Method for fault-tolerant control of a vapor compression system

Abstract

A method for controlling a vapour compression system (1) is disclosed, the vapour compression system (1) comprising an ambient temperature sensor (8) arranged to measure an ambient temperature. A time period is selected during which the ambient temperature sensor (8) is not exposed to solar heating. During the selected time period, a measurement of the ambient temperature is obtained by the ambient temperature sensor (8) and a measurement of at least one other parameter related to the vapour compression system (1) is obtained while the vapour compression system (1) is operated. Model parameters of a model for at least a part of the vapour compression system (1) are derived based on the obtained measurements, the model providing a correlation between the ambient temperature and the at least one other parameter. Subsequently, the vapour compression system (1) is operated based on measurements of the at least one other parameter and based on the ambient temperature derived by the model comprising the derived model parameters.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method for controlling a vapour compression system, the vapour compression system comprising an ambient temperature sensor arranged to measure an ambient temperature. The method according to the present invention is fault-tolerant with respect to solar heating of the ambient temperature sensor. BACKGROUND

[0002] Vapour compression systems are typically controlled based on measurements made by relevant sensors arranged in the vicinity of the vapour compression system or forming part of the vapour compression system. Such sensors can for example be pressure sensors and / or temperature sensors measuring the pressure or temperature of a refrigerant flowing in a refrigerant path of the vapour compression system at selected positions along the refrigerant path. Alternatively or additionally, temperature sensors can be arranged to measure the temperature in one or more refrigeration volumes and / or the ambient temperature. In any case, reliable sensor measurement results are essential in order to ensure that the vapour compression system is operated in an appropriate manner, which provides the required cooling or heating while minimizing energy consumption.

[0003] Accordingly, it is desirable to be able to determine whether a sensor is providing faulty measurements and to be able to mitigate or compensate for such faulty measurements. For example, the ambient temperature has an influence on the temperature and pressure conditions in the vapour compression system. More particularly, if the ambient temperature changes, the optimal refrigerant temperature and pressure values in the vapour compression system also change. Hence, the setpoint values for refrigerant pressure and / or refrigerant temperature applied when controlling the vapour compression system are sometimes calculated based on measurements of the ambient temperature. Accordingly, a faulty measurement of the ambient temperature can result in non-optimal setpoint values for the refrigerant pressure and / or the refrigerant temperature and thereby in non-optimal operation of the vapour compression system. Accordingly, it is particularly meaningful to obtain reliable measurements of the ambient temperature.

[0004] The ambient temperature sensor is typically arranged in an outdoor environment. Accordingly, the ambient temperature sensor is subjected to various weather conditions, including solar radiation. If the ambient temperature sensor is positioned in direct sunlight, or close to a surface heated by solar radiation (e.g. a roof, a metal shield, etc.), there is a risk that the ambient temperature sensor is heated. This can result in sensor measurements from the ambient temperature sensor indicating a higher, sometimes much higher, ambient temperature than the actual ambient temperature. It can be desirable to avoid this, or at least to be able to mitigate or compensate for this. SUMMARY

[0005] It is an object of embodiments of the present invention to provide a method for controlling a vapour compression system, wherein proper operation of the vapour compression system is ensured regardless of ambient conditions.

[0006] It is a further object of embodiments of the invention to provide a method for controlling a vapour compression system, wherein a faulty measurement of an ambient temperature can be mitigated and / or compensated.

[0007] The invention provides a method for controlling a vapour compression system, the vapour compression system comprising at least one compressor, a heat rejecting heat exchanger, at least one expansion device and at least one evaporator arranged in a refrigerant path, each evaporator being arranged in thermal contact with a refrigeration volume, and each expansion device being arranged to supply refrigerant to an evaporator, the vapour compression system further comprising an ambient temperature sensor arranged to measure an ambient temperature, the method comprising the steps of:

[0008] - selecting a time period during which the ambient temperature sensor is not exposed to solar heating,

[0009] - obtaining, during the selected time period, a measurement of the ambient temperature by the ambient temperature sensor and obtaining a measurement of at least one other parameter related to the vapour compression system while operating the vapour compression system,

[0010] - deriving, based on the obtained measurements, model parameters of a model for at least a part of the vapour compression system, the model providing a correlation between the ambient temperature and the at least one other parameter, and

[0011] - subsequently operating the vapour compression system based on measurements of the at least one other parameter and based on the ambient temperature derived by the model comprising the derived model parameters.

