Methods, systems and computer program products for control optimization of an air handling unit

Abstract

16 ABSTRACT The present disclosure relates to a method (100) for control optimization of an air handling unit, AHU. The method comprises obtaining (S100) at least one operational parameter of the AHU and data relating to at least one AHU performance parameter, predicting (S110) the at least one AHU performance parameter based on the obtained at least one operational parameter using 5 an AHU performance model configured to simulate the performance of the AHU with respect to the at least one AHU performance parameter, comparing (S120) the obtained data relating to the at least one AHU performance parameter to the predicted at least one AHU performance parameter, determining (S130) an optimized at least one operational parameter based on the comparison, and updating (S140) the current at least one operational parameter of the AHU to 10 the optimized at least one operational parameter. (Figure 1 for publication)

Description

[0001] METHODS, SYSTEMS AND COMPUTER PROGRAM PRODUCTS FOR CONTROL

[0002] OPTIMIZATION OF AN AIR HANDLING UNIT

[0003] TECHNICAL FIELD

[0004] The present disclosure relates to monitoring of the control performance of air handling units, AHUs.

[0005] BACKGROUND ART

[0006] Tuning control parameters of an air handling unit, AHU, to optimize the performance of the AHU can be challenging. Many AHU control systems employ Proportional-lntegral-Derivative, PID, controllers and changing PID parameters to achieve better performance can be quite difficult and may sometimes require expert knowledge. There is thus a need in the art for air handling units that facilitate the control of the performance of the AHU.

[0007] SUMMARY OF THE INVENTION

[0008] The present disclosure relates to a method for control optimization of an air handling unit, AHU, the method comprising obtaining a current at least one operational parameter of the AHU and data relating to at least one AHU performance parameter. The method further comprises predicting the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU with respect to the at least one AHU performance parameter. The method also comprises comparing the obtained data relating to the at least one AHU performance parameter to the predicted at least one AHU performance parameter. The method additionally comprises determining an optimized at least one operational parameter based on the comparison. The method further comprises updating the current at least one operational parameter of the AHU to the optimized at least one operational parameter.

[0009] By predicting the at least one AHU performance parameter using the AHU performance model, the method generates a prediction of optimal performance for the AHU given the current at least one operational parameter. The model prediction enables detection of real-world deviations from expected optimal performance, and determining an optimized at least one operational parameter based on a comparison between the model prediction and the actual performance of the AHU. By doing this, the method automatically compensates for deviations from optimal performance, such as a clogged filter or a powered component that is getting old and working sub-optimally. According to some aspects, the at least one AHU performance parameter comprises at least one of an energy consumption of the AHU, an energy consumption of at least one powered component of the AHU, and at least one air parameter. The at least one AHU performance parameter constitutes the target or objective parameter or function to be optimized by the proposed method for control optimization.

[0010] According to some aspects, the at least one operational parameter comprises at least one of a unit of measure comprising one of absolute humidity, relative humidity, and dew point, an operational mode configured to optimize one of energy cost, energy consumption, air parameter stability, and air parameter tolerance levels, control parameters for controlling at least one powered component of the AHU, and at least one set point for the at least one AHU performance parameter.

[0011] According to some aspects, comparing the obtained data to the predicted at least one AHU performance parameter comprises detecting a deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

[0012] By detecting deviations from expected variations, the method captures statistically expected behavior while simultaneously accounting for uncertainty in both simulation of optimal behavior and real-world measurements.

[0013] According to some aspects, the deviation comprises a deviation in a difference between an observed steady-state in the obtained data relating to the at least one AHU performance parameter and a predicted steady-state in the predicted at least one AHU performance parameter.

[0014] The method can thereby detect deterioration in performance over time, e.g. due to filters gradually getting clogged.

[0015] According to some aspects, the deviation comprises a deviation in a difference between respective temporal profiles of the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

[0016] The method can thereby detect time-dependent disturbances in AHU performance.

[0017] According to some aspects, determining an optimized at least one operational parameter is further based on the detected deviation.

[0018] Determining the optimized at least one operational parameter based on the detected deviation improves accuracy of the optimized at least one operational parameter and allows the determination to take into account both steady-state deviations and time-dependent disturbances of the AHU.

[0019] According to some aspects, determining the optimized at least one operational parameter is performed using a machine learning, ML, algorithm.

