdamper can affect others as it can provide a local pressure reduction and vice versa for the damper openings of the sections.

[0076] The invention considers controllability in the linear range, as well as the special information about whether a damper closes tightly or opens completely. This is information that is not handled in EP0321515. Nor is it handled that the mechanism may have gaps / play when moving in different directions, which may mean that one would want to measure both the opening and closing procedure, but this special case should have low value for the invention.

[0077] The invention also considers systems of dampers and combination of face-dampers and bypass-dampers while EP0321515 focuses on individual dampers. Hence, with the proposed invention, it can be predicted how the air is split / mixed between a set of dampers that shares the same air stream. This is hard to achieve in other ways since flow sensors might be impossible or very costly to implement.

[0078] The invention also proposes a pure electronical solution, while EP0321515 focuses on mechanical.

[0079] The scaling map was initially designed for modulating open-loop control of dampers for defrost of plate-heat-exchangers (PHE). In this application, the benefit is large, since it is important to be able to predict the air flow through the plate, and hence the defrost action, to avoid frost on the plates. In this application, dampers are commanded to close as a function of outdoor-air-temperature which is used as an indication of defrost demand. As the damper closes, less cold air passes through the PHE, and hence the efficiency of the PHE is reduced, and any frost is allowed to melt.

[0080] Without the scaling map, when an initial defrost demand is commanded, that is as the dampers start to close from fully open position, the damper moves into the previously described ’’dead-region’’. Hence, the defrost action has no influence on the air flow though the PHE which means that frost could start building up. A similar behavior could also be expected in the previously described ’’slow region” since the defrost actions, that is by closing dampers, have little influence on the actual air flow. The damper must move down to a 50 % opening until a sufficient relation between air flow and increased defrost demand is achieved. Further down, in the highly responsive region, the relation between increased defrost demand and air flow is too large, which means that the influence on efficiency can be more than expected. Naturally, when applying the recreated Scaling Map, all these regions are replaced with one active region, where the relation between increased defrost demand and the resulting change in air flow is constant and sufficient.

[0081] There are several other possible applications in which the proposed invention can be beneficial, such as mixing chambers, VAV-diffusers, etc. However, it is important to note that for each new damper design (and perhaps application as well) the scaling map needs to be updated since the mechanical aspect of the system needs to be accounted for. Hence, the potential patent will be referring to the method used to linearize the system characteristics in damper applications.

[0082] Also, new designs of actuated dampers could have new non-linear not yet anticipated air flow control characteristics. Having a control system that can analyze, test, measure, adapt and learn new non-linear actuated damper air flow control characteristics would be a big advantage as air flow configurations may have to be updated due to maintenance requirements, and changes in a building use, layout or modernizations required.

[0083] The invention provided can adapt to control new actuated damper air flow control units, as well as more complex installations, by providing access to non-linear characteristics via cloud data 90, data provided with a new actuated damper, potentially retrievable from the actuated damper itself. Test data and measurements performed in a laboratory may also be provided at the point for installation as machine readable or human readable information, to pre-configure the system to handle non-linear actuated damper air flow control units.

[0084] The proposed scaling map solution has been tested in a lab environment and has been found to be ready for distribution as field tests. As a first step, the scaling map can be implemented as software updates in existing installation units, to remedy units that have been recognized to have history of problems related to defrosting. The mechanism of the scaling map may of course be provided with all new actuated damper units, preferably with test data on nonlinearity from high-precision test lab set ups.

[0085] Other features and advantages of the disclosed embodiments will appear from the following detailed disclosure, from the attached dependent claims as well as from the drawings. Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a / an / the [element, device, component, means, step, etc.]" are to be interpreted openly as referring to at least one instance of the element, device, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.

[0086] The invention concerns an open-loop control system for managing non-linear air dampers in HVAC, VAV, and heat exchange systems. Traditional damper systems show nonlinear characteristics between actuator control signals (typically 0-10 V or 0-100% digital signals) and the resulting air flow through the damper. This non-linearity makes precise air flow control difficult, especially in multi-damper units or interconnected dampers sharing a common pressure volume. To address this, the invention introduces a "Scaling Map", a software-based linearization function that transforms a linear control signal into a non-linear signal that corrects for the physical non-linearity of the damper-actuator-airflow system. The result is that air flow changes proportionally to the control signal across the full range — a virtual linear damper response.

[0087] In existing systems, damper response varies significantly over the control range:

[0088] Some voltage ranges cause no airflow change (“dead regions”).

[0089] Others lead to overly aggressive changes (too responsive).

[0090] This variability hampers predictable control and leads to inefficiencies, especially when coordinating multiple dampers.

[0091] Feedback-based systems with air flow sensors can mitigate some issues but are costly, complex, and slow. The invention offers a sensorless, open-loop alternative by mapping a damper's non-linear behavior and translating linear commands into the correct non-linear control inputs.

[0092] The Scaling Map (also referred to as a linearization map) is the core innovation. It is a pre-calibrated or adaptive translation function between a desired air flow (linear signal) and the actual actuator signal needed to achieve that flow through a non-linear damper.

[0093] Key features:

[0094] Supports single dampers and multi-damper units (e.g., face and bypass dampers in the same unit).

[0095] Avoids need for airflow or pressure sensors in most applications.

[0096] Can be implemented in software, firmware, or hardware, and may use lookup tables, inverse functions, or adaptive filters.

[0097] Accounts for mechanical play, aging, and installation variability by allowing updates or verification steps.

[0098] The invention includes three core operational steps: a) LEARNING (S2000) where the non-linear damper is operated across its full range, the system measures actual air flow or pressure proxy vs control input and non-linear actuator- to-flow relationships (RAF vs ACS) are recorded. b) ADAPTING (S3000) where a Scaling Map is calculated from recorded data, where the map acts as a translator from linear control signals to nonlinear damper commands. This can be done locally or via cloud service using pre-calibrated data or machine learning. c) LINEAR OPERATION (S5000) where the control system now sends linear requests for airflow, e.g., 30% and the control unit uses the Scaling Map to translate that into the correct actuator command which achieves linearity in air flow response, enabling more predictable and precise control.

[0099] Optional steps include:

[0100] VERIFYING (S4000): Ensures Scaling Map accuracy post-installation.

[0101] FALLBACK (S1000 / S1100): Reverts to non-linear mode if the map fails verification.

[0102] Closed-loop support (S1200): Bypass damper regulation using sensor input for defrost or pressure maintenance.

[0103] The teachings herein provide several advantages. Providing predictable air flow in open-loop HVAC systems, where feedback sensors may be impractical.

[0104] Reduces commissioning time and cost by enabling pre-calibration.

[0105] Supports multi-damper coordination — can control multiple dampers as a logical unit.

[0106] Enhances system stability and responsiveness, especially in defrosting, mixing, or bypass scenarios.

[0107] Scaling Map may be derived from: Lab testing, Simulation, Field learning and Cloudbased historical data

[0108] The scaling Map can adjust for: Temperature, Humidity and Damper-specific deviations.

[0109] The teachings herein are applicable in devices such as: Plate Heat Exchangers (PHE), VAV diffusers, Plenum boxes and Chilled beams.

[0110] The proposed invention expands on earlier efforts like EP0321515 B1 in several ways, namely in that it supports multi-damper units, not just single dampers, incorporates learning and adaptive scaling vs static inverse functions, can use humidity sensors and Al-based learning, is designed for software implementation, allowing updates and remote configuration, and enables cloud-based distribution of calibration data and scaling maps for standardization and rapid deployment.

[0111] The invention significantly enhances air flow control precision in HVAC and related systems, particularly for non-linear dampers in open-loop environments. By employing a Scaling Map, damper systems behave as if they were linearly actuated, eliminating the need for complex sensor networks or feedback loops. This enables efficient control of both individual and grouped dampers in a variety of use cases, from commercial HVAC systems to specialized applications like defrost control in plate heat exchangers.

[0112] BRIEF DESCRIPTION OF DRAWINGS

[0113] The invention is described, by way of example, with reference to the accompanying drawings, which follows:

[0114] Fig. 1 , is a state diagram over method steps and states for the air flow control methods.

[0115] Fig. 2, visualizes the complete air control system 20, where in the air flow controller 80 is a subordinate-system.

[0116] Fig. 3 to 8. illustrates methods for air flow control, with method steps / states according to Fig. 1 . Fig. 3 illustrates the air control system 20 when in operation state=NO SCALING MAP, which also realizes the method steps S1000 with sub-step S1100 operation state=Open loop nonlinear.

[0117] Fig. 4 illustrates the air control system 20 when in operation state=LEARNING, which realizes the method steps S2000 Register Normal air flow control.

[0118] Fig. 5 illustrates the air control system 20 when in operation state=ADAPTING, which realizes the method steps S3000 Determining scaling map

[0119] Fig. 6 illustrates the air control system 20 when in operation state=VERIFYING, which also realizes the method steps S4000 Verifying scaling map at full range

[0120] Fig. 7 illustrates the air control system 20 when in operation state=LINEAR OPERATION, which also realizes the method steps S5000 Operating using scaling map at validated range. Fig. 8 illustrates the air control system 20 when in operation state=CLOSED LOOP, which also realizes the method steps S1200 Closed loop constant flow.

[0121] Fig. 9 depicts an air flow controller 80.

[0122] Fig. 10 illustrates non-linearity between a control signal ACS 63 for actuating a face- 32, 34, 36, 38 or bypass damper 52, with resulting non-linear relative air flow (RAF 64) curve 61 .

[0123] Fig 11 is the Scaling Map 60, showing the interior of a Scaling Map 60 with a hypothetical example of a minimalistic table representing actuator signals ACS 63 with related RAF relative air flow 64’ as a mapping, and the latest measured RAF as indicating deviations 64, where the table is associated with an identity ID 101.

[0124] DETAILED DESCRIPTION

[0125] The disclosed embodiments will now be described more fully hereinafter with reference to the accompanying drawings, in which certain embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Like reference numbers refer to like elements throughout.

[0126] It should be noted that even though the description given herein will be focused on a method for an air flow control system, for an air flow controller, including such air flow control system and air flow controller, the teachings herein may also be applied to usage in an HVAC system or, Air Handling Unit.

[0127] Detailed description of figures

[0128] Fig 1 . shows the method steps with dashed rectangles indicating optional method states, and black line rectangles indicates mandatory method steps, realizing a method to learn and compensate for non-linear actuated damper air flow characteristics.

[0129] Operation states of the new control system for improved response of air flow control using dampers in HVAC, VAV (baffle or plenum chambers) or heat exchange systems 20, typically executed as control logic 80, by a computer 80 using a software, FPGA, or electronic circuits 80. The state transitions from a mandatory LEARNING state S2000 to ADAPTING S3000, and later to an operational state LINEAR OPERATION S5000, where the control signals from the control system uses linear proportional control signals, determined by the Scaling Map function 60. Optional states are to use NO SCALING MAP S1000 which constitutes the prior art nonlinear control mode, and VERIFYING S4000, a state verifying and potentially improving the Scaling map as well as determining any deviations from the linear control function. During the VERIFYING, Verifying Scaling map at full range, the system uses the Scaling map 60 and linear control signals from the control system 27 and 47, that are transformed into rectifying inverse functions that linearizes the previously non-linear characteristics of the control signals 24, 44 to the damper actuators 25, 45 that resulted in a non-linear control of the air flows 72, 74, 76, 78, and 79.

[0130] Fig. 2 is a presentation of the complete control system 20 with typically non-linear actuated face-dampers, and bypass-damper. Non-linear means that a control signal is not directly proportional to an expected percentage of an air flow 70. The pressure can be measured using the air flow, air pressure, or differential air pressure sensor 66, where the air pressure of the air flow 70 entering damper units that is face-dampers and bypass-damper(s), can be measured. The differential air pressure sensor may measure the air pressure over an air fan 71 driving the air flow 70. The differential air pressure sensor may measure the air pressure over a heat exchanger unit that would also provide indications pressure variations in the damper unit air flow 70.