[0012] Thus, the invention provides a method for controlling a vapour compression system. In the context herein, the term “vapour compression system” shall be interpreted to mean any system in which a fluid medium flow, such as a refrigerant, is circulated and alternately compressed and expanded, thereby providing cooling or heating of a volume. Thus, the vapour compression system can be a refrigeration system, an air conditioning system, a heat pump, etc.

[0013] The vapour compression system comprises at least one compressor, a heat rejecting heat exchanger, at least one expansion device and at least one evaporator arranged in a refrigerant path. Each expansion device is arranged for supplying refrigerant to an evaporator. Accordingly, in case the vapour compression system comprises two or more evaporators, the refrigerant supplied to each evaporator can be controlled individually by means of the corresponding expansion device. Further, each evaporator is arranged in thermal contact with a refrigeration volume. The refrigeration volume(s) can for example be a display case in a supermarket, a room of a building provided with an air conditioning system, or any other suitable type of volume that needs to be cooled.

[0014] Accordingly, the refrigerant flowing in the refrigerant path is compressed by the compressor before being supplied to the heat rejecting heat exchanger. When passing through the heat rejecting heat exchanger, heat exchange takes place between the refrigerant and the secondary fluid stream across the heat rejecting heat exchanger in such a way that heat is rejected from the refrigerant. Thereby, the temperature of the refrigerant is reduced. The heat rejecting heat exchanger can be in the form of a condenser, in which case the refrigerant is at least partially condensed. Alternatively, the heat rejecting heat exchanger can be in the form of a gas cooler, in which case the refrigerant is cooled, but remains trans-critical or sub-cooled.

[0015] The heat rejecting heat exchanger can be arranged in an outdoor environment, such as on a roof or mounted on an outer wall of a building. In this case, the secondary fluid stream across the heat rejecting heat exchanger can be an ambient air stream driven by one or more fans.

[0016] The refrigerant leaving the heat rejecting heat exchanger can be supplied to the expansion device(s) via a high-pressure expansion device and a receiver, the refrigerant being subjected to expansion at the expansion device(s) before being supplied to the respective evaporator(s). Thereby, the refrigerant supplied to the evaporator(s) is in a gas-liquid mixture state. In the evaporator(s), heat exchange takes place between the refrigerant and the secondary fluid stream in the respective refrigeration volume in such a way that heat is absorbed by the refrigerant, while the liquid portion of the refrigerant is at least partially evaporated. Thereby, the refrigeration volume(s) is / are cooled.

[0017] Finally, the refrigerant is again supplied to the compressor(s).

[0018] Accordingly, the refrigerant flowing in the refrigerant path is alternately compressed by the compressor(s) and expanded by the expansion device(s), while heat exchange takes place in the heat rejecting heat exchanger and the evaporator(s).

[0019] The vapour compression system further comprises an ambient temperature sensor arranged to measure an ambient temperature. The ambient temperature sensor can be located outside a building housing the vapour compression system, e.g. on a roof or mounted on an outer wall of the building. For example, the ambient temperature sensor can be arranged in the vicinity of the heat rejecting heat exchanger. Thereby, the ambient temperature sensor is exposed to weather conditions, including solar radiation.

[0020] In the method according to the application, initially a time period is selected during which the ambient temperature sensor is not exposed to solar heating. As mentioned above, in case the ambient temperature sensor is arranged in direct sunlight, or close to a surface which is heated due to exposure to solar radiation, there is a risk that the temperature sensor is heated, in which case the temperature measurements provided by the ambient temperature sensor can indicate a temperature which is higher, possibly much higher, than the actual ambient temperature. Such measurements are generally considered to be faulty measurements. However, when the ambient temperature sensor is not exposed to solar heating, it can be assumed that the measurements provided by the ambient temperature sensor are reliable. Thereby, the time period during which the ambient temperature sensor is not exposed to solar heating can be considered to be a time period during which the ambient temperature sensor is operating under fault-free conditions.

[0021] It should be noted that in the context of the present document, the term "not exposed to solar heating" should be interpreted to cover both the case that the ambient temperature sensor is not located in direct sunlight, and the case that the structure in the proximity of the ambient temperature sensor is not heated by incident solar radiation to such an extent that it causes an unintended heating of the ambient temperature sensor, as mentioned above. For example, the selected time period can be a time period during which the sun has already set, the ambient temperature sensor is arranged in the shade, a cloudy condition occurs, or any other suitable condition which ensures that the ambient temperature sensor is not heated due to solar radiation.

