BACKGROUND OF THE INVENTION

The present invention relates to hydronic systems. It refers to a hydronic system according to the preamble of claim 1 .

It further refers to a method for operating such a hydronic system. PRIOR ART

Hydronic systems are part of the HVAC sector. In most cases, such hydronic systems comprise one or more control valves, which are used to control the flow of a fluid (liquid or gaseous) through a piping, which connects various parts of the hydronic system.

Related to these control valves is the well-known concept of so-called "valve authority".

As shown in Fig. 2 a hydronic system 20 in a general form comprises in a closed circuit a pump 1 1 , a two-port control valve 1 2 and a terminal unit, in this case a heat exchanger 1 3. Pump 1 1 , control valve 1 2 and terminal unit 1 3 are connected in series. When pump 1 1 pumps the fluid through the circuit with a certain pressure, there are pressure drops Δρ in the various sections of the system. These pressure drops or differential pressures can be divided into a first differential pressure Ap _va ive at the control valve 1 2 and a second differential pressure Aporcuit at the rest of the circuit (see Fig. 2). Now, when such a hydronic system 20 is commissioned, control valve 1 2 has to be chosen in accordance with the needs of the system:

When control valve 1 2 is undersized, the pressure drop of the entire system is increased which means that pump 1 1 would use a larger amount of energy to pump sufficient fluid through the system. On the other hand the accuracy of the control is increased as the entire control range of the valve may be utilized to achieve the desired control.

When the control valve is oversized, the amount of energy needed to pump the necessary flow through the system would be reduced. However, such energy savings will come at the cost of a decrease in control accuracy at the control valve 1 2, as the initial travel of the valve from fully open towards a more closed position would have no effect on the fluid flow. Thus, only a relatively small fraction of the entire control range of valve 1 2 is useful for control leading to an insufficient control'with poor stability and accuracy.

Thus, a trade off exists between the above two scenarios; and a proper sizing of control valve 1 2 requires a compromise between control accuracy and reduction of energy loss. This is, where the valve authority concept comes into play.

The valve authority N of a control valve like control valve 1 2 is defined as: where Ap _va ive is the pressure drop across the fully open control valve, Aporcuit is the pressure drop across the remainder of the circuit, kvs _va ive is the flow coefficient (in metric units) of the fully open control valve, and kvdrcuit is the respective flow coefficient of the remainder of the circuit outside the control valve. In other words, the valve authority N within a hydronic system indicates how much of the system's total pressure drop comes from the control valve. In practice a range of the valve authority N between 0.2 and 0.5 is considered acceptable. In accordance with equation ( 1 ) above, if valve authority N is too high (above 0.5 or 50%), then the control valve is likely to be undersized and so the hydronic system would benefit from a larger size valve in order to reduce losses that are driven by excessive pressure drop. If the value is too low (below 0.2 or 20%), then the valve movements will have only a marginal impact relative to the total system and hence the valve is likely to be oversized, yielding poor control.

In general, the flow coefficient kv of a part x of a hydronic system (as used in equation ( 1 )) is defined by the relation

for water as the fluid, having a specific gravity G= 1 , wherein Φ is the fluid flow through the part, and Δρ _χ is the pressure drop across part x.

Accordingly,

Φ

3) circuit

and

Φ

4) kv. valve Valve authority N has been used in the past in control schemes in an HVAC environment.

Document US 5,579,993 A is directed to a controller implemented in a heating, ventilation and air-conditioning (HVAC) distribution system, which provides improved control by implementing a general regression neural network (GRNN) to generate a control signal based on identified characteristics of components utilized within the HVAC system.

The local controller disclosed in US 5,579,993 A includes a feedforward means for generating a feedforward control signal based on the identified characteristics of a local component (e.g. damper or valve) and calculated system variables and a feedback means for generating a feedback control signal based on measured system variables. The controller then controls the component based on a combination of the feedforward control signal and the feedback signal.

