Method, device, equipment and medium for controlling heating, ventilation and air conditioning system

Abstract

The method comprises the following steps: determining an initial number of people, an initial indoor temperature and an initial indoor carbon dioxide concentration of a building served by a heating ventilation air conditioning (HVAC) system; determining a predicted number of people and a room occupancy state at a target time point according to the initial number of people; determining a predicted indoor temperature and a predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people; inputting the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration into a pre-trained model optimizer to obtain an indoor air temperature set value and a fresh air flow of the HVAC system at the target time point, and an optimization target of the model optimizer comprises a thermal comfort penalty, an indoor air quality penalty and an energy consumption penalty. Thus, an efficient balance among energy efficiency optimization, residential comfort and air quality can be achieved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of intelligent building environment management, and in particular to a heating, ventilation and air conditioning system control method, device, equipment and medium. BACKGROUND

[0002] Traditional building automation systems (BAS) ignore the dynamic variable of human behavior, relying on fixed schedules or oversimplified human behavior patterns. This approach can easily lead to excessive cooling, excessive heating or excessive ventilation, resulting in a large amount of energy waste. Human behavior is influenced by multiple factors such as culture, psychology and situation, and its high uncertainty makes it difficult for traditional systems to adapt. Some control systems have added people sensors and used simple feedback control logic (OBC) based on the number of people, which can adjust system set values according to real-time people data. However, this control logic lacks comprehensive consideration of personnel comfort feedback, real-time population, population prediction and dynamic prediction of building environment. It cannot solve the problem of conflict between multiple control objectives, such as balancing energy efficiency optimization, comfort and air quality. Although the development of Internet of Things (IoT) technology in recent years has accumulated a large amount of real-time data in the building environment, these data have not been systematically analyzed and modeled, and it is difficult to fully play a role in real-time prediction and control decisions. This lack of human-centered adaptability not only weakens the potential for building energy saving, but also may reduce personnel satisfaction. Therefore, the prior art still needs to be improved and developed. SUMMARY

[0003] Embodiments of the present disclosure provide a heating, ventilation and air conditioning system control method, device, electronic equipment and computer readable storage medium, aiming to at least solve one of the technical problems in the related art to some extent.

[0004] In a first aspect, embodiments of the present disclosure provide a heating, ventilation and air conditioning system control method, the method comprising:

[0005] determining an initial number of people, an initial indoor temperature and an initial indoor carbon dioxide concentration of a building served by a heating, ventilation and air conditioning system;

[0006] determining a predicted number of people and a room occupancy state at a target time point according to the initial number of people;

[0007] determining a predicted indoor temperature and a predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people;

[0008] input the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration into a pre-trained model optimizer to obtain an indoor air temperature set value and a fresh air flow rate of the HVAC system at the target time point, wherein an optimization objective of the model optimizer comprises a thermal comfort penalty, an indoor air quality penalty and an energy consumption penalty.

[0009] In a second aspect, the embodiments of the present disclosure further provide a HVAC system control device, which comprises:

[0010] a first determination for determining an initial number of people, an initial indoor temperature and an initial indoor carbon dioxide concentration of a building served by the HVAC system;

[0011] a second determination for determining a predicted number of people and a room occupancy state at a target time point according to the initial number of people;

[0012] a third determination for determining a predicted indoor temperature and a predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people;

[0013] a processing module for inputting the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration into a pre-trained model optimizer to obtain an indoor air temperature set value and a fresh air flow rate of the HVAC system at the target time point, wherein an optimization objective of the model optimizer comprises a thermal comfort penalty, an indoor air quality penalty and an energy consumption penalty.

[0014] In a third aspect, the embodiments of the present disclosure further provide an electronic device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the computer program is executed by the processor to implement the steps in the HVAC system control method described above.

[0015] In a fourth aspect, the embodiments of the present disclosure further provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps in the HVAC system control method described above.

[0016] In a fifth aspect, the embodiments of the present disclosure further provide a computer program product or a computer program, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to make the computer device execute the method provided in various optional implementation manners of the embodiments of the present disclosure.

[0017] In the embodiments of the present disclosure, the initial number of people, the initial indoor temperature and the initial indoor carbon dioxide concentration of a building served by the heating ventilation air conditioning system are determined; according to the initial number of people, the predicted number of people and the room occupancy state at a target time point are determined; according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people, the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point are determined; the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration are input into a pre-trained model optimizer to obtain the indoor air temperature set value and the fresh air flow of the heating ventilation air conditioning system at the target time point, wherein the optimization objectives of the model optimizer include: thermal comfort penalty, indoor air quality penalty and energy consumption penalty. Therefore, efficient balance between energy efficiency optimization, residential comfort and air quality can be achieved.