[0022] During the selected time period, measurements of the ambient temperature are obtained by the ambient temperature sensor. Since these measurements are obtained under fault-free conditions, it can be assumed that these measurements are reliable, i.e. they accurately reflect the actual ambient temperature which occurs. Furthermore, also during the selected time period, measurements of at least one other parameter related to the vapour compression system are obtained while the vapour compression system is operated. The other parameter can for example be a relevant temperature and / or pressure in the vapour compression system. This will be described in further detail below.

[0023] Thereby, while the vapour compression system is operated, simultaneous measurements of the ambient temperature and the at least one other parameter are obtained, and from these it can be derived a correlation between the ambient temperature and the at least one other parameter.

[0024] Next, based on the obtained measurement results, model parameters of a model for at least a part of the vapour compression system are derived. The model provides a correlation between the ambient temperature and at least one other parameter. This is possible because, as explained above, during operation of the vapour compression system, the measurement results provide correlated measurements of the ambient temperature and the at least one other parameter, and because the measurement results provided by the ambient temperature sensor are obtained under faultless conditions at least in terms of solar heating. Thus, the resulting model, including the derived model parameters, can be applied to subsequently derive some of the primary measurement parameters from measurement results of the other primary measurement parameters. For example, the ambient temperature can be derived from measurement results of the at least one other parameter by means of the model.

[0025] Thus, the vapour compression system is subsequently operated based on measurement results of the at least one other parameter and based on the ambient temperature derived by means of the model including the derived model parameters. Thereby, the vapour compression system can be operated based on reliable ambient temperature values, even in case the ambient temperature sensor is exposed to solar heating and thus the measurement results provided by the ambient temperature sensor are unreliable. Accordingly, the vapour compression system can be operated properly and in an energy efficient manner, even in conditions where it is not possible to obtain reliable measurement results from the ambient temperature sensor due to the ambient temperature sensor being exposed to solar heating.

[0026] The model can be a model reflecting behaviour of at least a part of the vapour compression system. The model can reflect behaviour of the entire vapour compression system, or only of a part of the vapour compression system, such as a particular component (e.g. a heat rejecting heat exchanger), the high pressure side or the low pressure side. For example, the model can reflect dynamic behaviour of at least a part of the vapour compression system, such as the behaviour of the system in response to various changes, such as changes in temperature level and / or pressure level in or near the vapour compression system. For example, the model can reflect dynamic behaviour of the vapour compression system in response to changes in ambient temperature.

[0027] As an alternative, the model can be a random model or another suitable "black box" model.

[0028] The model can be a model of at least the heat rejecting heat exchanger. For example, the model can reflect heat transfer from the refrigerant to a secondary fluid (e.g. in the form of an ambient air stream) flowing across the heat rejecting heat exchanger that occurs in the heat rejecting heat exchanger. This heat transfer is sensitive to changes in ambient temperature, and thus it makes sense to apply a model reflecting this when performing the method according to the present application.

[0029] The step of selecting a time period can comprise selecting a time period during night time. In the context of the present document, the term "night time" should be interpreted to mean the time period between the point in time at which the sun sets and the point in time at which the sun rises. Accordingly, by selecting a time period during night time as the fault-free time period, it is effectively ensured that the ambient temperature sensor is not exposed to solar heating, since the sun does not shine during night time.

[0030] Alternatively, the step of selecting a time period can comprise selecting a time period in which the ambient temperature sensor is in the shade, or selecting a time period in which a cloudy condition or a shaded condition is detected.

[0031] The step of selecting a time period can comprise the steps of:

[0032] - measuring, by the ambient temperature sensor, the ambient temperature during a continuous time interval of at least 24 hours,

[0033] - identifying time intervals with high ambient temperature as day time and time intervals with low ambient temperature as night time,

[0034] - calibrating the clock based on the identified day time and night time, and

[0035] - selecting a time period during night time based on the calibrated clock.

[0036] According to this embodiment, the distinction between time periods considered to be "day time" and time periods considered to be "night time" is empirically determined by continuously measuring the ambient temperature by the ambient temperature sensor during a continuous time period of at least 24 hours. Hence, the continuous time period encompasses at least one day and at least one night. It can be assumed that the ambient temperature during day time is generally higher than the ambient temperature during night time. Accordingly, time intervals with high ambient temperature are identified as day time and time intervals with low ambient temperature are identified as night time.