The local controller comprises two separate processes, an identification process and a control process. The identification process identifies certain characteristics of the local component. These identified characteristics are output to the control process. The control process accepts the identified characteristics, along with other signals, and outputs a control signal so as to provide global control of the HVAC system

Especially, the identification process utilizes a look-up table to store characteristics of the local component. These characteristics are the ratio of the pressure drop across the local component to the branch pressure drop when the component is fully open (vale authority in case of a valve), the percentage of flow through the component normalized to the maximum flow through the component. The control process is divided into a feedforward process and a feedback process. The feedback process accepts as input a calculated flow setpoint and also a feedforward control signal. These signals are utilized by the feedback process to generate a control signal.

The feedforward process starts by first receiving the fan static pressure setpoint. The fan static pressure setpoint is used to calculate the pressure loss for each of the i branches connecting the fan outlet and the individual local damper. Especially, the pressure loss for each of the i branches is determined adaptively, in real-time. To calculate the pressure loss for branch 1 , certain calculating steps are followed. The next step is to calculate the pressure loss of a second segment. This pressure loss is added to the pressure loss for the first segment to yield the pressure loss for the branch 1 . This method of calculating pressure loss applies for i additional branches connected to the main duct.

Document US 6,095,426 generally relates to control systems, and more particularly to control systems that are used in heating, ventilating and air conditioning fluid distribution systems.

US 6,095,426 discloses a controller for controlling the temperature within a room in a building having at least one space adjacent to the room, the building having a heating, ventilating and air conditioning (HVAC) system with a supply duct adapted to supply air to the room and a general exhaust duct adapted to exhaust air from the room. The system has a local component for controlling the supply air flow into the room, the room having at least one additional exhaust independent of the HVAC system. The controller comprises a feedforward means for generating a feedforward control signal based on a desired temperature and flow set points in the supply duct, the flow into and out of the room, the flow set point in the general exhaust duct, and based on identifying characteristics of the component and calculated system variables. The controller further comprises a feedback means for generating a feedback control signal based on measured system variables, and means for combining the feedforward control signal and the feedback control signal to achieve control of the local component.

US 6,095,426 also discloses a method of determining the value of a control signal in a controller for controlling the outlet air temperature from an air supply duct to a room, the air supply duct being part of an HVAC system of a building, the air duct having a heating coil adapted to heat the air moving through the duct and a flow valve for controlling the flow of hot water through the heating coil. The controller is of the type which has an identification means for periodically producing identified characteristics of the heating coil and valve and means for measuring the temperature of the air at the outlet of the duct, means for measuring the air flow rate through the duct and means for measuring the water pressure across the valve and in the system in which the valve is connected. The control signal is based on control set points and the identified characteristics of the heating coil and valve. The method comprises the steps of activating said identification means to determine the effectiveness of the coil in transferring heat to the air flowing through the duct, utilizing said coil characteristic to yield a desired water flow rate through the heating coil for a given measured duct outlet air temperature and air flow rate, measuring the pressure drop across the valve to the overall pressure drop in the system when the valve is fully open and determining the ratio of the former to the latter to derive the authority value for the valve, and generating said control signal as a function of the water flow rate and the valve authority.

Document EP 1 235 131 B1 discloses a process of controlling the room temperature, comprising a first temperature sensor for metering the room temperature, a second temperature sensor for metering the lead temperature of a heating medium, a third temperature sensor for metering the return temperature of the heating medium, and a control unit for actuating a valve for the flow of the heating medium. Within this process the operating characteristic of the valve is determined from the measured values of temperature sensors for the room, lead and return temperatures, with the control parameters of the room temperature control being adjusted to the operating characteristic in response to the point of operation of the valve.