[0131] In an alternative embodiment, the differential pressure sensor may be achieved by measuring current drawn by an electric motor powering the air fan 71 . In this alternative way to measure or rather estimate air pressure, the control system senses the current drawn by the air fan's 70 electrical motor at a certain RPM. Given characteristics of the air fan's ability to transfer air volume and build up air pressure at certain motor speeds RPM and currents, at defined power levels (voltage and current) and currents, the control unit can calculate and estimate the differential pressure over such an electrically powered air fan 71.

[0132] The Scaling map, is designed to support the method steps of identifying, learning, calculating compensation for control signal 28, that compensates for non-linear characteristics between the face-damper actuator control signal 24, as well as the bypass-actuator control signal 44, to the actual outlet air flow from face-dampers 72, 74, 76, 78, and bypass-damper outlet air flow 79.

[0133] Fig 3. shows an optional state=”No scaling map”, S1000, where No scaling map is being used. This means that the system operates in a traditional open-control loop with nonlinear actuated damper to air flow characteristics. Essentially a state that would run a control system based in a basic mode with normal precision in air flow control.

[0134] Fig 4. shows LEARNING an operational state=”Registering Normal non-linear air flow control”, S2000, where the control system operates records in a learning mode, the relationship between the input control signal ACS for an actuated damper resulting in a non-linear change in relative air flow 72, 74, 76, 78, and 79.

[0135] Scaling map data may be interchanged via a cloud data memory 90 via cloud interface 91 for networked coordination with cloud stored reference linearization data for the type of actuator damper 32, 34, 36, 38, 51 , unit 30, 50, or group 100, or for the specific individual actuated damper, unit, or group.

[0136] Fig 5. shows ADAPTING an operational state=”Determining Scaling Map”, S3000, where the control system calculates and determines a compensation function or map that corresponds to a Compensated Linearized Actuator Control Signal (CLACS ), that would control the non-linear actuated damper to air flow as a linear change of air flow as percentage.

[0137] Scaling map data may be interchanged via a cloud data memory 90

[0138] Fig 6. shows VERIFYING an optional operational state=” Verifying Scaling map at full range”, S4000, where the control system test runs using the Scaling map, working as a linearization function, and evaluates the linearity result.

[0139] Scaling map data may be interchanged via a cloud data memory 90

[0140] Fig 7. shows LINEAR OPERATION an operational state=” Operating using Scaling map at validated range”, S5000, where the actuator control signal is linearized by the adapted Scaling Map. All control signals sent from the control system are linear control signals, and the Scaling map translates and compensates for any non-linearity of the actuated damper control of air flow. Scaling map data may be interchanged via a cloud data memory 90.

[0141] Fig 8. shows an operational state- ’Closed loop constant flow”, S1200, wherein the Closed-loop control of bypass-damper 69 controls the Bypass-actor 45 to maintain a constant air flow over the bypass damper all the time, as measured by the air flow, pressure or differential pressure sensor 66. Face-damper air flow and differential pressure may also be monitored to identify freezing and ice at the face-dampers. As a result, all face-dampers can be individually controlled and actuated by a Manual input, or air flow control system 21 , or air flow controller 80, or an external controller, while maintaining a constant pressure over the crossflow plate heat exchanger 73, as preferably measured by differential sensors 66’, 66”.

[0142] Fig 9. illustrates an Air flow controller 80 with interfaces to sensors, actuators 22, 28, 20, 44, 48, 49, 66’, 66” and operating mode selection 21 connected, in a context for an air handling unit or HVAC system. The scaling map 60 may be intrinsic inside the Air flow controller (60), and / or extrinsic provided as a cloud service to the Air flow controller.

[0143] Fig 10. shows basic system without the linearization function, with the control signal ACS 63, resulting in non-linear relative air flow RAF 61 relation as depicted in the measured curve 61 . Due to characteristics and pressure variations over of the damper units, including face-dampers and bypass-damper(s), the relative air flow RAF 61 will not be linearly proportional to the control signal ACS 63 as seen in curve 61 , whereas the invention enables the system to be controlled using a proportional compensated control signal 62, where a demanded relative air flow can be requested with predictable results. Even a linearized system may have deficiencies in fading out non-linearities, the curvature 62 is not perfectly straight.

[0144] As a result of the non-linearity and non-proportional relation between control signal to the actuator and the desired regulation of the relative air flow thru the face-dampers and bypass-damper(s), four control zones 201 , 202, 203, and 204 are recognized to provide different feasibility for direct control of the dampers via actuators. The four control regions can be separated in the following regions with different challenges and usage:

[0145] Unused region 201 (too responsive). Very limited regulation of the actuator control signal results in a too big change in air flow response, which makes it hard to control the air flow with precision.

[0146] Active region 202. Feasible and good response between actuator control signal and relative air flow, making this regulation range very feasible for precision control of air flow.

[0147] Slow region 203. Slow response between actuator control signal and relative air flow.

[0148] Dead region 204. The actuator control signals have very little impact on adjustments of the air flow. This region is difficult to control as a several steps are required in the actuator control to get just a little change in relative air flow.

[0149] The control method comprises as step S2000 LEARNING, “Registering Normal nonlinear air flow control” where it discovers and registers the control range of actuated facedampers 31 and bypass-damper 51 . Next step is S3000 ADAPTING “Determining Scaling Map” where the Scaling Map 60, working as a linearization function 61 , 62, 67 is determined, calculated, and updated. Step S3200 will then “generate a scaling map 60 that linearizes actuator control signals”.

[0150] Fig 11 . illustrates a simplified table in the Scaling Map 60 The usage of the Scaling Map is to transform requested air flow signals 64’ to actual actuator control signals 63 which are linearized by the adapted Scaling Map.

[0151] All control signals sent from the control system are linear control signals, and the Scaling map translates and compensates for any non-linearity of the actuated damper control of air flow. Scaling map data may be interchanged via a cloud data memory 90.

[0152] A scaling map 60 which is needed to achieve the step S5000 LINEAR OPERATION with operational state=” Operating using Scaling map at validated range”. In step S5000 the actuator control signal is linearized by the adapted Scaling Map. All control signals sent from the control system are linear control signals, and the Scaling map translates and compensates for any non-linearity of the actuated damper control of air flow. Hence, the control system can control the actuator with a signal requesting a specific relative air flow, which is then translated by the Scaling map 60 into a likely non-linear actuator signal for the face-dampers, and bypassdamper.

[0153] To distribute Scaling Maps 60 data between installations, and to make commissioning of new systems quick and efficient, the Scaling map data can in some aspects of the invention be interchanged via a cloud data memory 90, where an identity 101 refers to: articles, production series, or individual damper groups 100, damper units 30, 50, and face dampers 32, 34, 36, 38 as well as bypass 52 dampers with or without related actuators. Thus, letting a calibrated reference installation set a preferable initial Scaling Map for now identical or very similar installations. Statistical analysis of a large set of systems exchanging Scaling Maps can also be a good source for detection of anomalies, errors, and performance issues in existing installations.

[0154] A first method for the control system 20

[0155] The present invention is a method designed to enhance the precision of air flow control in HVAC, VAV, and heat exchanger control systems that utilize open-control loop regulation and non-linear actuated dampers for air flow regulation, as defined in claim 1 . Also, the invention can be beneficial in systems with closed-loop controllers as well since values of proportional gain (P), integral time (I) and derivative time (D) that are optimum in the whole operational range of the damper can be set.

[0156] The method involves at the following least three steps: S2000 LEARNING, S3000 ADAPTING with sub steps S3100 and S3200, and S5000 LINEAR OPERATION.

[0157] Step S2000 LEARNING, where the method first recognizes typically a non-linear, or non-proportional, characteristics between the damper actuator signal in relation to the actual air flow through an actuated damper.

[0158] Step S3000 ADAPTING, where the method identifies and determines a transformation function, called the Scaling Map, that compensates for non-linearities between actuator signal controlling the blade, or similar air flow regulating mechanism, of an individual damper and resulting relative air flow achieved. As a result, the Scaling Map transforms a linear proportional control signal to a compensating actuator signal that results in a linear proportional relative air flow percentage [%] through the actual damper, being controlled.

[0159] Step S5000 LINEAR OPERATION, is a method step where the system, operates using linear proportional relationship between actuator signal in Volt or percentage [%], and resulting relative linearly proportional relative air flow, over the actuated damper, because of the rescaling of the control signal to the actuator signal, using the linearization function in the Scaling Map.

[0160] Each step is further described in detail as follows.

[0161] Step S2000 LEARNING:

[0162] In step Learning, the method is registering data obtained from the open-control loop regulation 23, 43 of actuated dampers 32, 34, 36, 38, 52 with non-linear air flow control 72, 74, 76, 78, 79 characteristics. This data is then registered in relation to the inlet air flow 70 and the corresponding actuator control signals 24, 44.

[0163] The non-linear characteristics, represented as Relative Air-Flow (RAF) [%] 61 and Actuator Control Signal (ACS) [V] as voltage or a digital signal 27, 47 for each actuated damper 32, 34, 36, 38, 52, actuated damper unit 30, 50, or actuated damper groups 100, are saved.

[0164] Preferably, this data is stored in the Scaling Map 60 or as cloud data 91 , or it can be saved directly in the actuators 25, 45 for easy retrieval, at a later step.

[0165] The step S2000 LEARNING is executed by: controlling a at least one damper with increments over range from fully closed to fully open while other sequences including random sequences are possible while, registering data from open-control loop regulation 23, 43 of actuated 25, 45 at least one face-damper 32, 34, 36, 38 and / or bypass-damper 52) with non-linear air flow control 72, 74, 76, 78, 79 characteristics, in relation to an inlet 70 or outlet air flow 72, 74, 76, 78, 79, with actuator control signals 24, 44, and saving the non-linear characteristics RAF[%] 61 and ACS [V] 27, 47 for the at least one actuated face-damper 32, 34, 36, 38 and / or bypass-damper 52 in actuated damper unit 30, 50 or actuated damper groups100, and saving in the Scaling Map 60 in the Air flow controller's 80 memory, as cloud data 91 , or in the actuator 25, 45 memory, for later for retrieval.

[0166] Step S3000 ADAPTING:

[0167] During this step, the method is retrieving non-linear characteristics RAF[%] 61 and ACS [V] 27, 47 for each actuated damper or for one damper in the set, if they are geometrically similar. The result from an individual damper can then be applied to the whole set. The method is also determining a linearization scaling map 60 for linear control of each actuated face-damper and bypass-damper air flow by the sub-step S3100 , or one common if the dampers are geometrically similar.

[0168] In sub-step S3100, the method is calculating a linearization map based on earlier recorded, or retrieved, normal air flow RAF [%] / ACS [V] plotted data NAF 61 .

[0169] In the following sub-step S3200 a scaling map 60 is generated such that it linearizes the actuated damper air flow control signal 28, 48. Such a linearization may for example be carried out by mapping a linearized actuator signal look up table with linear actuator signal values directly proportional to actuated damper air flow values, but more sophisticated methods exist, as will be mentioned later.

[0170] The S3000 ADAPTING is executed by:

[0171] Retrieving non-linear characteristics RAF[%] 61 and ACS [V] 27, 47 for at least one actuated damper32, 34, 36, 38, 52, and determining a linearization scaling map 60 for linear control of at least one actuated face-damper 32, 34, 36, 38 and bypass-damper 52 air flow 70, 72, 74, 76, 78, 79 by,

[0172] (S3100) Calculating a linearization map based on earlier recorded, or retrieved, Normal air flow RAF[%] / ACS[V] plotted data NAF 61 , and

[0173] (S3200) Generating a scaling map 60 that linearizes the actuated damper air flow control signal 28, 48, for example by mapping a linearized actuator signal look up table with linear actuator signal values directly proportional to actuated damper relative air flow values.

[0174] S5000 LINEAR OPERATION:

[0175] In this step, the method is operating the air flow control device 20 by receiving a linear actuator control signal 29, 49, while translating the linear control signal in the scaling map 60 function. Simultaneously the actuator control signal is linearized by the adapted scaling map 60 function to an inverse linearizing control signal 28, 48 adapted for controlling the actuator 25, 45 controlled non-linear damper 32, 34, 36, 38, 52, damper unit 30, 50, or damper group 100.