[0037] These findings are then applied to calibrate the clock, and subsequently a time period during night time can be selected based on the calibrated clock.

[0038] One advantage of this embodiment is that the method can be implemented in any vapor compression system without the need to provide information about local conditions such as time zone, duration of day or night, shaded conditions, etc. Instead, the method can be implemented in a "plug and play" fashion, wherein the system calibrates itself based on the measurements made by the available sensors.

[0039] As an alternative, the time period during the night time can be selected based on a clock calibrated to the time zone of the location where the vapour compression system is located, possibly in combination with information related to the sunset and sunrise times of that location.

[0040] The step of deriving the model parameters can comprise constructing a linear data driven model. In the context of this document, the term "data driven model" is to be interpreted to mean a model that allows finding a relationship between system variables, i.e. inputs and outputs, without explicit knowledge of the physical behaviour of the system. Instead, online system identification is performed. Accordingly, data driven models are suitable for certain systems where conditions are unknown. Correspondingly, applying a data driven model when performing the method according to the present application allows implementing the method in a "plug and play" fashion. Data driven models can be referred to as "black box" models. Data driven models are simple, ensuring fast computation times. Furthermore, convex optimisation leads to a guaranteed estimate of the best set of model parameters.

[0041] As an alternative, the model can be a non-linear data driven model, a stochastic model or any other suitable type of "black box" model.

[0042] As a further alternative, a "white box" model, i.e. a physical model in which physical parameters can be estimated, can be applied.

[0043] As yet a further alternative, a combination of a "white box" model and a data driven model or "black box" model can be applied. For example, the step of running the vapour compression system under fault-free conditions while estimating the ambient temperature can be considered a "white box" approach. Then, further simplifications can be applied by obtaining a linear model and estimating the model parameters using a data driven system identification method. This approach is easy to verify and ensures a robust and simple portability between different systems.

[0044] The at least one other parameter can comprise a fan speed of a fan driving the flow of secondary fluid across the heat rejecting heat exchanger and / or a temperature of the refrigerant leaving the heat rejecting heat exchanger.

[0045] The fan speed of the fan driving the flow of secondary fluid across the heat rejecting heat exchanger determines the flow rate of the secondary fluid, e.g. in the form of an ambient air flow. This has an influence on the heat transfer taking place in the heat rejecting heat exchanger. When the ambient temperature increases, it is also necessary to increase the fan speed to ensure sufficient heat transfer from the refrigerant to the secondary fluid, thereby maintaining the required temperature of the refrigerant leaving the heat rejecting heat exchanger. Accordingly, the fan speed is related to the ambient temperature and it is therefore suitable to use this fan speed as an other parameter in the method according to the present application.

[0046] The temperature of the refrigerant leaving the heat rejecting heat exchanger is also influenced by changes in the ambient temperature. For example, in case the secondary fluid stream across the heat rejecting heat exchanger is an ambient air stream, then the temperature of the refrigerant leaving the heat rejecting heat exchanger will never be lower than the ambient temperature. Accordingly, it is also suitable in the method according to the application to use the temperature of the refrigerant leaving the heat rejecting heat exchanger as the further parameter. Finally, the temperature of the refrigerant leaving the heat rejecting heat exchanger has a very strong correlation with the ambient temperature and is therefore very suitable to use this parameter as the further parameter.

[0047] The step of subsequently operating the vapour compression system can comprise deriving a setpoint value of the temperature and / or pressure of the refrigerant leaving the heat rejecting heat exchanger based on the derived ambient temperature and subsequently operating the vapour compression system in accordance with the derived setpoint value(s).

[0048] As mentioned above, the ambient temperature has an influence on the operation of the vapour compression system in terms of the optimum levels of refrigerant pressure and refrigerant temperature prevailing in the various parts of the vapour compression system. In particular, the optimum values of the temperature and pressure of the refrigerant leaving the heat rejecting heat exchanger are sensitive to changes in the ambient temperature. Therefore, it is advantageous to derive a setpoint value of the temperature and / or pressure of the refrigerant leaving the heat rejecting heat exchanger based on the true ambient temperature. In the method according to the application, the derived ambient temperature represents the true ambient temperature. Therefore, by deriving the setpoint value based on the derived ambient temperature, it is ensured that the setpoint value reflects the true ambient temperature even if reliable measurement results from the ambient temperature sensor cannot be obtained due to solar heating. Accordingly, it is ensured that the vapour compression system is always operated in an optimum manner irrespective of whether the ambient temperature sensor is malfunctioning or not.