[0018] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present disclosure, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 is a flowchart of a heating ventilation air conditioning system control method provided by the embodiments of the present disclosure;

[0021] Figure 2 is a flowchart of another heating ventilation air conditioning system control method provided by the embodiments of the present disclosure;

[0022] Figure 3 is an implementation architecture schematic diagram of a human-in-the-loop model optimization control method based on Internet of Things according to the present disclosure;

[0023] Figure 4 is a human-in-the-loop model optimization control scheme schematic diagram integrating random number of people model, real-time number of people and real-time personnel feedback according to the present disclosure;

[0024] Figure 5 is a thermal resistance and thermal capacity diagram showing the thermal dynamic structure of a building according to the present disclosure;

[0025] Figure 6 is a running schematic diagram of a real-time personnel comfort feedback system based on two-dimensional code according to the present disclosure;

[0026] Figure 7 is a structural schematic diagram of a heating ventilation air conditioning system control device provided by the embodiments of the present disclosure;

[0027] Figure 8 is a structural schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0028] Some embodiments of the present disclosure will be described in detail with reference to the drawings, of which examples are shown. The following description, in relation to the drawings, refers to the same or similar elements using the same or similar reference numerals unless otherwise indicated. Various changes, modifications and equivalents of the methods, devices and / or systems described herein will become apparent to those skilled in the art after understanding the present disclosure. For example, the order of the operations described herein is merely an example, and is not limited to those set forth herein, but can be changed as apparent after understanding the present disclosure, except for the operations that must be performed in a specific order. In addition, the description of features known in the art can be omitted for the sake of clarity and brevity.

[0029] The implementations described in some embodiments of the present disclosure below do not represent all implementations consistent with the present disclosure. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0030] It should be noted that the execution subject of the heating, ventilation and air conditioning system control method of the present embodiment can be a heating, ventilation and air conditioning system control device, which can be configured in any type of electronic device, which is not limited herein.

[0031] In the present embodiment of the present disclosure, the heating, ventilation and air conditioning system control method will be executed by the heating, ventilation and air conditioning system control device as the execution subject, which is not limited herein.

[0032] It should be noted that the order of the following embodiments is not limited as the priority order of the embodiments.

[0033] Figure 1 is a flowchart of a heating, ventilation and air conditioning system control method according to the first embodiment of the present disclosure.

[0034] As shown in Figure 1 , the method comprises:

[0035] Step 101, determining the initial number of people, the initial indoor temperature and the initial indoor carbon dioxide concentration of the building served by the heating, ventilation and air conditioning system.

[0036] The heating, ventilation and air conditioning system (HVAC) can be composed of a fan coil unit (FCU) and a main air handling unit (PAU), and the fresh air flow is adjusted by a variable air volume (VAV) system.

[0037] The initial number of people, the initial indoor temperature, and the initial indoor carbon dioxide concentration refer to the number of people, the indoor temperature, and the indoor carbon dioxide concentration of the building served by the heating ventilation air conditioning system in an initial state before multi-objective model predictive control is performed.

[0038] In the embodiments of the present disclosure, an indoor temperature sensor, a carbon dioxide sensor, a number of people sensor, and an Internet of Things temperature controller can be configured in the building served by the heating ventilation air conditioning system. The indoor temperature sensor and the carbon dioxide sensor are respectively used to monitor the indoor temperature and the carbon dioxide concentration, and are both installed at a central position of the room to ensure the representativeness of the measurement data. The number of people sensor is a camera-based number of people sensor used to detect the number of people in the room in real time, and is installed on the ceiling of the room to ensure that the entire room can be covered. The Internet of Things temperature controller is used to receive the optimized temperature set value and control the operation of the FCU, and is installed on the wall of the room for convenient personnel operation.

[0039] In the embodiments of the present disclosure, when the initial number of people, the initial indoor temperature, and the initial indoor carbon dioxide concentration of the building served by the heating ventilation air conditioning system are determined, reliable initial data support can be provided for subsequent control of the heating ventilation air conditioning system.

[0040] In step 102, the predicted number of people and the room occupancy state at a target time point are determined according to the initial number of people.

[0041] The target time point can refer to a time point at which the heating ventilation air conditioning system is to be controlled. In the embodiments of the present disclosure, one or more target time points can be determined within an optimization time range based on a preset time step.

[0042] The predicted number of people refers to the predicted number of people in the building served by the heating ventilation air conditioning system at the target time point. The room occupancy state can be used to indicate whether the building served by the heating ventilation air conditioning system is occupied and used by a user at the target time point. For example, when there is a person in the building served by the heating ventilation air conditioning system at the target time point, the room occupancy state can be represented by 1, and when there is no person in the building served by the heating ventilation air conditioning system at the target time point, the room occupancy state can be represented by 0.

[0043] For example, in the embodiments of the present disclosure, when the predicted number of people and the room occupancy state at the target time point are determined according to the initial number of people, the initial number of people and the current time characteristics (such as the day of the week, the beginning of the month, the end of the month, etc.) can be input into a pre-trained machine learning model (which can be trained based on historical number of people change data of the building served by the heating ventilation air conditioning system) to determine the predicted number of people and the room occupancy state at the target time point. Alternatively, the predicted number of people and the room occupancy state at the target time point can also be determined according to the initial number of people based on any other possible method, which is not limited in this regard.

[0044] Optionally, in some embodiments, when determining the predicted number of people and room occupancy status at the target time point based on the initial number of people, the following steps can be taken: The transition probability parameters are dynamically estimated using an adaptive B-spline method, where multiple transition probability parameters jointly form a transition probability matrix; an initial state probability distribution is determined based on the initial number of people; the predicted number of people at the target time point is determined based on the transition probability matrix and the initial state probability distribution; and the room occupancy status at the target time point is determined based on the predicted number of people at the target time point and a preset threshold. Therefore, a non-homogeneous Markov chain (IMC) method can be used for random number modeling, thereby achieving accurate prediction of the number of people and room occupancy status at the target time point.