[0020] Using a machine learning algorithm enables capturing complex relationships between the predictions of the AHU performance model and the at least one AHU performance parameter, such as detecting a deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

[0021] According to some aspects, determining the optimized at least one operational parameter is further based on historical operational data from other AHUs of the same type as the present AHU.

[0022] By integrating historical data from other AHUs, e.g. via a database collecting operational parameters, performance parameters and performance-related data from other AHUs, an improved at least one operational parameter can be determined by considering more data and more scenarios relating to suboptimal performance of AHUs .

[0023] According to some aspects, predicting the at least one AHU performance parameter further comprises obtaining machine-specific configuration data comprising information relating to physical dimensions and powered component layout of the AHU, providing the machine-specific configuration data to the AHU performance model, and using the machine-specific configuration data in the prediction of the at least one AHU performance parameter.

[0024] By taking the physical dimensions and powered component layout of the AHU into account, an accurate model of energy consumption, heat transport, airflow and their effects on humidity and dew point can be simulated to a high degree of accuracy, which in turn enables better detection of deviations from ideal operating conditions and may be used to pinpoint the cause of any detected deviation.

[0025] According to some aspects, updating the current at least one operational parameter of the AHU is preceded by a step of presenting a proposed action involving an update of the current operational parameters to an operator of the AHU, via a human-machine interface, HMI, of the AHU, and wherein updating the current at least one operational parameter of the AHU to the optimized at least one operational parameter is performed in response to an accept of the proposed action by the operator. The method can thereby get user feedback that evaluates the implicit tradeoffs involved with updating the current operational parameters. In addition to providing improved user control over the method, the user feedback can be used to improve the performance of the method, such as updating the parameters of a machine learning algorithm.

[0026] According to some aspects, the method further comprises registering each of the at least one operational parameter with a version control system configured to enable resetting the current at least one control parameter to any of a number of previously registered versions of control parameters.

[0027] A version control enables a user to quickly switch between different sets of operational parameters without needing to find a preferred set for a given situation. A further technical effect and advantage is that the use of version control reduces the potential for problems associated with transitioning to using the updated at least one operational parameter; if performance is unsatisfactory, the previous at least one operational parameter can be restored immediately.

[0028] The present disclosure further relates to a computer program product comprising a non- transitory computer-readable storage medium having stored / encoded thereon a computer program comprising program instructions, the computer program being loadable into a processor and configured to cause the processor to perform the method for control optimization of an air handling unit as described above and below.

[0029] The computer program product implements the disclosed method for control performance monitoring of an air handling unit and has all the associated technical effects and advantages.

[0030] The present disclosure further relates to a system for control performance monitoring of an air handling unit, AHU. The system comprises an AHU. The AHU comprises a control performance monitoring module configured to obtain at least one operational parameter of the AHU and data relating to at least one AHU performance parameter. The system further comprises computational circuitry configured to predict the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU, defined by the at least one AHU performance parameter, in a scenario defined by the at least one operational parameter. The system further comprises diagnostic circuitry configured to compare the obtained data relating to the at least one AHU performance parameter with the predicted at least one AHU performance parameter and determine an optimized at least one operational parameter based on the comparison. The control performance monitoring module is communicatively connected to the computational circuitry and the diagnostic circuitry. The control performance monitoring module is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter. The system implements the disclosed method for control performance monitoring of an air handling unit and has all the associated technical effects and advantages.

[0031] According to some aspects, the system further comprises a human-machine interface, HMI. The HMI is configured to present a proposed action involving an update of the current operational parameters to the optimized at least one operational parameter to an operator of the AHU. The control performance monitoring module is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter in response to an accept of the proposed action by the operator.

[0032] The system can thereby get user feedback via the HMI that evaluates the implicit tradeoffs involved with updating the current operational parameters. In addition to providing improved user control over the updating of operational parameters, the user feedback can be used to improve the performance of the system, such as updating the parameters of a machine learning algorithm.

[0033] According to some aspects, the system further comprises a machine database configured to provide machine-specific configuration data comprising information relating to physical dimensions and a powered component layout of the AHU to the control performance monitoring module. The control performance monitoring module is further configured to provide the machine-specific configuration data to the AHU performance model, whereby the computational circuitry may use the machine-specific data in the prediction of the at least one AHU performance parameter.