Document CN 1 05335621 A relates to an electric adjusting valve model selection method. The electric adjusting valve model selection method comprises the following steps: determining a use performance of an electric adjusting valve, selecting a flow property curve type of the valve according to the use performance, primarily selecting the diameter of a valve seat; according to the primarily-selected diameter of the valve seat, inquiring a design manual to obtain a valve adjustable ratio R, a flowing capability kv of the valve and valve authority S, determining the maximum aperture value K—90% and the minimum aperture value K= ^: 30% of the diameter of the valve seat; substituting the parameters including the R, kv, S, K=90% and K=30% into an actual flow property formula of the electric adjusting valve respectively to obtain a flow under the 30% aperture and a flow under the 90% aperture; determining whether a flow range Q mir Qmax of a cooling water system connection pipe ranges from Q ₃ o% to Q ₉₀ % or not; if the Qmin-Qmax ranges from Q3o% to Qgo%, finishing model selection; and if the Qmin-Qmax does not range from Q ₃₀ % to Q9o%, returning back to the step of primarily selecting the diameter of the valve seat and continually carrying out the model selection until the diameter of the valve seat meets the conditions. According to the method provided, model selection parameters of the valve and operation conditions of a cooling water system are matched, so that the valve can express a relatively good adjusting performance.

In general, a poor valve authority leads to poor system control and instability. Another problem is the so-called "hunting": The control of a hydronic circuit may be prone to unwanted oscillations, which also lead to poor system control and instability.

Document WO 2006/1 05677 A2 discloses a method and a device for suppressing vibrations in an installation comprising an actuator for actuating a flap or a valve used for metering a gas or liquid volume flow, especially in the area of HVAC, fire protection, or smoke protection. Vibrations of the flap or valve caused by an unfavorable or wrong adjustment or configuration of the controller and/or by disruptive influences are detected and dampened or suppressed by means of an algorithm that is stored in a microprocessor. Said algorithm is preferably based on the components recognition of vibrations, adaptive filtering, and recognition of sudden load variations.

SUMMARY OF THE INVENTION

It is an object of the invention, to provide a hydronic system, which avoids certain disadvantages of known hydronic systems and is in a simple way able to adapt to changes in hydraulic parameters of the system.

In it another object of the invention to teach a method for operating such a system.

These and other objects are obtained by claims 1 , 1 0, 1 1 , 1 6 and 1 9.

The hydronic system according to the invention comprises at least one hydronic circuit and a control for controlling the operation of said at least one hydronic circuit via a control path, whereby said control comprises a feed forward controller.

It is characterized in that said hydronic system further comprises a control improvement path running from said at least one hydronic circuit to said control, by means of which control improvement path said control can be improved in the case of said hydronic system becoming instable and/or showing poor system control.

According to an embodiment of the invention said at least one hydronic circuit comprises a control valve as a variable flow resistance and a static flow resistance, which are connected in series by a piping, whereby said control valve is controlled by a valve control device, in that a flow sensor is provided for measuring the flow of a fluid flowing through said circuit, and in that a valve authority determining device is associated with said hydronic circuit, whereby said valve authority determining device is connected to said valve control device in order to receive information about the actual opening position of said control valve, and whereby said valve authority determining device is further connected to said flow sensor in order to receive information about the actual fluid flow flowing through said circuit.

A storage may be associated with said valve authority determining device, which storage contains and provides said valve authority determining device with, information on a valve characteristic of said control valve.

Also, an outlet of said valve authority determining device may be connected to said feed forward controller.

According to an embodiment of the invention a frequency detector for detecting oscillations is provided in said hydronic system, and said frequency detector is in operative connection with said control.

Said control may comprise oscillation suppressing means, and said frequency detector may be in operative connection with said oscillation suppressing means. Furthermore, said feed forward controller may comprise a physical model of said hydronic circuit, and that said oscillation suppressing means may have an effect on input and/or output signals of said physical model.

Especially, said oscillation suppressing means may comprise at least one filter.

According to another embodiment of the invention said control may comprise an alternative controller, and said frequency detector may be in operative connection with switching means for switching between said feed forward controller and said alternative controller.