[0045] Among them, the transition probability parameter ( Let represent the probability of transitioning from state i to state j at time t. Here, the adaptive B-spline method is used to dynamically estimate these time-varying parameters. The transition probability matrix, defined in the non-homogeneous Markov chain (IMC) method, represents the likelihood of state transitions over time, as well as the initial state (or initial distribution) in the state space.

[0046] The preset threshold refers to a pre-configured threshold used to determine the occupancy status of a room. For example, it can be set to 0.5.

[0047] In this embodiment of the disclosure, when the predicted number of people and room occupancy status at the target time point are determined based on the initial number of people, reliable data support can be provided for subsequently determining the predicted indoor temperature and predicted indoor carbon dioxide concentration at the target time point.

[0048] Step 103: Determine the predicted indoor temperature and predicted indoor carbon dioxide concentration at the target time point based on the initial indoor temperature, the initial indoor carbon dioxide concentration, and the predicted number of people.

[0049] Among them, predicted indoor temperature and predicted indoor carbon dioxide concentration refer to the predicted indoor temperature and indoor carbon dioxide concentration of the building served by the HVAC system at the target time point.

[0050] It is understood that indoor temperature and carbon dioxide concentration may be affected by the number of people. Therefore, the embodiments of this disclosure can combine the initial indoor temperature, the initial indoor carbon dioxide concentration, and the predicted number of people to determine the predicted indoor temperature and predicted indoor carbon dioxide concentration at a target time point. This provides reliable reference information for subsequent HVAC system control.

[0051] In step 104, the room occupancy state, the predicted indoor temperature, and the predicted indoor carbon dioxide concentration are input into a pre-trained model optimizer to obtain an indoor air temperature set value and a fresh air flow rate of the HVAC system at a target time point, wherein the optimization objectives of the model optimizer include a thermal comfort penalty, an indoor air quality penalty, and an energy consumption penalty.

[0052] The model optimizer can be a pre-configured model for optimizing the control of the HVAC system based on the thermal comfort penalty, the indoor air quality penalty, and the energy consumption penalty.

[0053] Optionally, in some embodiments, the model optimizer performs multi-objective optimization control based on the following manner: constructing a first penalty function corresponding to the thermal comfort penalty, a second penalty function corresponding to the indoor air quality penalty, and a third penalty function corresponding to the energy consumption penalty; determining a first importance description value corresponding to the thermal comfort penalty, a second importance description value corresponding to the indoor air quality penalty, and a third importance description value corresponding to the energy consumption penalty; determining a first product value of the first penalty function and the first importance description value, a second product value of the second penalty function and the second importance description value, and a third product value of the third penalty function and the third importance description value; determining a first sum value of the first product value and the second product value, and a fourth product value of the first sum value and the room occupancy state; determining a second sum value of the fourth product value and the third product value; and minimizing the cumulative sum of the second sum values corresponding to multiple target time points within a prediction time range to obtain the indoor air temperature set value and the fresh air flow rate corresponding to each target time point. In this way, the weighting of multiple optimization objectives can be realized based on the room occupancy state and the importance description value corresponding to each optimization objective, thereby effectively improving the multi-objective optimization effect of the model optimizer.

[0054] The first importance description value is used to describe the relative importance of the thermal comfort penalty among the multiple optimization objectives. The second importance description value is used to describe the relative importance of the indoor air quality penalty among the multiple optimization objectives. The third importance description value is used to describe the relative importance of the energy consumption penalty among the multiple optimization objectives.

[0055] Optionally, in some embodiments, the first penalty function is used to evaluate the deviation between the indoor air temperature set value and a desired reference temperature; the second penalty function is used to indicate the case that the indoor carbon dioxide concentration exceeds a preset control threshold, wherein the second penalty function takes a value of zero when the indoor carbon dioxide concentration is less than the preset control threshold; and the third penalty function is used to indicate the energy consumption required for providing cooling for the room and for handling and supplying fresh air. In this way, the indication effect of the first penalty function, the second penalty function, and the third penalty function can be effectively improved.

[0056] The default value of the expected reference temperature may be set as 24°C, and then flexibly adjusted according to application scenarios. The preset control threshold value can also be flexibly adjusted according to application scenarios, for example, can be configured as 800 ppm.

[0057] Optionally, in some embodiments, comfort feedback data of the indoor temperature of the user in the building served by the HVAC system can also be obtained; the expected reference temperature is updated according to the comfort feedback data; and the expected reference temperature is reset to a preset default value based on a preset time interval. Thus, the real-time update of the expected reference temperature can be realized in combination with the actual comfort feedback data of the user in the building, so as to ensure that the HVAC system control responds to the individual thermal comfort of the user in the building, and the expected reference temperature is reset to the preset default value based on the preset time interval, so as to effectively avoid the drift of the temperature preference due to long-term accumulation.

[0058] The comfort feedback data can be used to indicate the user's suggestions related to the indoor temperature comfort, such as increasing the temperature, decreasing the temperature, keeping the temperature, etc. The preset time interval can be flexibly configured according to application scenarios, for example, can be 24 hours, or 7 days, etc., without limitation.

[0059] In this embodiment, the initial number of people, the initial indoor temperature, and the initial indoor carbon dioxide concentration of the building served by the HVAC system are determined; the predicted number of people and the room occupancy state at the target time point are determined according to the initial number of people; the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point are determined according to the initial indoor temperature, the initial indoor carbon dioxide concentration, and the predicted number of people; and the room occupancy state, the predicted indoor temperature, and the predicted indoor carbon dioxide concentration are input into the pre-trained model optimizer to obtain the indoor air temperature set value and the fresh air flow of the HVAC system at the target time point, wherein the optimization objectives of the model optimizer include thermal comfort penalty, indoor air quality penalty, and energy consumption penalty. Thus, efficient balance between energy efficiency optimization, residential comfort, and air quality can be achieved.