[0176] As a result, the mentioned method will achieve better precision using linear actuator control signals ACS [V] 49, instead of previously non-linear signals, an improvement resulting in predictable linear proportional changes in RAF [%] for actuated dampers 30, 50, and improved controllability of the damper blades, as non-linearities between control signal and actual resulting relative air flow can be more easily predicted and followed up.

[0177] Note that the actuator control signal ACS 49 may be a voltage signal, for example from 0 to 10 Volt, or a digital signal from 0 to 100%, or another value range. The actuator control signal ACS 49 controls an actuator, which then controls a damper.

[0178] By following this method, the air flow control precision of these systems can be significantly improved by incorporating the acquired data on the non-linear characteristics of the actuated dampers and using it for enhanced control and regulation of the air flow.

[0179] The S5000 LINEAR OPERATION method is executed by first: operating the air flow control device 20 by receiving a linear actuator control signal 29, 49, and translating the linear control signal in the scaling map 60 function where the actuator control signal is linearized with expect to air flow by the adapted scaling map 60 function to an adapted linearizing control signal 28, 48 controlling the actuator 25, 45 controlled non-linear damper 32, 34, 36, 38, 52, damper unit 30, 50, or damper group 100, for improved and predictable air flow control at face dampers and bypass damper, using a request for relative air flow RAF [%] 29, 49, compared to using an actuator control signal ACS 24, 44 for the non-linear dampers.

[0180] The method may also correlate the Scaling Map with different temperature conditions The control method according as described in claim 1 , may further implement a step LEARNING S2000, ADAPTING S3000, and LINEAR OPERATION S5000 where the method is correlating multiple measurements of NAF 61 , AFWSM 62, ACLACS 67, or air pressure sensors data 66, with sensed outdoor air temperature (OAT) sensor 65 data. This lets the method make use of the correlated temperature (OAT) sensor 65 data when determining which linearization function 67 to use in the Scaling map 60, as described in claim 2.

[0181] As a result, the method may compensate for different non-linearities at different air temperatures, when adapting the Scaling Map 60 to provide the ability to control the dampers using linear proportional control signals, that results in proportional linear relative air flows.

[0182] The method uses at least three sub-steps in the Learning step to learn the non-linear relationship between the actuator signals for the actuated dampers and the resulting relative air flow

[0183] The method for improving air flow control precision may further implement the following sub steps of the LEARNING step S2000, where the following sub-steps are included:

[0184] The LEARNING (S2000) method is executed by:

[0185] (S2100) Registering and storing in the Scaling map 60 at least an air flow, pressure or differential pressure 66 with measured relative air flow when at least one actuated damper is in a closed position 52’ for the outlet air flow 72, 74, 76, 78, 79, for measuring of non-linearities of the regulated air flow, measuring of a zero flow level 79’, for measuring effects from the at least one closed actuated damper 32, 34, 36, 38, 52, on the air flow for the remaining actuated dampers 32, 34, 36, 38, 52,

[0186] (S2200) Controlling a at least one damper with increments over range from fully closed to fully open while registering, calculating and storing all RAF[%] / ACS[V] air flow control signals 27, 47, 29, 49 and air flows 72, 74, 76, 78, 79, pressure or air flow sensor 66 data, over the at least one actuated damper 72, 74, 76, 78, 79 at regulation ranges 61 , 62, 63 regulating the relative air flow RAF [%] 64, and (S2300) Registering at least an air flow, pressure or differential pressure 66 within the air flow regulation range when at least one actuated damper is at a fully open position 32, 34, 36, 38, 52 for the outlet air flow 72, 74, 76, 78, 79, for measuring of non-linearities of the regulated air flow, measuring relative air flow 72, 74, 76, 78, 79 over an at least one fully open actuated damper, for measuring effects on the relative air flow from the at least one fully open actuated damper 72, 74, 76, 78, 79, on relative the air flow for the remaining actuated dampers 72, 74, 76, 78, 79.

[0187] As a result, during the LEARNING step, the method records typical non-linear relationships between the actuator control signals and actual relative air flow achieved, as a base to later determine how a linearization function should be determined, to enable a one-to- one translation using the Scaling Map 60 as a signal translator, between the actuator control signals and resulting relative air flow.

[0188] The method may also verify the linearization effects after the Learning step, in a Verifying step

[0189] In the VERIFYING step, the method is verifying the linearization function of the scaling map 60 at its full control range. The VERIFICATION step is added, or inserted, between the ADAPTING step S3000 and the LINEAR OPERATION step S5000. The VERIFICATION step comprises:

[0190] S4100 Verifying an air flow, pressure or differential pressure 66 with a normal air flow when at least one actuated damper is in a closed position 52’ for the outlet air flow 72, 74, 76, 78, 79, for measuring of non-linearities of the regulated air flow, measuring of a zero flow level 79', for measuring any damper leakages or measuring effects from the at least one closed actuated damper 32, 34, 36, 38, 52, on the air flow for the remaining actuated dampers 32, 34, 36, 38, 52,

[0191] (S4200) Verifying the scaling map 60 and storing deviations to all RAF[%] / ACS[V] air flow control signals 27, 47, 29, 49 and air flows 72, 74, 76, 78, 79, pressure or air flow sensor 66 sensing air flow, over the actuated damper 72, 74, 76, 78, 79 within regulation ranges 61 , 62, 63 regulating the relative air flow RAF [%] 64, and

[0192] (S4300) Verifying the air flow, pressure or differential pressure 66 within the air flow regulation range when at least one actuated damper is at a fully open position 32, 34, 36, 38, 52 for the outlet air flow 72, 74, 76, 78, 79, for measuring of deviations from linearities of the regulated air flow, measuring air flow 72, 74, 76, 78, 79 over an at least one fully open actuated damper, for measuring any damper imperfections in abilities to offer a fully open air flow, or measuring effects from the at least one fully open actuated damper 72, 74, 76, 78, 79, on the air flow for the remaining actuated dampers 72, 74, 76, 78, 79, and recording and storing that the linearization function in the Scaling Map 60 is verified. That is comprising at least storing: an identity 101 and a time stamp if available of individual damper group, damper unit, face dampers, or bypass damper, data indicating specified deviations, or verified at a certain date, time, air pressure, control signal data ranges, or temperature.

[0193] The verifying step provides support for fault monitoring, identification of variations in mechanical damper valve play, learning, performance monitoring, and quality inspection.

[0194] As a result, the recording that the linearization function in the Scaling Map 60 is verified. The verification may report that the verification was made as “verified with specified deviations”. Verification data also may contain a certain date, time, air pressure, control signal data ranges, or temperature, for fault monitoring, learning, performance monitoring, and quality inspection.

[0195] As such the verification step allows for quality control and feedback of any deviating linearization control function over time, to ensure continued operation. Reasons for deviations may be wear and tear as well as particle build ups in air ducts and dampers. Also, such deviations and needs for maintenance may be detected and reported to a user or maintenance function, at an early stage.

[0196] Method to operate with “no scaling map’’, as a fallback if verification results are outside limits

[0197] In case the verification results are outside limits of a useful and efficient linearization of damper actuator control signals (ACS) to relative air flow (RAF), then adjustments can be tested, but in case the operation is not within the bounds of the limit, then the method may return to a fallback where the control method operates without a linearization function as provided with a Scaling Map 60.

[0198] The method step S1000 is implemented such that the control system 20 is operating in a non-linear control mode with non-linear actuated damper air flow characteristics. The step S1000 is provided as an initial step before step LEARNING S2000 and step ADAPTING S3000, or as a safety-critical backup mode in case the control function 20 reports a fault or out-of- control range problem, or a failure during the VERIFYING S4000 step if present, while indicating that further LINEAR OPERATION S5000 cannot be executed with precision, errors, or other deviations from earlier linearization function.

[0199] Method step S1000 also serves as an initial operational state when the damper actuator control linearization function in the Scaling Map is not yet adjusted and updated to operate in a linearization mode.

[0200] In one embodiment of the invention, the step S1000 may act as an initial state, or a failsafe fallback, or alternative control mode, where no Scaling Map is used for the regulation of the damper-units.

[0201] In step S1 100 represents a system sub-state = Open loop non-linear, the control system operates as a basic fallback system state where no compensation for non-linearity is provided. If the air flow, pressure, differential pressure sensor 66 is present in the system, the system, typically orchestrated by the Air flow controller 80 may switch to system sub-state step 1200, for Closed loop constant flow, control.

[0202] In step S1200 the bypass-damper is controlled to maintain a constant flow, and to avoid pressure drop especially when all face-dampers closes. Face-dampers are controlled to avoid pressure drop over any Plate Heat Exchanger (PHE), or similar rotational heat exchanger. Step S1200 represents a system state = Closed loop constant flow, wherein the differential pressure sensor 66 measures pressure difference across the Plate Heat Exchanger (PHE) (as found in many HVAC systems), on the damper side to maintain constant pressure by controlling the bypass damper.

[0203] The Air flow controller 80 controls the damper actuators to ensure that the total flow through the PHE stays constant, provided that the flow resistance across the bypass-damper is the same as across all sectional face-dampers. The sectional face-dampers are controlled based on defrosting needs, and the bypass-damper is controlled to maintain a constant differential pressure.

[0204] As a result, the main task of the bypass-damper is to prevent the actuator throttled section face-damper regulation from leading to an increased total pressure drop. Yet another result, this leads to an improved effect on defrosting, because the pressure does not increase when the sectional face-dampers close

[0205] Hence, method step S1200 as described reduces the energy requirement due to increased pressure drop.

[0206] The regulation effect of S1200 is similar as the scaling map 60 in step S5000, but with a feedback regulation used in step S1200 instead of a scaling map. As the feedback loop in S1200 is somewhat slower due to inertia in a feedback regulation system, it is expected that step S5000 Operating using Scaling Map at validated range, will provide for a faster regulation mechanism, even without sensor 66.

[0207] Hence two different system modes are offered to achieve a very similar regulation, one with open control S5000 and one with feedback control S1200.

[0208] Typically the system state, would switch from step S5000 LINEAR OPERATION, to a step S4000 VERIFYING where the system would determine a situation such as a fail-state, or a degraded performance using the Scaling Map, and as a result step S4000 would make an operation state transition to step S1000 NO SCALING MAP, where the system would maintain a constant air flow according to the method, described. In case the system provides an air flow, or differential pressure sensor 66, the system 20, preferably the Air flow controller 80, will switch to sub-step S1200, rather than the default sub-step S1100, within step S1000.

[0209] A Scaling map 60 can be calibrated by S2000 LEARNING and S3000 ADAPTING, and S4000 VERIFYING. Then the system may operate in its Operational State S5000, that is Operating using Scaling map at validated range. As a result, the Scaling map 60 eliminates the need for pressure sensor 66, as the Scaling map can operate in an open-loop control mode.

[0210] The system 20 may switch between Operational States as a manual result by an operator that first calibrates a system with Scaling Map at a calibration lab, and then removes the pressure or air flow sensor 66, and then sets the system to operate in state S5000 for shipping and installation.

[0211] The system may also be equipped, or be installed with pressure sensor 66, and initiate its operation in initial Operational State S1000 without activating the Scaling Map, waiting for an operator and / or system signal to enter the S2000 LEARNING , S3000 ADAPTING, optional S4000 VERIFYING, and then the state S5000 Operating using Scaling map at validated range.

[0212] The purpose of scaling map is to get a perfect mirror between section damper and bypass damper, so that the pressure drop across the Plate Heat Exchanger (PHE) is always constant. Therefore, the essence is to operate the sectional face-damper and the bypassdampers together, so that the changed position of the sectional face-damper is met by an equivalent change in the bypass-damper. This means that the effect of regulating the facedampers will be optimal, even without a flow or (differential) pressure sensor.