[0034] The information relating to physical dimensions and powered component layout of the AHU greatly improves the accuracy of the AHU performance model and subsequently the comparison between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

[0035] BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 illustrates aspects of the disclosed method for control optimization of an air handling unit; and

[0037] Figure 2 illustrates aspects of the disclosed system for control optimization of an air handling unit.

[0038] DETAILED DESCRIPTION Figure 1 illustrates aspects of the disclosed method 100 for control optimization of an air handling unit, AHU. The method comprises obtaining S100 a current at least one operational parameter of the AHU and data relating to at least one AHU performance parameter. In other words, the method comprises obtaining information relating to the current settings on what the AHU is supposed to do, and which trade-offs it is supposed to make, as well as collect performance- related information that relates to how well the AHU is currently performing. Thus, according to some aspects, the at least one AHU performance parameter comprises at least one of the energy consumption of the AHU, the energy consumption of at least one powered component of the AHU, and at least one air parameter. According to some aspects, the at least one operational parameter comprises at least one of a unit of measure comprising one of absolute humidity, relative humidity, and dew point, an operational mode configured to optimize one of energy cost, energy consumption, air parameter stability, and air parameter tolerance levels, control parameters for controlling at least one powered component of the AHU, and at least one set point for the at least one AHU performance parameter.

[0039] The method further comprises predicting S110 the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU with respect to the at least one AHU performance parameter. The AHU performance model provides a theoretical approximation of an expected performance given a current at least one operational parameter, which can be used to compare with actual performance in order to determine both potential problems and improved operational parameters. The modelling of AHU behavior can be improved if the physical structure and / or layout of powered components is provided. This greatly facilitates modelling of air flow, energy expenditure and efficiency, and temperature distribution in both space and time. Thus, according to some aspects, predicting S110 the at least one AHU performance parameter further comprises obtaining S112 machine-specific configuration data comprising information relating to physical dimensions and powered component layout of the AHU, and providing S114 the machine-specific configuration data to the AHU performance model.

[0040] The method also comprises comparing S120 the obtained data relating to the at least one AHU performance parameter to the predicted at least one AHU performance parameter. The comparison detects deviations from expected performance and can suggest updated parameters for improved performance, and optionally provide a remedy for a current difference in current performance vs expected performance. For instance, if a filter is getting clogged, a potential at least one updated parameter may be directed at increasing air flow, while a potential remedy would be to clear the filter or replace it.

[0041] In practice, measurements include uncertainties that are preferably taken into account when comparing the obtained data to the predicted at least one AHU performance parameter. One way to do this is to determine what typical uncertainties look like, such as how large deviations one might expect during steady-state and / or what a typical behavior is across time. Thus, according to some aspects, comparing S120 the obtained data to the predicted at least one AHU performance parameter comprises detecting S122 a deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter. According to some aspects, the deviation comprises a deviation in a difference between an observed steady-state in the obtained data relating to the at least one AHU performance parameter and a predicted steadystate in the predicted at least one AHU performance parameter. According to some aspects, the deviation comprises a deviation in a difference between respective temporal profiles of the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter. The deviation in a difference between respective temporal profiles enables determining the presence of a time-dependent disturbance.

[0042] The method further comprises determining S130 an optimized at least one operational parameter based on the comparison. If detecting S122 a deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter, the detected deviation can significantly improve the determination S130 of the optimized at least one operational parameter, since the deviation directly captures the steady-state or dynamic nature of the deviation. Thus, according to some aspects, determining S130 an optimized at least one operational parameter is further based on the detected deviation. The quality of the determination S130 of the optimized at least one operational parameter can be further improved by using knowledge of how this and / or other AHUs of the same type has worked under similar circumstances and what the associated most likely problems and remedies are. Thus, according to some aspects, determining S130 the optimized at least one operational parameter is further based on historical operational data from other AHUs of the same type as the present AHU. The historical operational data may be obtained from a database comprising the historical operational data.

[0043] According to some aspects, determining S130 the optimized at least one operational parameter is performed using a machine learning, ML, algorithm. A machine learning algorithm enables capturing complex interdependencies of operational parameters and the at least one AHU performance parameter, and enables integration of machine-specific configuration data and historical operational data. The ML algorithm further provides a mechanism for detecting S122 the deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter. The method further comprises updating S140 the current at least one operational parameter of the AHU to the optimized at least one operational parameter.