[0060] Figure 2 is a flowchart of the HVAC system control method provided according to the first embodiment of the present disclosure.

[0061] As shown in Figure 2 , the method comprises:

[0062] Step 201, determining the initial number of people, the initial indoor temperature, and the initial indoor carbon dioxide concentration of the building served by the HVAC system.

[0063] Step 202, determining the predicted number of people and the room occupancy state at the target time point according to the initial number of people.

[0064] The description of steps 201 to 202 can be referred to the above-mentioned embodiments, which will not be repeated here.

[0065] In step 203, outdoor meteorological prediction data of the building served by the HVAC system at the target time point is obtained, wherein the outdoor meteorological prediction data includes solar radiation and outdoor temperature.

[0066] The outdoor meteorological prediction data can be obtained from a weather forecast.

[0067] It can be understood that the outdoor temperature and the solar radiation can affect the environmental parameters inside the building, and therefore, in the embodiments of the present disclosure, the outdoor meteorological prediction data of the building served by the HVAC system at the target time point is obtained, thereby providing data support in the outdoor meteorological dimension for subsequent construction of the building dynamic model.

[0068] In step 204, the number-related disturbance prediction data at the target time point is determined based on the predicted number of people, wherein the number-related disturbance prediction data includes indoor heat disturbance data and carbon dioxide generation data.

[0069] The number-related disturbance prediction data can be used to describe the influence of the change in the number of people on the relevant environmental parameters. The indoor heat disturbance data refers to the change in indoor heat caused by the change in the number of people. The carbon dioxide generation data refers to the change in indoor carbon dioxide concentration caused by the change in the number of people.

[0070] That is to say, in the embodiments of the present disclosure, after the predicted number of people at the target time point is determined, the number-related disturbance prediction data at the target time point is determined based on the predicted number of people, thereby providing data support of indoor human factors for subsequent construction of the building dynamic model.

[0071] In step 205, a building dynamic model based on a stochastic differential equation is constructed according to the outdoor meteorological prediction data and the number-related disturbance prediction data, wherein the building dynamic model includes a first model describing the change in indoor temperature and a second model describing the change in indoor carbon dioxide concentration.

[0072] The building dynamic model can be used to describe the dynamic evolution process of the room temperature and the carbon dioxide concentration.

[0073] For example, in the embodiments of the present disclosure, the building thermal dynamic model can be constructed based on an RC thermal network, and the change in indoor carbon dioxide concentration can be described based on a mass balance equation, wherein the RC thermal network is a network composed of thermal resistance (R) and thermal capacity (C), and based on the RC thermal network, the thermal system can be analyzed by the method of analyzing the circuit.

[0074] Optionally, in some embodiments, the first model includes a first control variable for describing the cooling supply of the room, the first control variable being less than or equal to the maximum cooling capacity of the fan-coil unit in the HVAC system; and the second model includes a second control variable for describing the fresh air flow supply, the second control variable being within a predefined range of air flow values. In this way, the operation constraints of the building dynamic model can be implemented to ensure the authenticity of the obtained first model and second model.

[0075] That is, in the embodiments of the present disclosure, after obtaining the outdoor weather forecast data and the number-related disturbance forecast data, the building dynamic model based on the stochastic differential equation can be constructed according to the outdoor weather forecast data and the number-related disturbance forecast data, thereby providing a reliable execution tool for subsequently determining the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point.

[0076] In step 206, the predicted indoor temperature at the target time point is determined based on the first model and the initial indoor temperature.

[0077] In step 207, the predicted indoor carbon dioxide concentration at the target time point is determined based on the second model and the initial indoor carbon dioxide concentration.

[0078] That is, in the embodiments of the present disclosure, the outdoor weather forecast data of the building served by the HVAC system at the target time point can be obtained, wherein the outdoor weather forecast data includes solar radiation and outdoor temperature; the number-related disturbance forecast data at the target time point is determined based on the predicted number of people, wherein the number-related disturbance forecast data includes indoor heat disturbance data and carbon dioxide generation data; a building dynamic model based on the stochastic differential equation is constructed according to the outdoor weather forecast data and the number-related disturbance forecast data, wherein the building dynamic model includes a first model describing the change of indoor temperature and a second model describing the change of indoor carbon dioxide concentration; the predicted indoor temperature at the target time point is determined based on the first model and the initial indoor temperature; and the predicted indoor carbon dioxide concentration at the target time point is determined based on the second model and the initial indoor carbon dioxide concentration. In this way, the building dynamic model can be constructed based on the outdoor weather forecast data and the number-related disturbance forecast data, and the accurate prediction of the indoor temperature and the indoor carbon dioxide concentration at the target time point can be realized.

[0079] In step 208, the room occupancy state, the predicted indoor temperature, and the predicted indoor carbon dioxide concentration are input into the pre-trained model optimizer to obtain the indoor air temperature set value of the HVAC system at the target time point and the fresh air flow, wherein the optimization objectives of the model optimizer include thermal comfort penalty, indoor air quality penalty, and energy consumption penalty.

[0080] The description of step 208 can be found in the above embodiments, and will not be repeated here.