[0213] Method implementing a learning step while recording data for later retrieval and adapting of the linearization function in the Scaling Map

[0214] In the step LEARNING S2000, the method implements a measuring and registering of air flow, pressure, or differential pressure sensor 66, or a combination of sensors, as a mean for measuring pressure drop, air flow, and / or outlet pressure over at least one actuated damper. This implements an improved measurement of each actuated damper air flow control characteristics, while storing the air flow control characteristics 61 , 62, 67, in at least one of the scaling map 60, cloud data 90 via interface 91 , in a memory in the actuated damper 25, 45. The information stored may be provided in a human or machine readable format such as RFID, QRC, diagram presenting the characteristics, or text format, for later retrieval in other method steps.

[0215] Other methods steps may then retrieve the actuated damper air flow control characteristics which are stored associated with a unique actuated damper identity identifying the specific actuated damper, a type of indicator identifying the type of actuated damper product such as a product number, or the identity of a unit comprising actuated dampers, or as group comprising unit comprising actuated dampers or actuated dampers.

[0216] Example usages of stored characteristics are to use these characteristics to adjust a linearization function in the Scaling Map 60, to determine changes during the lifetime of the operation of the damper units and control system, to implement an adjustment function updating the Scaling Map 60, or to identify deviations during the Verification step.

[0217] Method implementing an adapting method where the Scaling Map is retrieved from memory.

[0218] During the ADAPTING step, the method determines a Scaling Map S3000 using information retrieved from an information storage. The storage has previously stored information about air flow control characteristics 61 , 62, 67, from an earlier scaling map 60. The information may be retrieved a cloud data 90 via interface 91 , or from memory in the actuated damper 25, 45, or from a human or machine-readable format such as Radio Frequency Identification (RFID), Quick Response Code (QRC), or a diagram presenting the characteristics, or from information in text format.

[0219] Method where the Adapting step is executed to generate and update a linearizing Scaling Map 60 as a cloud data 90 or central processing remote from the control system.

[0220] The ADAPTING step S3000 may be preferably executed in a central server, server cluster, artificial intelligence mechanism such as a neural network configuration, or any other adaptive computing function with memory and capacity to execute the adapting step to determine a preferably Scaling Map. To execute the adapting step S3000 in a central server has the advantage that edge computing near or at the control system will require less computing power. The adapting step may even be determined at a satisfactory level during production testing and calibration before leaving the factory producing actuated damper unit systems, and control systems.

[0221] Yet another advantage of using a centralized allocation such as a cloud server 90 for the execution of the Adapting step S3000 is that information gathering of a large volume of units may provide a superior statistical information base for optimization and calibration of the adapting step and Scaling Map 60 updated. Also, a centrally placed and controlled ADAPTING method may also ensure that the latest most developed and up to data ADAPTING method, or rather adaptation method is executed.

[0222] The method is configured to process the Adapting step S3000 to where it is adapting and determining the Scaling Map 60, executed as a cloud or central processing remotely from the control system 20, and wherein the scaling map 60 is updated from a cloud server or processing system 90 via a cloud computing interface 91.

[0223] Method where the adapting step S3000 is using an adaptive constantly improving function for optimization of the linearization function for the Scaling Map.

[0224] The ADAPTING step S3000 is analyzing, optimizing suing statistical methods, and updating the Scaling Map 60 linearization control curve by analyzing stored interdependent actuator-damper air flow characteristics using artificial intelligence Convolution Neural Network (CNN), a Deep Learning Neural Network (DLNN), a Kalman filter, PID regulator, for adapting the compensating control signals to transform a linearized control signal to a control signal resulting in a linear proportional control of the air flow over the actuated dampers. The method step S3000 may also update a reversely lookup table where earlier recorded relative air flow in relation to actuator control signals sent to the damper actuators, can be looked up reversely. This lets the Scaling Map act as a reversely lookup table where desired relative air flow can be translated to previously recorded actual actuator control signals for corresponding damper actuators.

[0225] Method for VERIFYING step S4000 verifying a maintained precision of the scaling map and to carry out corrective actions

[0226] In this scaling map verifying method, the VERIFYING step S4000 is verifying a maintained precision of the scaling map's 60, precision in linearization of the actuated damper air flow control function. Concurrently the method step VERIFYING is triggered by a system event, error message, or a scheduling function, to carry out the following sub-steps: verifying scaling map at air flow as closed when at least one damper is closed as a reference, verifying scaling map at all control signal steps for ACS from min to max value while verifying and recording deviations from earlier recorded relations between RAF[%] as a result of ACS [V] or ACS as a digital control signal, verifying the scaling map when all dampers and air flow are fully open, and evaluating verification based on recorded deviations from previous calculated linearization scaling map.

[0227] Method for CLOSED LOOP CONSTANT FLOW operation step S1200

[0228] This method is operating using a differential pressure measuring pressure difference over the bypass damper 52 while regulating the bypass actuator damper to maintain constant a constant pressure independent of variations at the face-dampers. The difference pressure can be used for calculation of the air flow, and relative air flow over the damper. It is even possible to adjust the RPM speed of fan 71 to vary the absolute flow rate, but it has been found that by maintaining a constant pressure over the bypass damper.

[0229] The step (S1200) CLOSED LOOP CONSTANT FLOW is executed by: operating using a feedback loop wherein an air flow, pressure or differential pressure sensor 66 measures differential pressure over the bypass damper 66”, and the closed-loop control of bypass damper 69 is controlling the bypass-actuator 45 to steer the bypass-damper 52, wherein the airflow control system 20 or air clow controller 80 is maintaining a constant pressure difference over the bypass damper 52 by adjusting the bypass-actuator 45 to steer the bypass damper 52 to compensate for variations in face-damper 32, 34, 36, 38 air flows 72, 74, 76, 78.

[0230] In one alternative, using a differential pressure measurement over both a cross flow plate heat exchanger 73 and the bypass damper, may also provide detection of freezing at the plate heat exchanger.

[0231] The method may operate in both a closed loop and open loop mode at the same time, wherein the system adjusts the scaling map based in actual measurements from the sensors involved in the feedback loop for estimating relative air flow (RAF).

[0232] Air flow controller 80 as a component for regulating an Air Handling Unit (AHU), Heating, Ventilation, and Air Conditioning (HVAC), or Air Conditioning (AC) system

[0233] The air flow controller 80 is comprising a computing hardware and software 80 that executes the function but may be implemented as electrical hardwire system components for open loop control 23,43, and closed loop control 68, 69.

[0234] Furthermore the air flow controller 80 comprises the following interfaces, which may be combined into one or a few number of physical interfaces: an interface to a face damper relative air flow RAF request 29 and / or an interface to a face damper control 24 configured to control at least one face damper actuator 25 configured to steer an at least one face damper 32, 34, 36, 38, an interface to a bypass damper relative air flow RAF request 49 and / or an interface to a bypass damper control 44 configured to control at least one bypass damper actuator 45 configured to steer an at least one bypass damper 52, an interface to a face damper steering 28, an interface to a bypass damper steering 48, an interface for sensing face damper flow 66’ configured to receive data from an air flow, pressure, or differential pressure sensor 66 if connected, and an interface for sensing bypass damper flow 66” configured to receive data from an air flow, pressure, or differential pressure sensor 66 if connected.

[0235] The air flow controller 80 may contain a scaling map 60, or interface with an external scaling map 60, both configured to look up actual control signals ACS for face damper steering 28 and / or bypass damper steering 48 values, in the scaling map that predicts a desired face damper relative air flow RAF 66' and / or a desired bypass damper relative air flow RAF 66”.

[0236] A Manual input or Air flow Control System Mode 21 is provided in the air flow controller 80, configured to establish a current operation state S1000-S5000, comprising the operational states: OPEN LOOP NON-LINEAR,

[0237] CLOSED LOOP WITH CONSTANT FLOW,

[0238] LEARNING,

[0239] DETERMINING SCALING MAP,

[0240] VERIFYING, AND

[0241] LINEAR OPERATION.

[0242] The air flow controller 80 is furthermore configured to when set in operation state=LINEAR OPERATION: to receive: the face damper relative air flow RAF request 29 and the bypass damper relative air flow RAF request 49, and use the scaling map to look up actual control signals ACS stored in the scaling map 60 that predicts the requested face damper relative air flow RAF 66’ and desired bypass damper relative air flow RAF 66”, and to use the looked up actual control signals ACS to control the face damper steering 28, and the bypass damper steering 48, to steer any connected HVAC, AHU, or AC system to generate: a requested faced damper flow 66’ is equal to the face damper relative air flow RAF request 29, and a requested bypass damper flow 66” is equal to the bypass damper relative air flow RAF request 49.

[0243] As such the air flow controller 80 is acting as an electronic control unit (ECU) or integrated as an active software function in another ECU, to enable a user, external actor or a system to control an HVAC system, AHU, AC-system having a damper unit 100, by requesting desired relative air flow at the outlet after each face damper and / or bypass damper.

[0244] To control the damper group 100, damper groups 30, 50 with mentioned face dampers and bypass dampers, using a predictable request for relative air flow (RAF), is a much more precise and predictable method to control mentioned systems and units. In the past, dampers where controlled using typically voltage levels from 0 to 10 [V] to control a non-linear actuator to damper function resulting in a non-linear steered relative air flow.

[0245] Air flow controller 80 as a component for learning and registering non-linear relationship between actuator control signal and resulting relative air flow

[0246] According to an aspect, the Air flow controller 80, is configured to when set in operation state= LEARNING: to stepwise iterate and test each actuator control signal ACS between closed to fully open in any order for combinations of face damper steering 28 signals and bypass damper steering 48 signals for controlling an at least face damper actuator 25 steering a face damper 32, 34, 36, 38 and controlling a bypass damper actuator 25 steering a bypass damper 52 and, in the Scaling Map 60: store each face damper steering 24 signal value together with sensed face damper relative air flow RAF 66’, and store each bypass damper steering 48 signal value together with sensed bypass damper relative air flow RAF 66”.

[0247] The Air flow controller 80 is further configured to execute air flow control methods as described herein.

[0248] An air flow system 20 for open-loop control with improved air flow control precision of dampers

[0249] The invention also comprises an air flow open-loop control system 20 for improved air flow control precision for damper units in air flow controllers for HVAC, VAV and heat exchange systems.

[0250] As such, the air flow open-loop control system 20 is comprising at least the following parts during the steps S2000 LEARNING, and any S4000 VERIFYING if present: a damper group 100 or at least one damper group comprising: at least one face-damper actuator 25 configured to steer at least one actuated face-damper 32, 34, 36, 38 configured to control an inlet 70 or outlet air flow to at least one face-damper outlet air flow 72, 74, 76, 78, a bypass-actuator 45 configured to steer at least one actuated bypass-damper 52 configured to control an inlet air flow 70 to at least one bypass-damper outlet air flow 79, an air flow controller 80 wherein the interfaces 24, 28, 29, 44, 48, 49, 66, 66’ ,66” are connected the air flow control system's 20 sensors and actuators.

[0251] Specifically, the air flow control system 20may comprise: an Open-loop control of face dampers 23 and / or an Open-loop control of bypass dampers 43.

[0252] Mentioned open loop controls are configured for open-loop control of face dampers and bypass dampers without using any air flow sensor, nor differential pressure sensor 66.

[0253] When the open the air flow control system 20 is registered, and optionally verified, it may operate without any: air flow, pressure, or differential pressure sensor 66.

[0254] Hence no, air flow, nor pressure, nor differential pressure sensor 66 is needed to execute the step S3000 Determining Scaling Map.

[0255] And none of the air flow, pressure, nor differential pressure sensor 66 are needed to execute the actual open-loop control operation in step S5000, LINEAR OPERATION, where the control system 20 is Operating using Scaling map, in a true open-loop mode. In case more sensor data is available, the system may enter step S4000 VERIFYING where it is Verifying its Scaling map at full control range, even for run-time calibration. Reference values for the air flow, pressure, nor differential pressure sensor in relation to relative air flow can be determined in a calibration lab, during production, or provided as input manually, or automatically after commissioning.

[0256] The air flow controller 80 is configured to execute a control process according to the methods for the air flow control system as earlier described, also an implementation of the method as per herein.

[0257] The air flow controller 80 is configured for coordination of the operation state changes from LEARNING, ADAPTING, to LINEAR OPERATION.

[0258] The at least one actuated face-damper 32, 34, 36, 38 is configured for control of an inlet air flow 70 directed towards at least one face-damper outlet air flow 72, 74, 76, 78.