[0049] Alternatively or additionally, the step of subsequently operating the vapour compression system can comprise deriving an ambient temperature correction based on the model and the derived model parameters and correcting the ambient temperature measured by the ambient temperature sensor by the ambient temperature correction.

[0050] According to this embodiment, the ambient temperature does not have to be derived directly from the model and at least one other measured parameter. Instead, it is determined whether the measurement obtained by the ambient temperature sensor can be considered reliable or faulty. For example, the measurement obtained by the ambient temperature sensor can be compared to the ambient temperature value derived by the model. As long as there is a consistency between the measured ambient temperature and the derived ambient temperature, the measurement value can be considered reliable. However, if the measured ambient temperature and the derived ambient temperature differ from each other by a certain amount, it can be assumed that the measurement obtained by the ambient temperature sensor is faulty. Then, a correction can be derived based on this comparison and this correction can subsequently be added to the measurement performed by the ambient temperature sensor, thereby obtaining a reliable ambient temperature value. Assuming that the faulty change on the ambient temperature sensor is slower than other changes in the system, it can not be necessary to continuously derive the correction. Instead, the correction can be derived at certain intervals and a fixed correction can be added to the measured ambient temperature. This reduces the processing load.

[0051] The method can further comprise the steps of:

[0052] - comparing the ambient temperature derived by the model and the model parameters derived by the model to the ambient temperature measured by the ambient temperature sensor,

[0053] - in case the difference between the derived ambient temperature and the measured ambient temperature is below a predetermined threshold, operating the vapour compression system based on the measured ambient temperature, and

[0054] - in case the difference between the derived ambient temperature and the measured ambient temperature is above a predetermined threshold, operating the vapour compression system based on the derived ambient temperature.

[0055] According to this embodiment, the derived ambient temperature value is only applied in case there is a difference between the measured ambient temperature and the derived ambient temperature and, thus, the measured ambient temperature can be considered unreliable. However, as long as the measurement performed by the ambient temperature sensor can be considered reliable, the ambient temperature obtained in this way is applied.

[0056] The method can further comprise the steps of:

[0057] - repeating the steps of selecting a time period and obtaining a measurement during the selected time period, and

[0058] - updating the model parameters of the model based on the obtained measurements.

[0059] According to this embodiment, the model is repeatedly improved by collecting more data providing information on the correlation between the ambient temperature and the at least one other parameter in the above-described way. BRIEF DESCRIPTION OF DRAWINGS

[0060] The application will now be described in further detail with reference to the drawings, in which:

[0061] Figure 1 is a diagrammatic view of a vapour compression system controlled in accordance with a method according to an embodiment of the application,

[0062] Figure 2 illustrates heat transfer occurring in a heat rejecting heat exchanger of a vapour compression system controlled in accordance with a method according to an embodiment of the application,

[0063] Figure 3 is a block diagram illustrating a method according to a first embodiment of the application,

[0064] Figure 4 is a block diagram illustrating a method according to a second embodiment of the application, and

[0065] Figure 5 is a graph illustrating the temperature of the refrigerant leaving the heat rejecting heat exchanger, the measured ambient temperature and the derived ambient temperature as a function of time. DETAILED DESCRIPTION

[0066] Figure 1 is a diagrammatic view of a vapour compression system 1 controlled in accordance with a method according to an embodiment of the application. The vapour compression system 1 comprises a compressor 2, a heat rejecting heat exchanger 3, an ejector valve 4, an expansion device 5 in the form of an expansion valve, and an evaporator 6 arranged in a refrigerant path. A fan 7 is arranged to drive a secondary flow of ambient air across the heat rejecting heat exchanger 3.

[0067] The refrigerant flowing in the refrigerant path is compressed by the compressor 2 before being supplied to the heat rejecting heat exchanger 3. In the heat rejecting heat exchanger 3, heat exchange occurs between the refrigerant and a flow of ambient air driven by the fan 7 in such a way that heat is rejected from the refrigerant. The heat rejecting heat exchanger 3 can be in the form of a condenser, in which case the refrigerant is at least partially condensed as it passes through the heat rejecting heat exchanger 3. Alternatively, the heat rejecting heat exchanger 3 can be in the form of a gas cooler, in which case the refrigerant is cooled but remains supercritical or subcooled as it passes through the heat rejecting heat exchanger 3.