[0081] In this embodiment, outdoor meteorological forecast data of the building served by the HVAC system at a target time point is acquired. This outdoor meteorological forecast data includes solar radiation and outdoor temperature. Based on the predicted number of people, human-related disturbance forecast data for the target time point is determined. This human-related disturbance forecast data includes indoor thermal disturbance data and carbon dioxide generation data. Based on the outdoor meteorological forecast data and the human-related disturbance forecast data, a building dynamic model based on stochastic differential equations is constructed. This building dynamic model includes a first model describing indoor temperature changes and a second model describing indoor carbon dioxide concentration changes. Based on the first model and the initial indoor temperature, the predicted indoor temperature at the target time point is determined. Based on the second model and the initial indoor carbon dioxide concentration, the predicted indoor carbon dioxide concentration at the target time point is determined. Therefore, a building dynamic model can be constructed based on outdoor meteorological forecast data and human-related disturbance forecast data, thereby achieving accurate prediction of indoor temperature and indoor carbon dioxide concentration at the target time point.

[0082] In summary, the HVAC system control method proposed in this disclosure can employ camera-based personnel sensing technology to capture real-time occupancy data, while simultaneously allowing personnel to provide immediate comfort feedback through a QR code-based voting system. This real-time data is combined with a personnel pattern model trained on historical data. The system can optimize the operation of the HVAC system in real time. The system utilizes cloud and edge computing platforms to collaboratively process data, ensuring real-time responsiveness and control accuracy.

[0083] 1) System Architecture

[0084] This invention proposes an IoT-supported "human-in-the-loop" multi-objective model optimized control architecture (e.g., Figure 3 As shown, Figure 3 This is a schematic diagram of the architecture for the IoT-supported human-in-the-loop model-based optimized control method proposed in this disclosure. The Model Predictive Control (MPC) architecture consists of three layers: a data acquisition and field control layer, a data transmission layer, and an application layer based on cloud-edge collaboration. The core features of this architecture are:

[0085] (1) Human-centered control loop, including the collection of personnel data, personnel feature modeling and prediction, and model optimization control strategy based on personnel prediction and real-time voting feedback.

[0086] (2) Balancing multiple control objectives, including improving indoor occupant comfort, improving air quality and improving energy efficiency.

[0087] (3) The cloud and edge computing platforms work together to process the data. The cloud devices train models based on historical data, and the edge devices deploy optimization control algorithms based on real-time model predictions and send optimization control signals.

[0088] In the data acquisition and field control layer, this layer mainly realizes two functions: first, it collects real-time data from the physical world through IoT sensors, IoT thermostats, IoT chilled water meters, building automation systems (BAS), and QR codes; second, it transmits the optimized control setpoints to the actuators in the HVAC system through the IoT to realize field control.

[0089] At the data transmission layer, this layer is responsible for real-time data exchange and protocol conversion. It supports multiple communication protocols (such as MQTT, HTTP, and BACnet) through gateways and APIs, ensuring seamless interoperability between different subsystems. The MQTT protocol enables efficient and lightweight communication between IoT gateways and cloud servers and edge computing devices; the HTTP API is used to receive user feedback and weather forecast data; and the BACnet protocol supports bidirectional communication between servers and traditional BAS gateways, facilitating the monitoring and optimized control of HVAC systems.

[0090] The cloud-edge collaborative application layer integrates various software services, including real-time data processing and storage, data visualization, and advanced data analysis and model prediction optimization. Cloud devices train data-driven models based on historical data. Edge devices deploy optimization control algorithms based on real-time model predictions, combining real-time measurement data with data-driven models to dynamically optimize temperature and ventilation setpoints, ensuring the optimal balance between thermal comfort, indoor air quality, and energy efficiency.

[0091] 2) Model Predictive Control (MPC)

[0092] The "human-in-the-loop" MPC solution proposed in this invention integrates real-time resident numbers, resident feedback, and random resident prediction models (such as...). Figure 4 As shown, Figure 4 This is a schematic diagram of the human-in-the-loop model optimization control scheme based on the integrated random number model, real-time number of people, and real-time personnel feedback proposed in this disclosure.

[0093] 2.1 Random headcount prediction based on non-homogeneous Markov chains (IMC)

[0094] This invention employs the non-homogeneous Markov chain (IMC) method for modeling random number of people. A key feature of the IMC method is its time-varying transition probability matrix, which defines the probability of state transitions over time, as well as the initial state (or initial distribution) in the state space. State space denote different ranges of people. The transition probabilities between states at a given time step t form a transition matrix As shown in equation (1).

[0095]

[0096] The adaptive B-spline method is used to estimate the transition probability parameters . Then, the IMC-based stochastic population model is used for real-time online population prediction. Assuming that the current time is k, the current population measurement corresponds to the state , the initial state probability distribution at time k is , where and . The population state probability distribution of the subsequent n time steps is obtained by the forward Kolmogorov equation, as shown in equation (2):

[0097]

[0098] The expected population at the future time step is calculated by equation (3):

[0099]

[0100] where denotes the expected population at time t, t = k + 1, k + 2... k + n . In addition, the predicted room occupancy state is denoted by , which is 1 if the condition is met, and 0 otherwise.

[0101]

[0102] In the present invention, the threshold value α = 0.5 is used to determine the expected room occupancy state.