[0259] The open-loop control for face-dampers 23 is configured to control air flow and air distribution using at least one face-damper-actuator 25 actuated face-damper 32, 34, 36, 38.

[0260] The air flow, pressure, or differential pressure sensor 66, or a combination of these sensors, are configured to sense at least one of, or a combination of: air flow over a damper, damper unit, damper group, air pressure in an inlet air flow 70 to a plenum box, duct, cooling beam, or an air chamber for a face-damper unit 30 or bypass-damper unit 50, to sense differential pressure indicating an air flow.

[0261] The Scaling Map 60 is configured to translate linear air flow control signals 29 for controlling at least one non-linear actuated face-damper 32, 34, 36, 38 with air flow 72, 74, 76, 78, to result in a linear proportional change in relative air flow RAF over the actuated face-damper. The Scaling Map performs this linearization function when the Scaling Map has been “trained”, configured, or adapted during the earlier described Adapting step.

[0262] The air flow open-loop control system 20 controls state transitions from LEARNING to DETERMING SCALING MAP that is Adapting, to LINEAR OPERATION, to let the control system first learn from a non-linear actuator control mode of dampers resulting in in non-linear adjustment of relative air flow RAF, measured as a percentage [%]. The resulting scaling map can then be implemented on other dampers in the same set.

[0263] The air flow open-loop control system 20 may also incorporate more state transitions such as a VERIFYING step S4000 to Verify the Scaling Map 60 at full operational range of control signals for damper actuators, as well as for full range of operational control range of resulting relative air flow RAF [%] over each actuated damper, face-dampers, and bypassdamper.

[0264] Each of the states, corresponding to method steps are explained as follows.

[0265] When the operation state is in a LEARNING mode the open-loop controller for face-dampers 23 is configured to regulate the actuator control signal ACS 63 over its complete regulation range 0-10 [V], or 0-100% wherein the scaling-map 60 is configured to record and store the non-linear air flow control characteristics NAF 61 and register and store the NAF 61 with control signal 27, sensed air flow 66, and / or pressure or differential pressure sensor 66 data, and convert sensor values to a relative air flow RAF and then store the characteristics as preferably a table 63, 64’, 64, 101 in the scaling map 60, cloud 90, or a memory in the face-damper actuator 25.

[0266] This lets the air flow open-loop control system 20 record the characteristics of a nonlinearized control system regulating and actuating multiple dampers, to enable further compensation in the Scaling Map 60.

[0267] When the operation state is in ADAPTING mode the scaling map 60 is configured to determine a linearization function 67 that linearizes the non-linear air flow control characteristics NAF 61 for the actuated face-damper 25, 32, 34, 36, 38 to the Compensated Linearized Actuator Control Signal CLACS 67.

[0268] When the face-dampers are identical, it is sufficient to measure and determine the scaling map for one-face-damper and then apply the same linearizing scaling map when actuating all remaining face-dampers. The same principles apply if a bypass damper unit 50 is equipped with multiple bypass dampers 52.

[0269] Method steps to determine and adjust a feasible linearization function in the Scaling Map, would be known from regulation techniques in textbooks in the field. The linearization mechanism may be established by finding a linearization function that would translate an actuator signal for an actuated damper to its directly proportional relative air flow as a percentage.

[0270] A direct method to achieve such a linearization function is to record every pair of original actuator control signal with corresponding resulting relative air flow percentage, and then to implement the linearization function as a reverse lookup function that for a given requested actuated damper relative air flow (RAF), would look up the requested RAF to find the related Actuator Control Signal ACS as voltage, or digital signal, and then make use of this signal as input to the actual damper actuator, one of the face-damper actuators, or bypass actuator.

[0271] When the operation state is in LINEAR OPERATION mode, the scaling map 60 is configured to receive a linear proportional control signal 29, 49 defining a relative air flow 29 request, and to translate the control signal 29, 49 into: a Compensated Linearized Actuator Control Signal CLACS, as a steering signal from the scaling map 28 to control at least one actuated 25 face-damper 32, 34, 36, 38 relative air flow

[0272] 72, 74, 76, 78 or bypass-damper 52 relative air flow 79, resulting in a linear proportional face-damper air flow 72, 74, 76, 78 in relation to the linear relative air flow face-damper control signal 29 for the actuated face-damper 25, 32, 34, 36, 38, and / or resulting in a linear proportional bypass-damper air flow 79 in relation to the linear relative air flow bypass-damper control signal 49 for the actuated bypass-damper 79.

[0273] This provides an improvement in air flow control precision as result from the improved predictability of a linear control signal resulting in a linear proportional relative air flow over each face-damper 32, 34, 36, 38 and bypass-damper 79 in an open-loop control system.

[0274] An air flow open-loop control system 20 for improved air flow control precision of bypass-damper(s) The air flow control system 20 is normally extended to control not just the actuated facedampers, but also the actuated bypass-damper, or dampers, offering a control using one linear control signal(s) for each actuated damper. Thus, it would be possible to request a certain relative air flow RAF percentage over a certain damper such as at least one but also a combination of face-dampers and bypass-dampers, and even multiple bypass-dampers. When face-dampers are geometrically identical, it is sufficient to determine and implement one scaling map to linearize and control the relative air flow of all face-dampers in a face-damper unit.

[0275] The air flow control system 20 extends the implementation mentioned above.

[0276] As a result, and as an example, if a new request for a new RAF over a bypass-damper, expressed as a linearized actuator control signal 48 (a steering signal from the Scaling map 60, that compensates for any non-linear control characteristics from the bypass-damper actuators for the actuator for the bypass-damper, demands an air flow change from 10% to 30%, then it may be necessary to compensate and adjust the signals actuator control signals 28 (a steering signal from the Scaling map 60, that compensates for any non-linear control characteristics from the face-damper actuators) to the actuator for the face-dampers.

[0277] Orchestrating a linear control of multiple actuated face-dampers and bypass-dampers in a complex air-treatment or HVAC system may require a more advanced multi-variable linearization function in the Scaling Map.

[0278] With face-dampers of identical proportions, this control scenario can be simplified. Each face-damper actuator controls the relative air flow according to an identified linearization function shared as a Scaling Map for all face-dampers with identical geometry and calibrated opening angles.

[0279] For example, if we have a bypass and a section damper, the control system and ECU will be able to co-control several face-dampers using the same scaling map, to achieve a constant common flow, which is a significant advantage of the Scaling map. For example, when operating without a Scaling Map, such as during first startup of a new system or during commissioning, the system will typically when starting with a closed bypass and an open section damper, where the effect of opening the bypass becomes much greater than closing the section damper, leading to an overall increased flow. This can be avoided by closing the facedampers blade more than the bypass-damper blade opening angle. This linearization function is exactly what the scaling map accomplishes.

[0280] The air flow open-loop control system 20 offering linearization of bypass actuator control signals comprises of at least: at least one actuated bypass-damper 32, 34, 36, 38 controlling an inlet air flow 70 towards at least one bypass-damper outlet air flow 79, an open-loop control for bypass-dampers 43 configured to control air flow and air distribution using at least one bypass-actuator 45 actuated bypass-damper 51 .

[0281] The air flow open-loop control system 20 offering linearization of bypass actuator is configured to learn, adapt (determining scaling map) and to perform in a Linear Operation mode, configured as follows.

[0282] When the air flow open-loop control system 20 is in its operation state LEARNING mode then the open-loop controller for bypass-dampers 43 regulates the actuator control signal ACS 63 over its complete regulation range 0-10 [V], or 0-100% wherein the scaling-map 60 is configured to record and store the non-linear air flow control characteristics NAF 61 and register and store the NAF 61 with control signal 47, sensed air flow 66, and pressure or differential pressure sensor 66 data, in the scaling map 60, cloud 90, or a memory in the bypass-damper actuator 45.

[0283] When the air flow open-loop control system 20 is in its operation state DETERMINING SCALING MAP then the scaling map 60 determines a linearization function 67 that linearizes the non-linear air flow control characteristics NAF 61 for the actuated bypass-damper 45, 52 to the Compensated Linearized Actuator Control Signal CLACS 67.

[0284] The same procedure can also be implemented for all other dampers in the same set.

[0285] When the air flow open-loop control system 20 is in its operation state LINEAR OPERATION mode, then the scaling map 60 receives a linear proportional control signal 49 requesting an air flow 49 and translates signal into the linearized Compensated Linearized Actuator Control Signal CLACS, as a steering signal from the scaling map 48 to control the actuated 45 bypass-damper 52 air flow 79, and the remaining dampers in the set or damper unit 30, 50 or damper group 100, resulting in a linear proportional bypass-damper air flow 79 in relation to the linear air flow bypass-damper control signal 49 for the actuated bypass-damper 45, 52.

[0286] An air flow open-loop control system 20 for improved air flow control precision of actuated dampers with temperature compensation.

[0287] The air flow open-loop control system 20 may be further equipped with at least on outdoor air temperature OAT sensor 65, to determine properties of the air temperature and transported thru the face-dampers 72, 74, 76, 78 and bypass-damper(s) 79 as well as any fan 71 contributing to the inlet air flow 70 and inlet air pressure to the damper units.

[0288] The air flow control system (20) comprises sensors for measuring airflow or differential pressure indicating an air flow, preferably: an outdoor air temperature OAT sensor 65, or an air flow, pressure, or differential pressure sensor 66 configured to sense and determine a defrost demand on a PHE 72,

[0289] The air flow open-loop control system 20 is further configured to make use of the OAT sensor, where the scaling map 60 is configured to receive a sensor signal from the outdoor air temperature OAT sensor 65, or the air flow, pressure, or differential pressure sensor 66 indicating a defrost demand at the cross flow plate heat exchanger PHE 72, and to determine a Compensated Linearized Actuator Control Signal ACLACS 67 correlated with the temperature 65, and the scaling map 60 translates the Compensated Linearized Actuator Control Signal the ACLACS 67 correlated with the outdoor air temperature OAT sensor 65 temperature, or the air flow, pressure, or differential pressure sensor 66.

[0290] This lets the Scaling Map 60 operate its linearization in a state or mode, that compensates for temperature changes in the air flow, otherwise altering the density of the air flow along with resulting further non-linearities between control signal and actual resulting relative air flow percentage.

[0291] The air flow open-loop control system 20 is further configured to incorporate functions for multi variable linearization functions.

[0292] As a result, the air flow open-loop control system 20 may incorporate a scaling map 60 linearization function making use of more complex multi variable analysis, thus enabling the Scaling Map to receive multiple input signals and generate multiple dependent output signals where complex internal dependencies may exist. This function is determined through a linearizing differential equation system, which compensates for non-linearities resulting from the interaction between various interdependent components such as actuated dampers 25, 45, damper units 30, 50, and damper groups 100. The purpose of this compensation is to address the non-linearities caused by the air flow and pressure system interaction between communicating dampers 32, 34, 36, 38, and 52, damper units 30, 50, and damper roups 100.

[0293] A second embodiment may also incorporate control and regulation of a single fan, or multiple fans, in concert orchestration with the face-damper unit and bypass-damper unit. As a result, a requested new linear air flow through face-damper number 2 may have to be compensated with an increase in fan speed capacity. Hence, the controller of a fan, or multiple fans may be integrated in the same control function.

[0294] The air flow control system 20 may be implemented using alternatives to pure hardware electronics with software.

[0295] This air flow control system 20 may perform its operation in an open-control loop where the scaling map 60 linearization function is determined using and implemented by at least: a P-, PI-, PID-regulator, a Kalman filter, a linear regression analysis, a convolutional neural network CNN, translation of lookup values, a non-linear regression technique, piecewise linear approximation, or a partial differential equation, for each actuated 25, 45 damper 32, 34, 36, 38, 52, damper unit 30, 50, or damper groups 100.