[0068] The refrigerant leaving the heat rejecting heat exchanger 3 is supplied to the ejector valve 4, where the refrigerant is subjected to expansion and the pressure is reduced. The refrigerant is then supplied to the expansion device 5, where the refrigerant is subjected to further expansion before being supplied to the evaporator 6. Accordingly, the refrigerant supplied to the evaporator 6 is in a liquid-vapour mixture state.

[0069] In the evaporator 6, a heat exchange between the refrigerant and the air within the refrigeration volume takes place in such a way that heat is absorbed by the refrigerant, while a liquid portion of the refrigerant is at least partially evaporated. Finally, the refrigerant is supplied again to the compressor 2.

[0070] During operation of the vapour compression system 1, the ambient temperature T amb is measured by means of the ambient temperature sensor 8. gc Furthermore, the temperature T gc of the refrigerant leaving the heat rejection heat exchanger 3 is measured by means of the temperature sensor 9, and the pressure P gc of the refrigerant leaving the heat rejection heat exchanger 3 is measured by means of the pressure sensor 10.

[0071] The temperature T gc and the pressure P amb of the refrigerant leaving the heat rejection heat exchanger 3 are setpoint values which are derived from the ambient air temperature T amb . Then, the vapour compression system 1 is operated in accordance with the derived setpoint values.

[0072] However, in case the ambient temperature sensor 8 is exposed to solar heating, the measurements by means of the ambient temperature sensor 8 can be unreliable. Therefore, when this is the case, an estimated or derived value of the ambient air temperature is applied instead. The estimated or derived value of the ambient air temperature is obtained in the following way.

[0073] A time period is selected in which the ambient temperature sensor 8 is not exposed to solar heating. This can for example be a time period during night time, a time period in which it is known that the ambient temperature sensor 8 is in the shade, or a time period in which cloudy conditions have been detected. In any case, since the ambient temperature sensor 8 is not exposed to solar heating during the selected time period, it is ensured that the ambient temperature sensor 8 is in a faultless condition at least with respect to solar heating, and therefore it can be considered that the measurements by the ambient temperature sensor 8 during the selected time period are reliable.

[0074] Therefore, during the selected time period, the vapour compression system 1 is operated while measurements are obtained by means of the ambient temperature sensor 8, the temperature sensor 9 and the pressure sensor 10. Accordingly, simultaneous measurements of the ambient temperature T amb , the temperature T gc of the refrigerant leaving the heat rejection heat exchanger 3 and the pressure P gc of the refrigerant leaving the heat rejection heat exchanger 3 are obtained under the condition that the measurements of the ambient temperature T amb are considered to be reliable.

[0075] Based on the obtained measurements, model parameters of a model for at least the heat rejection heat exchanger 3 are derived. The model provides a relationship between the ambient air temperature T ambthe temperature T gc and pressure P gc of the refrigerant leaving the heat rejecting heat exchanger 3.

[0076] Subsequently, when the ambient temperature sensor 8 can be exposed to solar heating, the ambient temperature can be derived by the model and based on measurements of the temperature T gc and pressure P gc of the refrigerant leaving the heat rejecting heat exchanger 3. Then, when deriving the set point value, the derived ambient temperature is applied. Thus, also when the ambient temperature sensor 8 is exposed to solar heating and thereby cannot provide reliable measurements, proper control of the vapour compression system 1 can be obtained.

[0077] Figure 2 The heat transfer taking place in the heat rejecting heat exchanger is illustrated. The heat rejecting heat exchanger can for example be Figure 1 the heat rejecting heat exchanger 3 of the vapour compression system 1 illustrated.

[0078] The upper part 11 of the figure represents the flow of refrigerant within the heat rejecting heat exchanger, while the lower part 12 of the figure represents the secondary flow of ambient air across the heat rejecting heat exchanger.

[0079] The refrigerant enters the heat rejecting heat exchanger in a substantially gaseous state with a mass flow rate and a specific enthalpy h c . As the refrigerant passes through the heat rejecting heat exchanger, the refrigerant is at least partially condensed. Thereby a gaseous zone, a mixed zone and a liquid zone are formed. The refrigerant leaves the heat rejecting heat exchanger with a mass flow rate and a specific enthalpy h l which is necessarily the same as the mass flow rate of the refrigerant entering the heat rejecting heat exchanger.