[0103] 2.2 Stochastic Differential Equation (SDE) based building dynamic modeling

[0104] In terms of building dynamic modeling, the present invention uses stochastic differential equations (SDE) to describe the dynamic evolution process of room temperature and carbon dioxide concentration. The present invention builds a building thermal dynamic model based on RC thermal network, and describes the change of indoor carbon dioxide concentration based on mass balance equation. As shown in Figure 5 , the thermal resistance and thermal capacity diagram showing the thermal dynamic structure of the building is proposed according to the present disclosure. The SDE model describing the building thermal dynamics is given in equations (5) and (6), while the SDE model of room carbon dioxide concentration is given in equation (7). Figure 5

[0105] (5) ​

[0106] (6)

[0107] (7)

[0108] The continuous-time SDE model is discretized to ensure more efficient computation. The stochastic discrete-time model for room temperature and carbon dioxide concentration takes the form

[0109]

[0110] where, is the state vector of room temperature and external wall temperature at time step t, denoted as ; is the disturbance vector, containing the number of occupants, solar radiation, and outdoor temperature, denoted as . In the online optimization, the future number of occupants is predicted using the IMC model (equations 2, 3), while the solar radiation and outdoor temperature can be obtained from weather forecasts. These predictions form the predicted disturbance vector . is the control variable, representing the cooling supply to the room . This control variable is constrained by the maximum cooling capacity of the FCU.

[0111]

[0112] In practical implementation, the indoor air temperature is controlled by adjusting the temperature setpoint on the FCU thermostat.

[0113] For carbon dioxide concentration modeling (equation 9), is the system state of indoor carbon dioxide concentration at time step t; is the control variable, representing the fresh air flow supply . The fresh air flow is constrained to ensure it is within predefined minimum and maximum limits.

[0114]

[0115] The main disturbance affecting carbon dioxide concentration is the number of occupants . In the online optimization, the predicted future number of occupants is used as the disturbance variable. , , , , and It is the corresponding matrix of the discrete-time model. and The process noise that indicates indoor air temperature and CO2 concentration.

[0116] 2.3 MPC Structure and Multi-Objective Optimization

[0117] The optimized controller developed in this invention aims to optimize the prediction range. N Intra-time step t Indoor air temperature setting value ( ) and fresh air flow ( This is achieved by solving a multi-objective rolling optimization problem at each time step. The weighted sum of the following penalty terms is minimized: Thermal comfort penalty (i.e., the first penalty function mentioned above); Indoor air quality penalty (i.e., the second penalty function mentioned above); and Energy consumption penalty (i.e., the third penalty function mentioned above). Parameters (i.e., the first importance descriptive value mentioned above) (i.e., the second importance descriptive value mentioned above) and (i.e., the third importance descriptor value mentioned above) defines the relative importance of each objective, which can be adjusted by operations personnel to reflect a specific priority. The overall MPC structure is shown in formulas (12) and (13). The constraints of the optimization control problem include: occupancy prediction constraints, building dynamic constraints, HVAC system constraints, and initial condition constraints.

[0118]

[0119] in, That is, the first product value mentioned above. That is, the second product value mentioned above. That is, the third product value mentioned above. This refers to the room occupancy status described above. That is, the first sum mentioned above, corresponding to Then it is the fourth product value mentioned above.

[0120] subject to

[0121]

[0122] The three optimization objectives are as follows.

[0123] (1) Thermal comfort penalty )

[0124] Thermal comfort is quantified by a quadratic penalty function, as shown in Equation (14).

[0125]

[0126] This function evaluates the deviation between the indoor air temperature setpoint and the desired reference temperature In this work, The default value is set to 24 °C. However, the voting feedback of the occupants influences the value of , so that the occupants’ preferences are better satisfied.

[0127] (2) Indoor air quality penalty

[0128] The indoor air quality is represented by the CO2 concentration as an indicator of the indoor pollutant level. The cost function for controlling the IAQ is given in equation (15).

[0129]

[0130] where is the control threshold for CO2, set to 800 ppm. To ensure computational efficiency, the max operation in the penalty function is linearized by introducing an auxiliary variable and an additional constraint.

[0131]

[0132]

[0133]

[0134] (3) Energy consumption penalty

[0135] The energy consumption includes the energy consumption required for providing cooling to the room and for handling and supplying fresh air. The energy consumption definition is given in equation (19).

[0136]

[0137] The cooling demand of fresh air is calculated as the product of the specific enthalpy difference between the supply fresh air and the outdoor air and the fresh air flow rate . Here, and denote the specific enthalpy of the supply fresh air and the outdoor air, respectively. The enthalpy of the outdoor air is dynamically calculated using the predicted outdoor temperature and RH data. The coefficient of performance (COP) represents the overall efficiency of the chiller and the auxiliary cooling distribution system.

[0138] ​​The MPC formulation as described in equations (12) and (13) is converted into a quadratic programming (QP) problem and solved using an optimization solver. Model predictive control (MPC) employs an iterative rolling optimization method, optimizing over a moving prediction horizon. The optimization at each time step generates a sequence of control signals, where only the first control signal is implemented in the actual system. At the next time step, the optimization is recalculated using updated real-time data.

[0139] 3) Real-time occupant comfort feedback

[0140] While traditional MPC systems usually assume a fixed comfort temperature range, the present invention introduces a real-time occupant feedback mechanism, incorporating occupant's personalized needs into the control loop. The implementation of this system is shown in Figure 6 Figure 6 is a schematic diagram of the operation of a real-time occupant comfort feedback system based on QR codes according to the present disclosure. Occupants can participate in the voting by scanning the QR code to report their personalized comfort needs. The system dynamically adjusts the reference temperature (Tref) according to the voting results and incorporates it into the MPC optimization process, ensuring that the control strategy can respond to occupant preferences in real time. In addition, to prevent the temperature preference from drifting due to long-term accumulation, the system resets the reference temperature to the default value (24°C) every day at midnight.