[0044] According to some aspects, updating S140 the current at least one operational parameter of the AHU is preceded by a step of presenting S142 a proposed action involving an update of the current operational parameters to an operator of the AHU, via a human-machine interface, HMI, of the AHU, and updating S140 the current at least one operational parameter of the AHU to the optimized at least one operational parameter is performed in response to an accept of the proposed action by the operator. The disclosed method thus relieves the operator of the complex task of diagnosing the current performance of the AHU and determine if and how the current at least one operational parameter should be updated, while still providing the operator full control of the process. Furthermore, updating S140 the current at least one operational parameter in response to the accept allows the AHU to collect operator preferences for different scenarios, which could be stored in a database and / or be integrated into a machine learning algorithm of the AHU and / or the HMI.

[0045] According to some aspects, the method further comprises registering S150 each of the at least one operational parameter with a version control system configured to enable resetting the current at least one control parameter to any of a number of previously registered versions of control parameters. Version control allows the operator to revert to previous control parameters and thereby reduces any perceived risks or performance related issues, since the operator does not need to find a better at least one control parameter than the current at least one control parameter - the operator can simply revert to a previous at least one control parameter.

[0046] The present disclosure further relates to a computer program product comprising a non- transitory computer-readable storage medium having thereon a computer program comprising program instructions, the computer program being loadable into a processor and configured to cause the processor to perform the method for control optimization of an air handling unit as described above and below.

[0047] Figure 2 illustrates aspects of the disclosed system 200 for control optimization of an air handling unit, AHU 210.

[0048] The system comprises an AHU 210. The AHU 210 comprises a control performance monitoring module 220 configured to obtain at least one operational parameter of the AHU and data relating to at least one AHU performance parameter.

[0049] The system further comprises computational circuitry 230 configured to predict the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU, defined by the at least one AHU performance parameter, in a scenario defined by the at least one operational parameter. Depending on the computational demands of the AHU performance model, the computational circuitry 230 may be located at the AHU or at a server to which the AHU is communicatively connected. Thus, according to some aspects, the AHU 210 comprises the computational circuitry 230. According to some aspects, the system comprises a server, wherein the server comprises the computational circuitry 230. According to some aspects, the computational circuitry 230 comprises a processor 232 and a memory 234. Placing the computational circuitry 230 at a server reduces the energy consumption of the AHU and allows the server to potentially function as a hub, serving multiple AHUs at the same time.

[0050] The system also comprises diagnostic circuitry 236a, 236b configured to compare the obtained data relating to the at least one AHU performance parameter with the predicted at least one AHU performance parameter, and determine an optimized at least one operational parameter based on the comparison. Examples of how the comparison can be made and how the optimized at least one operational parameter are determined can be found in relation to Figure 1 , which describes details of the method implemented by the disclosed system 200.

[0051] The diagnostic circuitry 236a, 236b may be comprised within the AHU 210 or be comprised within an external server of the system 200. If the system comprises an external server, the external server may thus comprise both the computational circuitry 230 and the diagnostic circuitry 236b. In other words, according to some aspects, the system 200 comprises a server, wherein the server comprises the diagnostic circuitry 236b. According to some aspects, the diagnostic circuitry 236a, 236b comprises a processor 238a, 238b and a memory 239a, 239b.

[0052] The control performance monitoring module 220 is communicatively connected to the computational circuitry 230 and the diagnostic circuitry 236a, 236b.

[0053] According to some aspects, the system 200 further comprises a machine database 250 configured to provide machine-specific configuration data comprising information relating to physical dimensions and powered component layout of the AHU to the control performance monitoring module 220. The control performance monitoring module 220 is further configured to provide the machine-specific configuration data to the AHU performance model. The computational circuitry 230 is configured to use the machine-specific data in the prediction of the at least one AHU performance parameter. Alternatively, the machine database 250 is communicatively connected to the computational circuitry 230 and configured to provide the machine-specific configuration data to the AHU performance model.

[0054] According to some aspects, the machine database 250 is configured to provide the machinespecific configuration data to the AHU performance model based on a request originating from the performance monitoring module 220. According to some aspects, the AHU 210 comprises the machine database 250. According to some alternative aspects, the system further comprises a server, wherein the server comprises the machine database 250.