[0296] The furthermore, the air flow open-loop control system 20 may be designed with a scaling map 60 linearization function configured to represent and determined relationships between ACS and RAF for all actuated face dampers and bypass dampers as a linearizing differential equation system. A linearizing differential equation system where interdependent actuated 25, 45 face damper 32, 34, 36, 38, and bypass damper 52, damper unit 30, 50, or damper groups 100 are compensated for non-linearity, for compensation of non-linearities due to air flow and pressure system interaction between communicating dampers 32, 34, 36, 38, and 52, damper units 30, 50, and damper groups 100.

[0297] The air flow open-loop control system 20 is further incorporating an Air flow controller hardware (HW) and software (SW)

[0298] The air flow open-loop control system 20 usually incorporates an Air flow controller hardware (HW) and software (SW) 80. This controller comprises an operation state verification feature. Control system 20, 80 is configured to assess the effectiveness of the linearizing function in the scaling map 60 by measuring relative changes in air flow. This is achieved by using a linear proportional air flow control signal to request a proportional air flow while employing the scaling map 60 to translate and linearize the control signal into a damper actuator control signal.

[0299] As a result, proportional air flow changes are observed across face-dampers 72, 74, 76, 78 and the bypass-damper 79. Simultaneously, any deviations from air flow control linearity are identified, recorded, and communicated to the control system, supervision system, or an operator for information or alerts.

[0300] The air flow control system (20) for open-loop control for Verification of scaling map 60

[0301] The air flow control system 20 for open-loop control can be provided with a function for Verification of current or part scaling map 60 correctness. Then the Air flow controller 80 compromises an operation state=Verification as a system mode.

[0302] When the system is in the operation state= Verification: the control system 20, 80 is designed to verify the effect of linearizing function in the scaling map 60 by measuring relative air flow changes while using a linear proportional air flow control signal to request a linear proportional air flow while using the scaling map 60 to translate and linearize the linear control signal to a damper actuator control signal, resulting in proportional air flow changes over face-dampers 72, 74, 76, 78 and bypass-damper 79, while identifying deviations from air flow control linearity, recording deviations, and, informing or alerting the control system, supervision system or an operator.

[0303] A closed loop air flow control system (20)

[0304] A closed loop air flow control system (20) is designed to comprise: a Closed-loop control of face-dampers 68, and / or a Closed-loop control of bypass-damper 69 and an air flow, pressure, or differential pressure sensor 66 for: measuring relative air flow over at least one face damper 66’, and / or measuring relative air flow over a bypass damper 66”, wherein the airflow control system 20 is configured for closed-loop control of face dampers and bypass dampers, wherein the Air flow controller 80 or Closed loop control of bypass-damper 69 controls the bypass damper actuator 45 to maintain a constant differential pressure 66” between the air intake 70 air pressure and the bypass damper air stream 79. The Air flow controller 80 or Closed loop control of bypass-damper 69 may also control the bypass damper actuator 45 to prevent a pressure drop over the differential pressure sensor 66 measuring differential pressure 66” between the air intake 70 air pressure and the bypass damper air stream 79. If a sudden pressure drop occurs this is likely due to icing at face dampers, clogging of the dampers or other error situation. An action to prevent a progressing differential pressure drop 66” is to proactively open the bypass damper, thus releasing air thru the bypass damper 52. Depending on the design of an AHU, HVAC system or AC system using the invention, the bypass damper 52 may further release air into a dump area, to the outside of a building, back in the outdoor channel, or back to a heat exchanger for energy recovery.

[0305] An Air handling unit AHU comprising the air flow control system 20 or controller 80

[0306] An Air handling unit AHU, HVAC system or AC system, can be configured with a flow control system (20) or an air flow controller 80 where the Air handling unit AHU is comprising: an air flow control system 20, or an air flow control system 20 with air flow control system interfaces are connected with damper actuators, and connected with at least one air flow, pressure, and differential pressure sensor 60, 66’, 66”, configured to operate according to method as per herein.

[0307] Scaling map

[0308] The scaling map 60 for the Air flow control system 100, and or Air flow controller 20 may in its simplest form, but not limiting, comprise at least: a conversion table with table records comprising: identity 101 identifying an actual damper group 100, damper group 30, 50, face damper 32, 34, 36, 38, and bypass damper 52, a face damper or bypass damper control air flow control ACS signal [V] 63, a face damper or bypass damper's relative air flow [%] 64’, and a measure face damper's or bypass damper's relative air flow [%] 64.

[0309] The scaling map 60 may be further designed according to more sophisticated designs, as previously mentioned, such as comprising CNN, differential equation system representations, tables, and matrixes, as fell as database designs and memory structures, supporting multi variable look up and retrieval of identities 101 , face dampers, bypass dampers, damper units, and damper groups, as well as actuators with mechanical designs.

[0310] Reference signs list

[0311] Description of reference numbers used in drawings are defined as follows. First a figure reference number is specified, and then a short name of the item, arrow, flow of information or method step is presented, and in some cases also briefly described.

[0312] The complete system described here, is the “work tool carrier sensors control system” with reference number 20.

[0313] To avoid risks of mixing figure numbers with reference sign numbers, numbers 1 -19 are omitted. Method steps are numbered as S1000, S2000, and on.

[0314] The control system:

[0315] 20= An air flow open-loop control system for improved response precision for an actuated damper air flow control in HVAC, VAV (baffle or plenum chambers) and heat exchange systems.

[0316] Supervision system and input signals:

[0317] 21= A manual input or, input signal from an air flow control system, that requests a system state, a certain air flow over an actuated damper 31 , an actuated damper unit 30, 50, or an actuated damper group 100.

[0318] 22= Request as a manual input or from a control system 21 , requesting a relative air flow from an actuated face-damper 32, 34, 36, 38, bypass-damper 52, damper unit 30, 50, or damper group 100. The request is either a steering signal 22 to a damper actuator such as a face-damper actuator 25, a bypass-damper 52 actuator 45, or when using the scaling map 60 in open-loop control 23, 43 a request 29, 49 for a linear proportional relative air flow control signal 62.

[0319] Face-damper control:

[0320] 23= Open-loop control for face-dampers.

[0321] 24= The damper control signal, usually non-linear in relation to the desired facedamper air flow 72-78, sent to the face-damper actuators 25. Typically, the damper control signal is a voltage from 0 closed, to 10 [Volt] fully open, or as a digital signal transferring a signal from 0 closed to 100% fully open for the actuator movement range.

[0322] 25= A face-damper actuator, usually with a non-linear control of connected facedampers' 32-38 air flow 72-78. Face-damper control signal [V] 0.0 to 10.0 [V].

[0323] 26= The actuation force and movement from the face-damper actuator moving the face-damper sections 32, 34, 36, 38, usually with a non-linear relation between actuated movement, rotation, or force, in relation to the change in related face-damper air flow.

[0324] 27= Open loop control signal sent to the scaling map 60. Before the Scaling map is adapted to linearize actuator signals, this signal will be a non-proportional direct signal to set the face-damper actuators 25 in certain positions, based on voltage such a 0 closed to 10 [Volt] fully open, or as a digital signal transferring a signal from 0 closed to 100% open of the actuator movement range.

[0325] When the Scaling map is adapted to function as a Scaling map 60 functioning as a linearization map, then the control signal 27, may be turned into a linear proportional to a requested 29 and delivered air flow at each individual face-damper 32, 34, 36, 38, or acting on a whole facedamper unit 31 , or a group of dampers including both a face-damper unit 31 , and an bypass- damper 50.

[0326] 28= A face damper steering signal from the Scaling map 60, that compensates for any non-linear control characteristics from the face-damper actuators 25 to the face-dampers 30 regulating the air flow 72, 74, 76, to 78 as linear proportional to a requested air flow 27.

[0327] 29= A linear and direct proportional face-damper control signal corresponding to a requested and linear proportional delivered air flow. This control signal then replaces the earlier non-linear control signal 27.

[0328] Face-damper mechanics:

[0329] 30= Face-damper unit comprising face-dampers and a duct or chamber connected to a source of air flow 70, such as a constant pressure fan or compressor.

[0330] 31= Face-dampers.

[0331] 32= First face-damper, for section 1 , regulating the air flow from the source 70 to its individual section 72.

[0332] 34= Second face-damper, for section 2, regulating the air flow from the source 70 to its individual section 74.

[0333] 36= Third face-damper, for section 3, regulating the air flow from the source 70 to its individual section 76.

[0334] 38= Fourth face-damper, for section 4, regulating the air flow from the source 70 to its individual section 78.

[0335] Bypass-damper control:

[0336] 43= Open-loop control for bypass-damper.

[0337] 44= The bypass-damper control signal, usually non-linear in relation to the desired bypass-damper air flow 79, sent to the bypass-damper actuator 45. Typically the damper control signal is a voltage from 0 closed, to 10 [Volt] fully open, or as a digital signal transferring a signal from 0 closed to 100% fully open for the actuator movement range.

[0338] 45= A bypass-damper actuator, usually with a non-linear control of connected bypass-damper 52 air flow 79. Bypass-damper control signal 0.0 to 10.0 [V], 46= The actuation force and movement from the bypass-damper actuator moving the bypass-damper 52, usually with a non-linear relation between actuated movement, rotation, or force, in relation to the change in related bypass-damper air flow.

[0339] 47= Open loop control signal sent to the scaling map 60. Before the Scaling map is adapted to linearize actuator signals, this signal will be a non-proportional direct signal to set the bypass-damper actuator 45 in certain positions, based on voltage such a 0 closed to 10 [Volt] fully open, or as a digital signal transferring a signal from 0 closed to 100% open of the actuator movement range.

[0340] When the Scaling map is adapted to function as a Scaling map 60 functioning as a linearization map, then the control signal 47, may be made linear proportional 49 to a requested and delivered air flow at the bypass-damper 52, or acting on a whole face-damper unit 51 , or a damper group including both a face-damper unit 31 , and a bypass-damper 51 , as an actuated damper group 100.

[0341] 48= A bypass damper steering signal from the Scaling map 60, that compensates for any non-linear control characteristics from the bypass-damper actuator 45 to the bypass- damper unit 50 regulating the air flow 79 as linear proportional to a requested air flow 47.

[0342] 49= A linear and direct proportional bypass-damper control signal corresponding to a requested and linear proportional delivered air flow. This control signal then replaces the earlier non-linear control signal 47.

[0343] Bypass-damper mechanics:

[0344] 50= Bypass-damper unit comprising bypass-damper and a duct 51 or chamber connected to a source of air flow 70, such as a constant pressure fan or compressor, or pressure-controlled volume, using the Air flow, pressure, or differential pressure sensor 66 to determine the actual air pressure before entering the bypass-damper, as well as Face-dampers. 51 = Bypass-damper duct with bypass-damper 52

[0345] 52= Bypass-damper, regulating the air flow from source 70 to the bypass channel without passing the heat-exchanger directly to a heat-battery where the air is heated before entering the building. The function is to provide sufficient air flow to the building when the heatexchanger cannot heat the total air flow without freezing.

[0346] A Scaling Map calibrated by Learning for a damper unit, eliminates the need for sensors and the need to measure air flow, pressure, or differential pressure over the damper units, or between the damper units, outdoor, or room air pressure.

[0347] In one embodiment, the bypass-damper is a mechanism to reduce air-pressure in the plenum and air flow 70, and to keep the pressure constant or at least regulated. As input to the pressure control, the sensor 66 measures preferably a differential pressure over the damper units.

[0348] Alternatively, the sensor may measure air flow, or differential pressure over the damper unit and a cross flow plate heat exchanger.

[0349] 52 - Bypass-damper in a closed position, having a zero-flow (79’).

[0350] 60= Scaling map, working as linearization function, and look up table between

[0351] Actuator Control Signal (ACS) usually but not limited to be expressed as a voltage from 0 to 10 Volt, and Relative Air Flow (RAF) usually expressed as a percentage of the total airflow.

[0352] 61= Non-linear Air flow (NAF), as a result when controlling a non-linear actuated damper.

[0353] 62= Air flow with Scaling Map (AFWSM), as a result when controlling a non-linear actuated duct, using a linear control signal where the linear control signal has been transformed using the Scaling Map to control a non-linear actuated damper, resulting in a linear proportional controlled air flow.

[0354] 63= Actuator Control Signal (ACS), in the range of 0 to 10 [Volt], or a digital or analogue signal indicating an actuator signal from 0 to 100%. Controls an actuator for the face damper or bypass damper.