A method for operating a hydronic system according to the invention, which comprises a control valve as a variable flow resistance and a static flow resistance, which are connected in series by a piping, whereby said control valve is controlled by a vaive control device, in that a flow sensor is provided for measuring the flow of a fluid flowing through said circuit, and in that a valve authority determining device is associated with said hydronic circuit, whereby said valve authority determining device is connected to said valve control device in order to receive information about the actual opening position of said control valve, and whereby said valve authority determining device is further connected to said flow sensor in order to receive information about the actual fluid flow flowing through said circuit, comprises the steps of a. providing a valve characteristic of said control valve, which comprises the dependency of the flow coefficient (kv) of said valve on the opening position of said valve; b. moving said control valve into a first opening position having a first flow coefficient ( kVvalvej ) ; c. measuring the flow ( Φ ₁ ) of said circulating fluid through said control valve in said first opening position; d. moving said control valve into a second opening position having a second flow coefficient (kv _va ive,2); e. measuring the flow ( Φ ₂ ) of said circulating fluid through said control valve in said second opening position; f. determining from said measured flows ( Φ, ,Φ ₂ ) and the respective flow coefficients ( kvvaive.1 , kv _va ive,2) the valve authority (N) using the formula

with

Av circuit

and kvsvaive being the flow coefficient of the fully opened valve.

Another method for operating a hydronic system according to the invention, which comprises a control valve as a variable flow resistance and a static flow resistance, which are connected in series by a piping, whereby said control valve is controlled by a valve control device, in that a flow sensor is provided for measuring the flow of a fluid flowing through said circuit, and in that a valve authority determining device is associated with said hydronic circuit, whereby said valve authority determining device is connected to said valve control device in order to receive information about the actual opening position of said control valve, and whereby said valve authority determining device is further connected to said flow sensor in order to receive infornnation about the actual fluid flow flowing through said circuit, comprises the steps of: a. Providing a shape of a valve characteristic of said control valve, which comprises the principal dependency of the flow coefficient (kv) of said valve on the opening position of said valve; b. moving said control valve into a first opening position; c. measuring the flow ( Φ ) of said circulating fluid through said control valve in said first opening position; d. moving said control valve into a second opening position different from said first position; e. measuring the flow ( Φ ₂ ) of said circulating fluid through said control valve in said second opening position; f. moving said control valve into a third opening position different from said first and second opening position; g. measuring the flow ( Φ ₃ ) of said circulating fluid through said control valve in said third opening position; h. determining from the three measured flows ( Φ, , Φ ₂ , Φ ₃ ) the flow coef ^¬ ficients of the circuit, kv _d rcuit, and the fully opened control valve ( 1 2), kvs _va ive; and i. determining the valve authority (N) of said control valve ( 1 2) using the formula

(^arcultf +( ) ²

Said valve authority may be determined at predetermined times during the lifetime of said hydronic system.

Especially, said valve authority may be determined during a commissioning of said hydronic system.

In addition, said valve authority may be determined at least a second time during the lifetime of said hydronic system. Furthermore, said valve control device may comprise a feed-forward part, and said determined valve authority may be used as a parameter in said feed-forward part of said valve control device.

Another method for operating a hydronic system according to the invention, wherein a frequency detector for detecting oscillations is provided in said hydronic system, and said frequency detector is in operative connection with said control, comprises the steps of: a. monitoring a flow through said hydronic system and/or a set point signal by means of said frequency detector; b. acting on said control, when an oscillation is detected by said frequency detector. Especially, oscillation suppressing means may be activated in said control, when an oscillation is detected by said frequency detector.

Alternatively, said feed forward controller may be replaced by an alternative controller, when an oscillation is detected by said frequency detector.

Another method for operating a hydronic system according to the invention, wherein a frequency detector for detecting oscillations is provided in said hydronic system, and said frequency detector is in operative connection with said control, and wherein said control comprises an alternative controller, and said frequency detector is in operative connection with switching means for switching between said feed forward controller and said alternative controller, comprising the steps of: a. monitoring a flow through said hydronic system and/or a set point signal by means of said frequency detector; b. replacing said alternative controller by said feed forward controller, when an oscillation is detected by said frequency detector.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is now to be explained more closely by means of different embodiments and with reference to the attached drawings.