[0141] 4) Laboratory actual deployment and Internet of Things transformation

[0142] The present invention has been successfully deployed in an office living laboratory of a certain university, which includes a large office, a small office and a conference room, and the HVAC system is composed of fan coil units (FCUs) and a main air handling unit (PAU). During the transformation, LoRa-based Internet of Things sensors, temperature controllers, and QR code feedback systems were installed. The conference room was chosen as the target space for testing the "man-in-the-loop" MPC because of its highly irregular use pattern, with significant randomness and intermittency. The experimental test period covers three control strategies: traditional fixed set value control (FIX), occupant feedback control based on simple rules (OBC), and MPC control proposed by the present invention. The test results show that the present invention performs significantly better under different living conditions.

[0143] Based on the above embodiments, the present disclosure can at least include the following technical effects:

[0144] (1) By optimizing pre-cooling and dynamically adjusting fresh air flow, the system significantly reduces unnecessary energy consumption;

[0145] (2) Significantly improves thermal comfort. The system also dynamically adjusts the temperature set point through real-time occupant feedback, ensuring that the occupant's personalized comfort needs are met;

[0146] ​(3) Avoids excessive ventilation and insufficient ventilation. The system dynamically adjusts the fresh air flow to ensure that the indoor air quality is always within the comfortable range;

[0147] (4) The system quickly responds to personalized comfort feedback through a real-time voting mechanism, ensuring a comfortable experience for personnel. Personnel can provide feedback by scanning a two-dimensional code, and the system dynamically adjusts the control strategy based on the feedback to ensure the flexibility and response speed of the system.

[0148] To better implement the HVAC system control method of the present disclosure, the present disclosure also provides a HVAC system control device based on the above-mentioned HVAC system control method. The meanings of the terms are the same as in the above-mentioned HVAC system control method, and specific implementation details can be referred to the description in the method embodiment.

[0149] Please refer to Figure 7 , Figure 7 is a structural schematic diagram of the HVAC system control device provided by the embodiment of the present disclosure. The HVAC system control device 700 comprises:

[0150] The first determination module 701 is configured to determine the initial number of people, the initial indoor temperature, and the initial indoor carbon dioxide concentration of the building served by the HVAC system.

[0151] The second determination module 702 is configured to determine the predicted number of people and the room occupancy state at the target time point according to the initial number of people.

[0152] The third determination module 703 is configured to determine the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration, and the predicted number of people.

[0153] The processing module 704 is configured to input the room occupancy state, the predicted indoor temperature, and the predicted indoor carbon dioxide concentration into the pre-trained model optimizer to obtain the indoor air temperature set value and the fresh air flow of the HVAC system at the target time point, wherein the optimization objectives of the model optimizer include thermal comfort penalty, indoor air quality penalty, and energy consumption penalty.

[0154] Optionally, the second determination module 702 is specifically configured to:

[0155] Adopting the adaptive B-spline method to dynamically estimate the transition probability parameters, wherein the plurality of transition probability parameters jointly constitute a transition probability matrix;

[0156] Determine the initial state probability distribution according to the initial number of people;

[0157] Determine the predicted number of people at the target time point based on the transition probability matrix and the initial state probability distribution;

[0158] Determine the room occupancy state at the target time point based on the predicted number of people at the target time point and a preset threshold.

[0159] Optionally, the third determination module 703 is specifically configured to:

[0160] Obtain outdoor weather forecast data of the building served by the HVAC system at the target time point, wherein the outdoor weather forecast data includes solar radiation and outdoor temperature.

[0161] Determine the number-related disturbance prediction data at the target time point based on the predicted number of people, wherein the number-related disturbance prediction data includes indoor thermal disturbance data and carbon dioxide generation data.

[0162] Construct a building dynamic model based on a stochastic differential equation according to the outdoor weather forecast data and the number-related disturbance prediction data, wherein the building dynamic model includes a first model describing changes in indoor temperature and a second model describing changes in indoor carbon dioxide concentration.

[0163] Determine the predicted indoor temperature at the target time point based on the first model and the initial indoor temperature.

[0164] Determine the predicted indoor carbon dioxide concentration at the target time point based on the second model and the initial indoor carbon dioxide concentration.

[0165] Optionally, wherein,

[0166] The first model includes a first control variable for describing the cooling supply of the room, and the first control variable is less than or equal to the maximum cooling capacity of the fan-coil unit in the HVAC system.

[0167] The second model includes a second control variable for describing the supply of fresh air flow, and the value of the second control variable belongs to a predefined air flow value range.

[0168] Optionally, the model optimizer performs multi-objective optimization control in the following manner:

[0169] Construct a first penalty function corresponding to the thermal comfort penalty, a second penalty function corresponding to the indoor air quality penalty, and a third penalty function corresponding to the energy consumption penalty.

[0170] Determine a first importance description value corresponding to the thermal comfort penalty, a second importance description value corresponding to the indoor air quality penalty, and a third importance description value corresponding to the energy consumption penalty.

[0171] Determine a first product value of the first penalty function and the first importance description value, a second product value of the second penalty function and the second importance description value, and a third product value of the third penalty function and the third importance description value.