[0355] 64= Measured Relative Air Flow (RAF) in [%] over the Face / Bypass damper.

[0356] 64’= Lookup index for earlier calibrated and stored Relative Air Flow (RAF) in [%].

[0357] Linearized Face / Bypass damper Relative Air flow request as [%] for reverse look up.

[0358] 65= Outdoor air temperature (OAT) sensor.

[0359] 66= Air flow, pressure, or differential pressure sensor. Preferably the sensor 66 is a differential pressure sensor, indicating a pressure drop. As input to the pressure control of a plenum chamber, airduct for the air flow 70, sensor 66 measures at least one unit out of an air flow, a pressure or differential pressure over the damper units, or between the damper units, outdoor, or room air pressure. Alternatively, the differential pressure sensor may measure the pressure over the fan 71 building up air pressure. Also, a current or RPM sensor for the fan 71 with motor, may be used as an indirect differential pressure sensor where the fan's electrical motor current variations, or shifts in RPM indicate variations in air pressure created by the fan.

[0360] 66 - Differential pressure sensor over Face damper(s), in relation to Intake air flow pressure at the damper unit 70.

[0361] 66”= Differential pressure sensor over Bypass damper, in relation to Intake air flow pressure at the damper unit 70.

[0362] 67= Compensated Linearized Actuator Control Signal (CLACS), in the range of 0 to

[0363] 10 [Volt], or a digital or analogue signal indicating an actuator signal from 0 to 100%.

[0364] This signal is essentially the inverse function of the Measured RAF 64, or a calculated actuator signal that compensates for non-linearity in the non-linear actuated damper to air flow control characteristics.

[0365] 68= Closed-loop control of at least one face-damper 32, 34, 36, 38 using a sensor 66 sensing air flow, pressure, or differential pressure to maintain a constant air flow, or pressure drop within a desired range, and to avoid pressure drops due to ice and frost building up at a Plate Heat Exchanger (PHE).

[0366] 69= Closed-loop control of bypass-damper 52, using a sensor 66 sensing air flow, pressure, or differential pressure to maintain a constant air flow, and to avoid pressure drop beyond a desired range, as the face-dampers 32, 34, 36, 38 are closing.

[0367] 70= Intake air flow into the damper unit, usually from a fan 71 , from a VAV, or HVAC system. The air intake is shared between the face-dampers (30) and bypass-damper (50).

[0368] 71= Fan, driving the air flow and building up air pressure. The fan is typically powered by an electric motor. The fan 71 may be used as an indirect differential pressure sensor where the fan's electrical motor current variations indicate variations in air pressure created by the fan. This means that the fan current may represent a differential pressure drop sensor 66.

[0369] 72= Outlet air flow from a first face-damper, section 1 .

[0370] 73= Heat Exchanger, or Plate Heat Exchanger (PHE), alternatively a rotational heat exchanger. The heat exchanger, or any similar function such as a heat pump with coils, may be placed in the system to recover or transfer heat, as well for cooling purposes. Placement of the heat exchanger, or equivalent part in the system, may vary according to system set-up.

[0371] Typically, heat exchanger 73 provides heat or cooling effect to air flow 70 propelled by a fan 71 , while withdrawing air from outdoor. The bypass-damper 52 may open a bypass-channel to regulate an air flow 79’ into the extract air, channel, typically to prevent freezing, and to regulate the pressure drop based on sensor 66.

[0372] 74= Outlet air flow from a second face-damper, section 2.

[0373] 75= Outdoor air.

[0374] 76= Outlet air flow from a third face-damper, section 3, if any.

[0375] 78= Outlet air flow from a fourth face-damper, section 4, if any.

[0376] 79= Outlet air flow from the bypass-damper. Usually the bypass-damper regulates air flow to an air intake of a heat exchanger, AC-system, a VAV, a heat pump, a plenum chamber, a baffle, or it may lead air to a dump area where excess air can be transferred without negative consequences such as a separate room, or a drill hole for a heat pump system.

[0377] 79 - A zero flow level (79’) from a closed bypass-damper (52’). A similar zero-flow

[0378] (zero-air flow) can be measured on any type of closed damper (32, 34, 36, 38) resulting in zeroflows 72’, 74’, 76’, and 78’. Zero-flow can be described as air flow and pressure leakage through a closed damper valve.

[0379] 80= Air flow controller, comprising hardware (HW) and potentially software (SW), and data for operation, coordination and execution of method steps and control algorithms.

[0380] The Air flow controller can be implemented as analogue or digital electronics, Field programmable gate array, computer circuit, ASIC, hybrid- or integrated circuits. It may even comprise a neural network function such as Convolutional Neural-Network (CNN) or similar adaptive constructions.

[0381] 90= Cloud data, and cloud database for information transfer of Scaling maps and damper-actuator product data.

[0382] 91= Wireless or wired communication interface transferring information between cloud data 90 and the Scaling map, linearization function 60.

[0383] 100= A group of actuated dampers, or actuated damper units. Several actuated damper units, and actuated damper may be sharing the same intake air flow through a plenum chamber, duct, or any other connections in a VAV, AC system or HVAC system.

[0384] 101= An identity (ID) of a group of dampers, a damper unit, a face damper, or a bypass damper. The ID is connecting pre-recorded characteristics for damper characteristics in the Scaling Map, including measurements and calculations made to store RAF% to ACS relationship to enable pre-calibration and recording of Scaling Map data to compensate for nonlinearities between ACS and RAF%.

[0385] 201= Unused region (too responsive). Very limited regulation of the actuator control signal results in a too big change in air flow response, which makes it hard to control the air flow with precision.

[0386] 202= Active region. Feasible and good response between actuator control signal and adjusted air flow, making this regulation range very feasible for precision control of air flow.

[0387] 203= Slow region. Slow response between actuator control signal and adjusted air flow.

[0388] 204= Dead region. The actuator control signals have very little impact on adjustments of the air flow.

[0389] Method steps:

[0390] S1000= OPERATING WITHOUT ANY SCALING MAP. Operational State = No scaling map. The system operates with actuated damper having non-linear relationship between control signal and air flow. Two subordinate states exist.

[0391] S1100= OPERATING WITHOUT ANY SCALING MAP. In step S1100, operation state Operating without any scaling map. The system operates in an open-loop control mode, where non-linearity of control of face-dampers 32, 34, 36, 38 and the control of bypass-damper 52, are controlled by an open-loop control curve for face-dampers that that does not compensate for non-linearities between actuator signals 24, 44, and control results for relative air flow at each damper, face-dampers, and bypass-damper. The state S1100, represents a typical operation state, with no added technical effect, besides acting as a basic rough operational mode. A mode that may be used as an initial state, during start-up, or a state that may act as a fallback in case of an error occurs, or that the Scaling Map operation in step S5000 operates outside its range, or if the validation step S4000 identifies a risk factor for continued operation at step S5000. In case air flow, pressure, or differential pressure sensor 66 is available, the Air flow controller 80 controlling method steps, that are state changes in the system, can switch to step S1200 Closed loop constant flow, as follows.

[0392] S1200= CLOSED LOOP CONSTANT FLOW. Controlling the air flow towards a constant air flow, or a pressure drop. Each face-damper is controlled by the Air flow controller 80, using a Closed-loop control of face-dampers 68 to avoid pressure drop due to ice building up at a Plate Heat exchanger (PHE) if present in the HVAC system 20. The Air flow controller 80 uses as Closed-loop control of bypass-damper 69 to maintain constant differential pressure, especially to avoid partial pressure beyond a predefined desired limit. As such, this means that the closed loop control of the bypass-damper prevents a situation from occurring, where totally closed face-dampers would lead to a pressure fall.

[0393] Step S1200 may act as a fail-safe fallback state, when air flow, pressure, or differential pressure sensor 66 is available, and the validation step S4000 identifies a risk factor for continued operation at step S5000.

[0394] S2000= LEARNING. Registering Normal non-linear air flow control without using the scaling map, while saving each measured sensor data in the scaling map. This records non-linearities between damper actuator control signal and resulting air flow rate as a percentage of the maximal air flow, possible.

[0395] S2100= Register Normal air flow as closed.

[0396] S2200= Register all RAF[%] / ACSO as normal air flow.

[0397] S2300= Register Normal air flow as fully open.

[0398] S3000= DETERMINING THE SCALING MAP.

[0399] S3100= Calculate linearization map based on Normal air flow RAF[%] / ACS[V] plot.

[0400] S3200= Generate scaling map that linearizes actuator control signal.

[0401] The linearization function or scaling map corresponds to a Compensated Linearized Actuator Control Signal (CLACS) curve, with in a valid control range, and given other conditions such as outdoor temperature sensor data. Each actuated damper, damper unit, and damper group may have its own scaling maps, and CLACS curve.

[0402] S4000= VERIFYING. Verifying Scaling map at full range and storing results.

[0403] S4100= Verify Scaling map at air flow as closed.

[0404] S4200= Verify Scaling map at all RAF[%] / ACS[V], S4300= Verify Scaling map at air flow as fully open. S5000= LINEAR OPERATION.

[0405] Operating using Scaling map at validated range

[0406] Where the actuator control signal is linearized by the adapted Scaling Map

Claims

CLAIMS1 . A method for controlling airflow through a non-linear actuator-controlled damper in an HVAC, VAV, or heat exchange system, comprising the steps of:(a) learning the non-linear relationship between actuator control signals and actual airflow (relative air flow percentage) by registering measured values across the operational range of the damper;(b) generating a Scaling Map that translates a linear airflow request signal into a non-linear actuator control signal;(c) operating the damper in an open-loop control mode by applying the translated actuator control signal to achieve a linear proportional airflow response to the input signal; wherein the method enables linear airflow control without requiring direct airflow feedback sensors.

2. The method of claim 1 , further comprising the step of: updating the Scaling Map based on sensor measurements or performance deviation over time, by re-learning the non-linear response characteristics of the damper and re-calculating the map accordingly.

3. The method of claim 1 or 2, wherein the Scaling Map is applied to a unit or group of multiple dampers (face and / or bypass dampers), such that a single or coordinated linear input signal results in a proportional combined airflow response.

4. The method of any preceding claim, further comprising the step of: retrieving Scaling Map data from a remote memory or cloud server, the data having been generated based on damper type, production series, or individual calibration, and applying said map to the controller at installation.

5. The method of any preceding claim, wherein the Scaling Map selection is dynamically adjusted based on environmental parameters such as outdoor air temperature or relative humidity, to improve linearization accuracy under varying conditions.

6. An airflow control device for HVAC, VAV, or heat exchange systems, comprising: at least one actuator-controlled damper; a control unit comprising computing hardware and software, the control unit configured to receive a linear airflow request signal; a memory storing a Scaling Map that translates said linear signal into a non-linear actuator control signal; wherein the control unit is further configured to apply the translated signal to the actuator to produce a linear airflow response over the operational range.

7. The device of claim 6, wherein the control unit includes a learning module configured to:measure the actual airflow response to actuator control signals, generate a Scaling Map based on said measurements, and store the Scaling Map for subsequent operation.

8. The device of claim 6 or 7, further comprising: a network interface configured to retrieve Scaling Map data from a cloud server based on damper identity, system configuration, or historical performance data.

9. The device of any of claims 6 to 8, wherein the control unit is configured to operate multiple dampers, including at least one face damper and one bypass damper, using coordinated Scaling Maps to achieve proportional airflow distribution.