[0055] The control performance monitoring module 220 is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter.

[0056] According to some aspects, the system 200 further comprises a human-machine interface, HMI, 240. The HMI 240 is configured to present a proposed action involving an update of the current at least one operational parameter to the optimized at least one operational parameter to an operator of the AHU. The control performance monitoring module 220 is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter in response to an accept of the proposed action by the operator.

[0057] By compartmentalizing the task of presenting the proposed action and receive the accept of the proposed action to the HMI, the additional complexity associated with generating the message presenting the proposed action, such as the natural language message processing illustrated below, is removed from the control performance monitoring module 220, which keeps the energy consumption of the control performance monitoring module 220 to a minimum while simultaneously introducing a flexibility in the system 220 in that the HMI can be modified or replaced without having to change the control performance monitoring module 220.

[0058] According to an exemplary embodiment, the HMI comprises a generative Al module configured to generate, transmit, receive and interpret natural language messages relating to the operational state of the AHU using a first machine learning algorithm trained to map natural language to machine-specific knowledge. Thus, rather than informing an operator of the AHU about the specific current at least one operational parameter of the AHU that is updated S140 to the optimized at least one operational parameter, the HMI is configured to generate a natural language message arranged to explain the change and the expected technical effect in natural language, thereby removing the need for the operator to possess expert knowledge. In other words, the HMI is arranged to receive a natural language, wherein the natural language request comprises an intention expressed in natural language for a desire to adjust and / or diagnose an operational state of the AHU. The HMI is further configured to interpret the natural language and translate the message to instructions for the AHU carry out the natural language request. The HMI is also configured to provide a natural language response comprising a natural language description of the operational state of the AHU after carrying out the natural language request.

[0059] According to some aspects, the HMI comprises a graphical user interface, GUI, configured to receive user input arranged to cause the AHU to initiate the disclosed method 100 for control optimization of an air handling unit, providing information relating to a current operational state of the AHU, and accept a response for a proposed update relating to the current at least one operational parameter of the AHU.

[0060] To summarize, the system is configured to carry out the disclosed method for control optimization of an air handling unit. Although described above with reference to a number of discrete circuitries or modules configured to carry out different steps of the method, it should be appreciated that the method is a computer-implemented method and that the logic for carrying out the method could be centralized or distributed among different “modules” or “circuitries” of the system in any conceivable manner. Essentially, the system comprises one or more processors configured to carry out the steps of the method upon execution of different code segments of the above-mentioned computer program.

[0061] The AHU 210 of the present disclosure may be any type of device involved in the process of treating air to be provided into a defined space, including but not limited to devices involved in air treatment processes such as heating, cooling, humidifying, dehumidifying, filtering and ventilating. In an exemplary embodiment, the AHU 210 is a dehumidifier. For example, the AHU 210 may be a dehumidifier including a desiccant wheel or rotor. Such a rotor comprises a sorption media, also called rotor media, which typically consists of corrugated panels forming axially extending channels through the rotor media. The rotor rotates slowly between a process airstream and a regeneration airstream and the gas sorption rotor typically comprises a process section and a regeneration section. The process air flows through the channels of the rotor media and the rotor media either adsorbs or absorbs moisture. When the gas sorption rotor rotates, the rotor media is heated by the hot regeneration air, and the rotor media releases its moisture into the regeneration air. Following regeneration, the gas sorption rotor rotates back into the process airstream, where the process repeats itself.

Claims

CLAIMS1. A method (100) for control optimization of an air handling unit, AHU, the method comprising obtaining (S100) a current at least one operational parameter of the AHU and data relating to at least one AHU performance parameter, predicting (S110) the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU with respect to the at least one AHU performance parameter, comparing (S120) the obtained data relating to the at least one AHU performance parameter to the predicted at least one AHU performance parameter, determining (S130) an optimized at least one operational parameter based on the comparison, and updating (S140) the current at least one operational parameter of the AHU to the optimized at least one operational parameter.

2. The method according to claim 1 , wherein the at least one AHU performance parameter comprises at least one of an energy consumption of the AHU, an energy consumption of at least one powered component of the AHU, and at least one air parameter.