Fig. 1 shows in a generalized configuration a hydronic system according to an embodiment of the invention comprising a hydronic circuit and a control interacting by means of a control path and a control improvement path; shows a basic hydronic circuit comprising a pump, a control valve and a heat exchanger; shows a "learning" hydronic system according to an embodiment of the invention based on the circuit of Fig. 2 and further comprising control means capable of reacting to changes of certain parameters of the hydronic circuit by valve authority learning; shows a diagram related to a first method of authority learning used in the present invention; shows a diagram related to a second method of authority learning used in the present invention; shows a "learning" hydronic system according to another embodiment of the invention; shows a feed forward control scheme, which may be used to implement the valve authority learning method according to the present application; shows a modified feed forward control scheme, which may be used to suppress unwanted oscillations of the system; and shows another way of dealing with unwanted oscillations of the system by switching between different controllers. DETAILED DESCRIPTION OF DIFFERENT EMBODIMENTS OF THE INVENTION

Fig. 1 shows in a generalized configuration a hydronic system HS according to an embodiment of the invention. Hydronic system HS comprises a hydronic circuit HC, which is usually associated with a building and includes piping, valves, heat exchangers, pumps, and the like, and a control CT interacting by means of a control path CP and a control improvement path CIP. Control path CP is related to the communication between control CT and hydronic circuit, and is the path for exchanging control signals from control CT to hydronic circuit HC, and operating parameters from hydronic circuit HC to control CT. Control CT comprises a feed forward controller FFC, which contains a physical model of hydronic circuit HC. Control CT may further comprise an alternative controller AC, which may replace feed forward controller FFC, and vice versa. The switching between the two controllers FFC and AC is symbolized in Fig. 1 by a selector switch.

An improvement of the control may be achieved in different ways, depending on the situation in the hydronic circuit:

• During commissioning of the system it may be necessary and/or advantageous to adapt the control CT to certain parameters of the system, which were unknown prior to commissioning.

• Operation of the system for a longer time may result in a change of important system parameters and/or a degradation, which may lead to poor system control and instability.

There are especially two cases, which are of concern with regard to the controllability of the hydronic system: 1 ) As long as control valves are part of the hydronic circuit, the so-called "valve authority" is an important parameter: Poor valve authority leads to poor system control and instability

2) Sometimes hydronic systems are prone to undesirable oscillations, so-called "hunting": Hunting signals, too, lead to poor system control and instability.

According to the invention, negative implications of a change of valve authority over time or an insufficient knowledge of the actual valve authority will be avoided by a respective improvement of the control.

As has been already described in the introductory part Fig. 2 shows a hydronic system 20 in its most general form, which comprises in a closed circuit a pump 1 1 , a two-port control valve 1 2 and a terminal unit, in this case a heat exchanger 1 3. Pump 1 1 , control valve 1 2 and terminal unit 1 3 are connected in series. When pump 1 1 pumps the fluid through the circuit with a certain pressure, there are pressure drops Δρ in the various sections of the system. These pressure drops or differential pressures can be divided into a first differential pressure Ap„ _a iv _e at the control valve 1 2 and a second differential pressure Apdrcuit at the rest of the circuit.

In such a circuit the valve authority N is the pressure drop across the fully open valve in relation to the pressure drop across the whole system. Valve authority N, which is defined by equations ( 1 ) to (4) above, indicates how good the hydronic system is controllable (the higher the valve authority N, the better the hydronic system can be controlled). However, valve authority N is not a parameter, which is constant through the lifetime of the system. When valve authority N changes as a result of changes in the system, it will be advantageous to have a valve authority learning capability of the system in order to adapt the control mechanism of the system to the changing system environment.

The present invention deals with such valve authority learning.

Within the scope of the present invention at least two different procedures of valve au- thority learning are possible. Both of them include active measurements at the valve in the hydraulic circuit, meaning the valve is actively moved between different valve positions.

A first of these at least two different procedures is chosen, when the whole valve characteristic is known. In this case the curve kv vs. valve position shown in Fig. 4(a) is known. Further, as a primary assumption, there shall be a constant pressure across the relevant zone of the system.