[0080] The air of the ambient air stream enters the heat rejecting heat exchanger with a temperature T amb , a specific heat capacity C 空气 and a mass flow rate equal to the ambient temperature. The air of the secondary ambient air stream leaves the heat rejecting heat exchanger with a temperature T 空气，o , a specific heat capacity C 空气 and a mass flow rate .

[0081] As the refrigerant passes through the heat rejecting heat exchanger and the ambient air stream flows across the heat rejecting heat exchanger, heat exchange takes place between the refrigerant and the ambient air stream in such a way that heat is transferred from the refrigerant to the ambient air stream. The heat transfer takes place through a heat transfer rate and From the gaseous zone, the mixing zone and the liquid zone. The heat transfer rates of the individual zones are likely to be different from each other, and thus the total heat transfer from the refrigerant to the ambient air stream depends, among other things, on the distribution of the zones along the heat rejection heat exchanger. However, as heat is transferred from the refrigerant to the ambient air stream, h l <h c , and T 空气，o > T amb .

[0082] From certain assumptions, it can be derived that Figure 2 The energy balance model in the presented system, and the derived energy balance can be applied to build a data-driven model of the heat rejection heat exchanger from which the ambient temperature can be derived.

[0083] Figure 3 is a block diagram illustrating a method according to a first embodiment of the present invention. The gas cooler system 13 represents the operation of a heat rejection heat exchanger of a vapour compression system, e.g. similar to the presented system Figure 2 . During the operation of the vapour compression system, the temperature T gc of the refrigerant leaving the heat rejection heat exchanger and the fan speed N gc driving the secondary ambient air stream across the heat rejection heat exchanger are measured and supplied to an active fault-tolerant control (AFTC) loop 14. In addition, the ambient temperature T amb is also measured and supplied to the AFTC loop 14. The AFTC loop 14 calculates a correction to the ambient temperature T gc based on a model derived essentially in the above described manner and based on T gc and N amb . The AFTC loop 14 outputs the corrected ambient temperature and supplies it to the gas cooler system 13. Then, T is applied instead of the measured ambient temperature T amb during the control of the vapour compression system.

[0084] S AFTC (z) represents the transfer function of the ATFC loop 14, and S gc (z) represents the transfer function of the gas cooler system 13 from to N gc and T gc . S AFTC (z) is designed such that its dynamics are significantly slower than the dynamics of the system being controlled in order to avoid instability.

[0085] Figure 4 is a block diagram illustrating a method according to a second embodiment of the present invention. The gas cooler process 15 controls the temperature T gc and the pressure Pgc The setpoint value is derived by the reference controller 16 in the following way.

[0086] Ambient temperature T amb is measured and supplied to the "day and night" detection module 17, where it is determined whether it is daytime or nighttime. In case of nighttime, it is concluded that the ambient temperature sensor is not exposed to solar heating, and training of the model of the vapour compression system takes place at the "online subspace identification" module 18. The training of the model is based on the measured ambient temperature T amb and the measured value of the temperature of the refrigerant leaving the heat rejecting heat exchanger T gc , and the fan speed N gc driving the fan of the secondary ambient air stream across the heat rejecting heat exchanger.

[0087] The resulting model L is supplied to the "observer" module 19. Subsequently, the ambient temperature sensor can be exposed to solar heating, and thus the measurements by the ambient temperature sensor can be unreliable. The "observer" module 19 then derives the ambient temperature gc based on the derived model L, and the measured values of T gc and N The derived ambient temperature is supplied to the "fault detection" module 20, where the measured ambient temperature T amb is compared to the derived ambient temperature Based on this comparison, a correction to the ambient temperature is calculated

[0088] The true ambient temperature is supplied to the reference controller 16, and the reference controller 16 derives the setpoint value based on this.

[0089] Figure 5 is a graph showing the temperature of the refrigerant leaving the heat rejecting heat exchanger 22, the measured ambient temperature 23 and the derived ambient temperature 24 as a function of time. The derived ambient temperature 24 is based on a model and is derived essentially in the manner described above. All three signals vary in a similar sinusoidal manner, according to the expected variation of the temperature with time of day. Thus, the peak of the signal corresponds to midday, while the trough of the signal corresponds to the point in time immediately before sunrise.