3. The method according to claim 1 or 2, wherein the at least one operational parameter comprises at least one of• a unit of measure comprising one of absolute humidity, relative humidity, and dew point,• an operational mode configured to optimize one of energy cost, energy consumption, air parameter stability, and air parameter tolerance levels,• control parameters for controlling at least one powered component of the AHU, and• at least one set point for the at least one AHU performance parameter.

4. The method according to any of the preceding claims, wherein comparing (S120) the obtained data to the predicted at least one AHU performance parameter comprises detecting (S122) a deviation from an expected variation in a difference between the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

5. The method according to claim 4, wherein the deviation comprises a deviation in a difference between an observed steady-state in the obtained data relating to the at least one AHU performance parameter and a predicted steady-state in the predicted at least one AHU performance parameter.

6. The method according to claim 4 or 5, wherein the deviation comprises a deviation in a difference between respective temporal profiles of the obtained data relating to the at least one AHU performance parameter and the predicted at least one AHU performance parameter.

7. The method according to any of claims 4-6, wherein determining (S130) an optimized at least one operational parameter is further based on the detected deviation.

8. The method according to any of the preceding claims, wherein determining (S130) the optimized at least one operational parameter is performed using a machine learning, ML, algorithm.

9. The method according to any of the preceding claims, wherein determining (S130) the optimized at least one operational parameter is further based on historical operational data from other AHUs of the same type as the present AHU.

10. The method according to any of the preceding claims, wherein predicting (S110) the at least one AHU performance parameter further comprises obtaining (S112) machine-specific configuration data comprising information relating to physical dimensions and a powered component layout of the AHU, and providing (S114) the machine-specific configuration data to the AHU performance model, and using (S116) the machine-specific configuration data in the prediction of the at least one AHU performance parameter.

11. The method according to any of the preceding claims, wherein updating (S140) the current at least one operational parameters of the AHU is preceded by a step of presenting (S142) a proposed action involving an update of the current at least one operational parameter to an operator of the AHU, via a human-machine interface, HMI, of the AHU,wherein updating (S140) the current at least one operational parameter of the AHU to the optimized at least one operational parameter is performed in response to an accept of the proposed action by the operator.

12. The method according to any of the preceding claims, further comprising registering (S150) each set of operational parameters with a version control system configured to enable resetting the current at least one control parameter to any of a number of previously registered versions of control parameters.

13. A computer program product comprising a non-transitory computer-readable storage medium having stored thereon a computer program comprising program instructions, the computer program being loadable into a processor and configured to cause the processor to perform the method for control optimization of an air handling unit according to any of claims 1-12.

14. A system (200) for control performance monitoring of an air handling unit, AHU, the system comprising an AHU (210), the AHU comprising• a control performance monitoring module (220) configured to obtain at least one operational parameter of the AHU and data relating to at least one AHU performance parameter, computational circuitry (230) configured to predict the at least one AHU performance parameter based on the obtained at least one operational parameter using an AHU performance model configured to simulate the performance of the AHU, defined by the at least one AHU performance parameter, in a scenario defined by the at least one operational parameter, diagnostic circuitry (236a, 236b) configured to compare the obtained data relating to the at least one AHU performance parameter with the predicted at least one AHU performance parameter, and determine an optimized at least one operational parameter based on the comparison, wherein the control performance monitoring module (220) is communicatively connected to the computational circuitry (230) and the diagnostic circuitry (236a, 236b), wherein the control performance monitoring module (220) is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter.

15. The system according to claim 14, further comprising a human-machine interface, HMI, (240), wherein the HMI (240) is configured to• present a proposed action involving an update of the current operational parameters to the optimized at least one operational parameter to an operator of the AHU, wherein the control performance monitoring module (220) is further configured to update the current at least one operational parameter of the AHU to the optimized at least one operational parameter in response to an accept of the proposed action by the operator.

16. The system according to claim 14 or 15, further comprising a machine database (250) configured to• provide machine-specific configuration data comprising information relating to physical dimensions and a powered component layout of the AHU to the control performance monitoring module (220), wherein the control performance monitoring module (220) is further configured to provide the machine-specific configuration data to the AHU performance model, the computational circuitry (230) being configured to use the machine-specific data in the prediction of the at least one AHU performance parameter.

Images