To evaluate the actual valve authority of control valve 1 2, the valve is moved to two different positions. These positions are in Fig. 4 characterized through two respective kv- values, namely kv _va ive.i (for position 1 ) and kv _va ive,2 (for position 2). In each of these two positions the related flow Φ is measured (Fig. 4(b)) and stored together with its associated kv-value, thus giving pairs of values Φ _ρ /ί ^, and ₂ , /< ' _{3/ 82} .

Based on these pairs of values, the actual valve authority N can be calculated by means oft he following formulas: kv ( ' - Φ, ² )

circuit ^'

(6) Φ Φ

In?

n ^v valve, ™valve

A second of these at least two different procedures is chosen, when only the shape of the valve characteristic is known, but no scaling is available. In this case the curve kv vs. valve position (shown in Fig. 5(a)) is a function F(kvs, n _g i) of the value kvs and a parameter n _g i, which is a measure of how sharply the characteristic curve is curved. For example, when the curve represents a valve with an "equal percentage characteristic", Curves with other values of n _g i are shown in Fig. 5(a) with dash and dot-and-dash lines.

Again, as a primary assumption, there shall be a constant pressure across the relevant zone of the system.

Now, the valve is moved to three (different) positions (Fig. 5(b)). The respective flows Φ, , Φ ₂ and Φ ₃ are measured at these positions and stored.

Finally, an equation system with 3 unknowns kvdrcuit, kvs _va ive and Δρ can be solved using the stored flows.

To move control valve 1 2 into the different positions and measure the respective flow circulating through piping 1 9 and said valve a valve control device 14 and a flow sensor 1 8 are provided in a hydronic system 1 0 in accordance with Fig. 3. Both devices are connected to a valve authority determining device 1 6, which controls the measuring action of control valve 1 2 and flow sensor 1 8. Storage 1 5 may be used to store certain parameters of the valve characteristic, which are necessary for a valve authority calculation, as explained above. The valve authority measured and calculated by valve authority deter- mining device 1 6 from time to time may be transferred to a valve authority using unit 1 7, as indicated. Valve authority unit 1 7 then may control valve control device 14, according- ly-

Valve authority N may be determined at predetermined times during the lifetime of hy- dronic system 1 0. Furthermore, valve authority N may be determined during a commissioning of hydronic system 1 0, and, preferably, at least a second time during the lifetime of said hydronic system.

As valve control device 14 comprises (besides a possible feedback) a feed -forward part 23, as shown in Fig. 6, said determined valve authority N may be used as a parameter in feed-forward part 23 of valve control device 14.

Hydronic circuit 1 0, as shown in Fig.6, may be a simple circuit with a pump 1 1 , a control valve 1 2 and a heat exchanger 1 3. However, there may be further circuit elements 21 and branches comprising further piping 1 9' and circuit elements 22.

Finally, the arrangement of control valve 1 2, valve control device (or actuator) 14, flow sensor 18 and valve authority determining device 1 6 and storage 1 5 may be combined in one unit, which is known as "energy valve" EV (see for example EP 2 896 899 A1 ).

Fig. 7 shows a feed forward control scheme, which may be used to implement the valve authority learning strategy explained so far. Central part of forward control scheme 24 of Fig. 7 is a physical model 27 of the hydronic system in question. When a flow set value F _sv is given, the physical model 27 generates a feed forward position set value PS _Fsv by using flow set value F _sv , the measured actual flow, F, valve authority 28 and other input parameters 29, e.g. the valve characteristic. Added to said feed forward position set value PS _FS v is a deviation of valve position set value, APS _SV , which is determined by deviation part 30 from the difference between flow set value F _sv and measured actual flow F. Deviation part 30 comes up for small deviations due to a mismatch of physical model 27 and reality. The sum of PSF _S V and APS _SV is finally used as valve position set value PS _S v for controlling the controlled system flow 25. The resulting actual flow F is measured by flow sensor 26.