[0090] At the first peak, around t = 1010, it can be seen that the measured ambient temperature 23 and the derived ambient temperature 24 are essentially identical. This indicates that the measurements by the ambient temperature sensor are reliable.

[0091] At the two peaks near t = 1035 and near t = 1060, there is a significant difference between the measured ambient temperature 23 and the derived ambient temperature 24. Furthermore, the measured ambient temperature 23 is higher than the temperature 22 of the refrigerant leaving the heat rejecting heat exchanger, which is physically impossible. Accordingly, it can be concluded that the measurement by the ambient temperature sensor is unreliable. Therefore, during these time intervals, the derived ambient temperature 24 (rather than the measured ambient temperature 23) is applied to control the vapor compression system.

Claims

1. A method for controlling a vapour compression system (1) in the form of a refrigeration system, the vapour compression system (1) comprising at least one compressor (2), a heat rejecting heat exchanger (3), at least one expansion device (5) and at least one evaporator (6) arranged in a refrigerant path, each evaporator (6) being arranged in thermal contact with a refrigeration volume and each expansion device (5) being arranged to supply refrigerant to an evaporator (6), and the heat rejecting heat exchanger (3) being arranged in an outdoor environment, the vapour compression system (1) further comprising an ambient temperature sensor (8) arranged to measure an ambient temperature, the ambient temperature sensor (8) being arranged in the outdoor environment, the method comprising the steps of: - selecting a time period during which the ambient temperature sensor (8) is not exposed to solar heating, - during the selected time period, obtaining measurements of the ambient temperature by the ambient temperature sensor (8) and obtaining measurements of at least one other parameter related to the vapour compression system (1) while operating the vapour compression system (1), - deriving model parameters of a model for at least a part of the vapour compression system (1) based on the obtained measurements, the model providing a correlation between the ambient temperature and the at least one other parameter, and - subsequently operating the vapour compression system (1) based on measurements of the at least one other parameter and based on an ambient temperature derived by the model comprising the derived model parameters.

2. The method of claim 1, wherein, The model is a model reflecting behaviour of at least a part of the vapour compression system (1).

3. The method of claim 1 or 2, wherein, The model is a model of at least the heat rejecting heat exchanger (3).

4. The method according to any of the preceding claims, wherein, The step of selecting a time period comprises selecting a time period during night time.

5. The method of claim 4, wherein, The step of selecting a time period comprises the steps of: - measuring ambient temperature by the ambient temperature sensor (8) during a continuous time interval of at least 24 hours, - identifying time intervals with high ambient temperature as day time and time intervals with low ambient temperature as night time, - calibrating a clock based on the identified day time and night time, and - selecting a time period during night time based on the calibrated clock.

6. The method according to any one of the preceding claims, wherein, The step of deriving model parameters comprises constructing a linear data driven model.

7. The method according to any one of the preceding claims, wherein, The at least one other parameter comprises a fan speed driving a flow of secondary fluid across a fan (7) of the heat rejecting heat exchanger (3) and / or a temperature of refrigerant leaving the heat rejecting heat exchanger (3).

8. The method of any of the preceding claims, wherein, The step of subsequently operating the vapour compression system (1) comprises deriving a setpoint value of a temperature and / or pressure of refrigerant leaving the heat rejecting heat exchanger (3) based on the derived ambient temperature and subsequently operating the vapour compression system (1) in accordance with the derived setpoint value(s).

9. The method according to any of the preceding claims, wherein, The step of subsequently operating the vapour compression system (1) comprises deriving an ambient temperature correction based on the model and the derived model parameters and correcting the ambient temperature measured by the ambient temperature sensor (8) by the ambient temperature correction.

10. The method according to any of the preceding claims, further comprising the step of: - identifying a time period during which the ambient temperature sensor (8) is not exposed to solar heating, - comparing the ambient temperature derived by the model and the model parameters derived with the ambient temperature measured by the ambient temperature sensor (8), - in case the difference between the derived ambient temperature and the measured ambient temperature is below a predetermined threshold, operating the vapour compression system (1) based on the measured ambient temperatures, and - in case the difference between the derived ambient temperatures and the measured ambient temperatures is above the predetermined threshold, operating the vapour compression system (1) based on the derived ambient temperatures.

11. The method according to any of the preceding claims, further comprising the steps of: - repeating the steps of selecting a time period and obtaining measurement results during the selected time period, and - updating the model parameters of the model based on the obtained measurement results.

Images