10. The device of any of claims 6 to 9, further comprising: a verification module configured to test the accuracy of the Scaling Map during operation and, in response to detected deviation beyond threshold limits, revert the device to a non-linear open-loop fallback mode or activate closed-loop feedback regulation if sensors are present.11 . A method for improving air flow control precision in open-control loop systems, such asHVAC, VAV, or heat exchanger control systems, the open-control loop systems comprising at least one non-linear actuated damper with non-linear air flow control (72, 74, 76, 78, 79) characteristics, the method comprising utilizing a scaling map (60) for linear control of the at least one non-linear damper(32, 34, 36, 38, 52) providing air flow (70, 72, 74, 76, 78, 79) to transform linear control signals into control signals configured to enable a transformation of characteristics of the at least one non-linear actuated damper into a proportional relative air flow by in a measurement phase analyzing the non-linear actuated damper based on measurement values and mapping control signals and resulting non-linear air-flow rates in the scaling map,12. A method according to claim 11 , wherein the method comprising:(S2000) LEARNING by controlling the at least one damper with increments over range from fully closed to fully open while, registering data from open-control loop regulation (23, 43) of the at least pone non-linear actuated damper (32, 34, 36, 38, 52) in relation to an inlet (70) or outlet air flow (72, 74, 76, 78, 79), and saving the non-linear characteristics RAF[%] (61) and ACS [V] (27, 47) for the airflow of the at least one non-linear actuated damper (32, 34, 36, 3852) ,(S3000) ADAPTING bydetermining the scaling map (60) by,(S3100) calculating a linearization map based on the saved earlier recorded, or retrieved, air flow RAF[%] / ACS[V], and(S3200) generating the scaling map (60) so that it linearizes the actuated damper air flow control signal (28, 48),(S5000) LINEAR OPERATION by operating the air flow control device (20) by receiving a linear actuator control signal (29, 49), and translating the linear control signal in the scaling map (60) function where the actuator control signal is linearized with expect to air flow by the scaling map (60) to an adapted linearizing control signal (28, 48) controlling the actuator (25, 45) controlled non-linear damper (32, 34, 36, 38, 52), damper unit (30, 50), or damper group (100), thereby providing an improved and predictable air flow control at face dampers and bypass damper, using a request for relative air flow (RAF) [%] (29, 49).

13. The method according to claim 11 or 12, characterized in that generating a scaling map (60) that linearizes the actuated damper air flow control signal (28, 48), comprises mapping a linearized actuator signal look up table with linear actuator signal values directly proportional to actuated damper relative air flow values14. The method according to any of claims 11-13, characterized in that LEARNING (S2000), ADAPTING (S3000), and LINEAR OPERATION (S5000) comprises correlating multiple measurements of NAF (61), AFWSM (62), ACLACS (67), or air pressure sensors data (66), with sensed outdoor air temperature (OAT) sensor (65) data, and comprises utilizing correlated temperature (OAT) sensor (65) data when determining which linearization function (67) to use in the Scaling map (60).

15. The method according to any of claims 11 - 14, wherein LEARNING (S2000) further comprises, for at least one actuated face damper and by pass damper:(S2100) registering and storing in the Scaling map (60) at least an air flow, pressure or differential pressure (66) with measured relative air flow when at least one actuated damper is in a closed position 52’ for the outlet air flow (72, 74, 76, 78, 79), for measuring of non-linearities of the regulated air flow, measuring of a zero flow level (79’), for measuring effects from the at least one closed actuated damper (32, 34, 36, 38, 52), on the air flow for the remaining actuated dampers (32, 34, 36, 38, 52),(S2200) controlling at least one damper with increments over range from fully closed to fully open, while registering, calculating and storing all RAF[%] / ACS[V] air flow controlsignals (27, 47, 29, 49) and air flows (72, 74, 76, 78, 79), pressure or air flow sensor (66) data, over the at least one actuated damper (72, 74, 76, 78, 79) at regulation ranges (61 , 62, 63) regulating the relative air flow RAF [%] (64), and (S2300) registering at least an air flow, pressure or differential pressure (66) within the air flow regulation range when at least one actuated damper is at a fully open position (32, 34, 36, 38, 52) for the outlet air flow (72, 74, 76, 78, 79), for measuring of non-linearities of the regulated air flow, measuring relative air flow (72, 74, 76, 78, 79) over an at least one fully open actuated damper, for measuring effects on the relative air flow from the at least one fully open actuated damper (72, 74, 76, 78, 79), on relative the air flow for the remaining actuated dampers (72, 74, 76, 78, 79).

16. The method according to any of claims 11 - 15, further comprises VERIFYING (S4000) the linearization function of the scaling map (60) at full control range between ADAPTING (S3000) and LINEAR OPERATION (S5000) the VERIFYING comprising:(S4100) verifying an air flow, pressure or differential pressure (66) with a normal air flow when at least one actuated damper is in a closed position 52’ for the outlet air flow (72, 74, 76, 78, 79), for measuring of non-linearities of the regulated air flow, measuring of a zero flow level (79’), for measuring any damper leakages or measuring effects from the at least one closed actuated damper (32, 34, 36, 38, 52), on the air flow for the remaining actuated dampers (32, 34, 36, 38, 52),(S4200) verifying the scaling map (60) and storing deviations to all RAF[%] / ACS[V] air flow control signals (27, 47, 29, 49) and air flows (72, 74, 76, 78, 79), pressure or air flow sensor (66) sensing air flow, over the actuated damper (72, 74, 76, 78, 79) within regulation ranges (61 , 62, 63) regulating the relative air flow RAF [%] (64), and(S4300) verifying the air flow, pressure or differential pressure (66) within the air flow regulation range when at least one actuated damper is at a fully open position (32, 34, 36, 38, 52) for the outlet air flow (72, 74, 76, 78, 79), for measuring of deviations from linearities of the regulated air flow, measuring air flow (72, 74, 76, 78, 79) over an at least one fully open actuated damper, for measuring any damper imperfections in abilities to offer a fully open air flow, or measuring effects from the at least one fully open actuated damper (72, 74, 76, 78, 79), on the air flow for the remaining actuated dampers (72, 74, 76, 78, 79), and recording and storing that the linearization function in the Scaling Map (60) is verified, the recording and storing comprising: an identity (101) and a time stamp if available of individual damper group, damper unit, face dampers, or bypass damper,data indicating specified deviations, or verified at a certain date, time, air pressure, control signal data ranges, or temperature, for fault monitoring, identification of variations in mechanical damper valve play, learning, performance monitoring, and quality inspection.

17. The method according to any of claims 11 - 16, further comprising“NO SCALING MAP” operation, wherein an operation state= OPEN LOOP NON-LINEAR is set as an initial default operation mode entered before LEARNING (S2000) and ADAPTING (S3000), , the method further characterized in that the method further comprises operating the air flow controller (80), or airflow control system (20), in a non-linear control mode with actuated dampers with non-linear air flow characteristics, or a robust mode with no optimization of the non-linear damper control signals (24, 44).

18. The method according to any of claims 11 - 17, where in LEARNING (S2000) further comprises measuring and registering at least an air flow, pressure, or differential pressure sensor (66) measuring pressure drop, air flow, or outlet pressure over at least one actuated damper, for improved measurement of each actuated damper air flow control characteristics, and storing the air flow control characteristics (61 , 62, 67), and identity (101), in at least one of the scaling map (60), cloud data (90) via interface (91), in a memory in the actuated damper (25, 45), or in a human or machine readable format such as RFID, QRC, a table (63, 64, 64’, 101), or diagram presenting the characteristics, or text format, for later retrieval, wherein the actuated damper air flow control characteristics are stored associated with a unique actuated damper identity identifying the specific actuated damper, if available, or an indicator identifying the type of actuated damper product such as a product number, calibrated production batch, or the identity of a unit comprising actuated dampers, or as group comprising unit comprising actuated dampers or actuated dampers.

19. The method according to claim any of claims 11 - 18, characterized in ADAPTING, Determining Scaling Map (S3000) comprises retrieving data the air flow control characteristics (61 , 62, 67), for the scaling map (60), identity (101), cloud data (90) via interface (91), or from memory in the actuated damper (25, 45), or from a human or machine readable format such as RFID, QRC, diagram presenting the characteristics, or text format, optionally analyzing and improving recorded information using statistical methods, and storing the improved data in the Scaling Map (60).

20. The method according to any of claims 11 - 19, further characterized in that ADAPTING Determining Scaling Map (S3000) comprises executing its function as a cloud or central processing remotely from the control system (20), and wherein the scaling map (60) is updated from a cloud (90) interface (91), for allowing remote retrieval of scaling map (60) data the actual identity (101), damper group(100), damper unit (30, 50), face dampers (32, 34, 36, 38), bypass damper (52)21 . The method according to claim any of claims 11 - 20, further characterized in that ADAPTING (S3000) Determining the Scaling Map (60) comprises linearization control curves by analyzing stored actuator-damper air flow characteristics using artificial intelligence Convolution Neural Network (CNN), a Deep Learning Neural Network (DLNN), a Kalman filter, PID regulator, for adapting the compensating control signals, and to transform a linearized control signal to a control signal resulting in a linear proportional control of the air flow over the actuated dampers, for preparing the Scaling Map (60) with tables for look up of requested relative air flow (29, 49) for each damper (32, 34, 36, 38, 52), to retrieve damper actuation control signals (ACS) (63) that results in a desired relative air flow (64’).

22. The method according to claim any of claims 11 - 21 , characterized in that VERIFYING (S4000) comprises verifying a maintained precision of the scaling map's (60), precision in linearization of the actuated damper air flow control function, wherein the method step VERIFYING is triggered by a system event, error message, manual input, or a scheduling function, comprising: verifying scaling map at air flow as closed when at least one damper is closed as a reference, verifying scaling map at all control signal steps for ACS from min to max value, or during operation at available control range, while verifying and recording deviations from earlier recorded relations between RAF[%] as a result of ACS [V] or ACS as a digital control signal, verifying the scaling map when all dampers and air flow are fully open, and evaluating verification based on recorded deviations from previous calculated linearization scaling map.

23. The method according to any of claims 11 - 22 characterized in that the method is configured to operate in both a closed loop and open loop mode at the same time, wherein the system adjusts the scaling map based in actual measurements from the sensors involved in the feedback loop for estimating relative air flow (RAF).

24. An air flow controller (80) for regulating an Air Handling Unit (AHU), Heating, Ventilation, and Air Conditioning (HVAC), or Air Conditioning (AC) system, comprising: a computing hardware and software (80) configured to perform the method according to any preceding claim.

25. An air flow controller (80) for regulating an Air Handling Unit (AHU), Heating, Ventilation, and Air Conditioning (HVAC), or Air Conditioning (AC) system, comprising: a computing hardware and software (80), an interface to a face damper relative air flow (RAF) request (29) and / or an interface to a face damper control (24) configured to control at least one face damper actuator (25) configured to steer an at least one face damper (32, 34, 36, 38),an interface to a bypass damper relative air flow (RAF) request (49) and / or an interface to a bypass damper control (44) configured to control at least one bypass damper actuator (45) configured to steer an at least one bypass damper (52), an interface to a face damper steering (28), an interface to a bypass damper steering (48), an interface for sensing face damper flow (66’) configured to receive data from an air flow, pressure, or differential pressure sensor (66) if connected, an interface for sensing bypass damper flow (66”) configured to receive data from an air flow, pressure, or differential pressure sensor (66) if connected, a scaling map (60), configured to look up actual control signals (ACS) for face damper steering (28) and / or bypass damper steering (48) values, in the scaling map that predicts a desired face damper relative air flow (RAF) (66’) and / or a desired bypass damper relative air flow (RAF) (66”), an Manual input or Air flow Control System Mode (21) configured to establish an operation state (S1000-S5000) comprising the operational states:OPEN LOOP NON-LINEAR,CLOSED LOOP WITH CONSTANT FLOW,LEARNING,DETERMINING SCALING MAP,VERIFYING, ANDLINEAR OPERATION, wherein the air flow controller (80) is configured, to when set in operation state=LINEAR OPERATION: to receive: the face damper relative air flow (RAF) request (29) and the bypass damper relative air flow (RAF) request (49), and use the scaling map to look up actual control signals (ACS) stored in the scaling map (60) that predicts the requested face damper relative air flow (RAF) (66’) and desired bypass damper relative air flow (RAF) (66”), and to use the looked up actual control signals (ACS) to control the face damper steering (28), and the bypass damper steering (48), to steer any connected HVAC, AHU, or AC system to generate: a requested faced damper flow (66) is equal to the face damper relative air flow (RAF) request (29), and a requested bypass damper flow (66”) is equal to the bypass damper relative air flow (RAF) request (49).

Images