The valve authority 28 put into the physical model 27 is the valve authority determined by the methods explained above. In this way the feed forward control can react to changes of this relevant system parameter in order to improve system control and stability.

However, as already mentioned above, other characteristics of the system than valve authority may trigger an action on the feed forward control scheme. For example, document WO 2006/1 05677 A2 discloses a method and a device for suppressing vibrations in an installation comprising an actuator for actuating a flap or a valve used for metering a gas or liquid volume flow, especially in the area of HVAC, fire protection, or smoke protection. Vibrations of the flap or valve caused by an unfavorable or wrong adjustment or configuration of the controller and/or by disruptive influences are detected and dampened or suppressed by means of an algorithm that is stored in a microprocessor. Said algorithm is preferably based on the components recognition of vibrations, adaptive filtering, and recognition of sudden load variations.

Specifically, according to the document, a regulating variable from the regulating path is provided, whereby said regulating variable corresponding to the effective liquid volume flow. Further, a predefined control signal corresponding to the required liquid volume flow is provided. The predefined control signal and the regulating variable are compared and a regulator output variable is calculated therefrom. The regulator output variable is monitored by a vibration detection algorithm. If the vibration detection algorithm does not detect vibrations of the regulator output variable, the regulator output variable is fed to an actuating device which is actuating a flap or a valve in the pipe for dosing the gas or liquid volume flow. If, on the other hand, the vibration detection algorithm detects vibrations of the regulator output variable, the regulator output variable is fed to an adaptive filter and the adaptive filter suppresses the vibration and generates a control signal with suppressed or damped vibrations of the regulator output variable, which is then used at the actuating device instead of the regulator output variable.

In the present case of a feed forward control scheme the situation is different: As shown in Fig. 8, a modified feed forward control scheme 24a may be used to suppress unwanted oscillations of the system. To detect unwanted oscillations of the system, a frequency detector 31 may be connected to flow sensor 26. When frequency detector 31 detects unwanted oscillations in the system, appropriate input or output signals of the physical model 27 will be compensated or filtered (e.g. with lead or lag filters or a combination thereof). As an example, Fig. 8 shows two filters 32 and 33 (dashed lines) at the input of the flow set value F _sv and the output of floe signal F of flow sensor 26. A further filtering means 35 may be used to filter the valve position set value (PS _SV ). Other locations of filtering and/or compensating are possible.

In addition, setpoint signals flow set value F _sv and/ or position set value PS _SV may be monitored by frequency detector 31 .

Another way of dealing with unwanted oscillations of the system is shown in Fig. 9: When unwanted oscillations are detected by frequency detector 31 of feed forward control scheme 24b, it actuates a disabling means 34 (e.g. a switch) to shut down the actual feed forward control and replace it with an alternative, more suitable and oscillation-free controller AC. The switching may be reversed in other situations, so that the system switches from an alternative controller AC to a feed forward controller to improve stability and control.

List of reference numerals

1 0,20 hydronic circuit

1 1 pump

1 2 control valve

1 3 heat exchanger

14 valve control device (or actuator)

1 5 storage

1 6 valve authority determining device

7 valve authority using unit

8 flow sensor

9, 19' piping

1 ,22 circuit element

3 feed-forward part (valve control device)

4 feed forward control scheme

4a,b feed forward control scheme

5 controlled system flow

6 flow sensor

7 physical model

8 valve authority

9 other input parameters (e.g. valve characteristic)

0 deviation part

1 frequency detector 32,33 filter

34 disabling means (e.g. switch)

35 filtering means

AC alternative controller

CIP control improvement path

CP control path

CT control

EV energy valve

F flow

Fsv flow set value

FFC feed forward controller

HC hydronic circuit

HS hydronic system

APSsv deviation of valve position set value PSsv valve position set value

PSF _SV feed forward valve position set value kvvaive flow coefficent of control valve

Φ flow through control valve

Ap ai e pressure drop at control valve

Apcircuit pressure drop at circuit outside control valve

♦