[0172] determining a first sum value of the first product value and the second product value, and determining a fourth product value of the first sum value and the room occupancy state;

[0173] determining a second sum value of the fourth product value and the third product value;

[0174] performing a minimization process on a cumulative sum of the second sum values corresponding to a plurality of target time points within a prediction time range, to obtain an indoor air temperature set value corresponding to each target time point and a fresh air flow.

[0175] Optionally, wherein,

[0176] the first penalty function is used to evaluate a deviation between the indoor air temperature set value and the expected reference temperature;

[0177] the second penalty function is used to indicate a case where the indoor carbon dioxide concentration exceeds a preset control threshold, wherein the second penalty function takes a value of zero when the indoor carbon dioxide concentration is less than the preset control threshold;

[0178] the third penalty function is used to indicate an energy consumption required for providing cooling for the room and for processing and supplying fresh air.

[0179] Optionally, the apparatus further comprises:

[0180] an acquisition module configured to acquire comfort feedback data of indoor temperatures corresponding to users in a building served by the HVAC system;

[0181] an updating module configured to update the expected reference temperature according to the comfort feedback data;

[0182] a resetting module configured to reset the expected reference temperature to a preset default value based on a preset time interval.

[0183] In the embodiments of the present disclosure, the initial number of people, the initial indoor temperature and the initial indoor carbon dioxide concentration of a building served by the HVAC system are determined; the predicted number of people and the room occupancy state at a target time point are determined according to the initial number of people; the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point are determined according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people; the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration are input into a pre-trained model optimizer to obtain the indoor air temperature set value and the fresh air flow of the HVAC system at the target time point, wherein the optimization objectives of the model optimizer include a thermal comfort penalty, an indoor air quality penalty and an energy consumption penalty. In this way, an efficient balance among energy efficiency optimization, residential comfort and air quality can be achieved.

[0184] In addition, the present disclosure also provides an electronic device, such as Figure 8As shown in the figure, it shows a structural schematic diagram of an electronic device related to the present disclosure, in particular:

[0185] The electronic device can include a processor 801 with one or more processing cores, a memory 802 with one or more computer readable storage media, a power supply 803, and an input unit 804, etc. Those skilled in the art can understand that, Figure 8 The electronic device structure shown in the figure does not constitute a limitation on the electronic device, and can include more or fewer components than the figure, or combine certain components, or different component arrangements. Among them:

[0186] The processor 801 is the control center of the electronic device, which connects all parts of the electronic device through various interfaces and lines, executes various functions of the electronic device and processes data by running or executing software programs and / or modules stored in the memory 802, and calling data stored in the memory 802, thereby overall monitoring the electronic device. Optionally, the processor 801 can include one or more processing cores; preferably, the processor 801 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application program, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 801.

[0187] The memory 802 can be used to store software programs and modules, and the processor 801 executes various functions and data processing by running the software programs and modules stored in the memory 802. The memory 802 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), etc.; the data storage area can store data created according to the use of the electronic device, etc. In addition, the memory 802 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device. Accordingly, the memory 802 can also include a memory controller to provide access for the processor 801 to the memory 802.

[0188] The electronic device also includes a power supply 803 for powering various components, and preferably the power supply 803 can be logically connected to the processor 801 through a power management system, so as to realize the functions of managing charging, discharging, and power consumption management, etc. through the power management system. The power supply 803 can also include one or more direct current or alternating current power supplies, a recharging system, a power supply device debugging circuit, a power supply converter or inverter, a power supply state indicator, etc. Any component.

[0189] The electronic device can further include an input unit 804, which can be used to receive inputted digital or character information, and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0190] Although not shown, the electronic device can further include a display unit, etc., which will not be described here. In particular in the present embodiment, the processor 801 in the electronic device will load the executable file corresponding to the process of one or more application programs into the memory 802 according to the following instructions, and run the application program stored in the memory 802 by the processor 801, thereby implementing the steps in any of the HVAC system control methods provided by the embodiments of the present disclosure.

[0191] The specific implementation of each of the above operations can refer to the previous embodiments, which will not be described here.

[0192] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by instructions controlling relevant hardware, which can be stored in a computer readable storage medium and loaded and executed by a processor.

[0193] To this end, the present disclosure provides a computer readable storage medium, which stores a computer program capable of being loaded by a processor to execute the steps in any of the HVAC system control methods provided by the present disclosure.

[0194] The specific implementation of each of the above operations can refer to the previous embodiments, which will not be described here.

[0195] The computer readable storage medium can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0196] Since the instructions stored in the computer readable storage medium can execute the steps in any of the HVAC system control methods provided by the present disclosure, the beneficial effects that can be achieved by any of the HVAC system control methods provided by the present disclosure can be achieved, which will be described in detail in the previous embodiments, which will not be described here.

[0197] The above describes in detail the HVAC system control method, device, electronic device and computer readable storage medium provided by the present disclosure. The principles and implementation manners of the present disclosure are described by using specific examples. The above description of the examples is only used to help understand the method of the present disclosure and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present disclosure, the specific implementation manners and application ranges can be changed. In conclusion, the content of the specification should not be understood as a limitation of the present disclosure.

Claims

1. A heating, ventilation, and air conditioning system control method, characterized by, The method comprises the following steps: determining the initial number of people, the initial indoor temperature and the initial indoor carbon dioxide concentration of a building served by a heating, ventilation and air conditioning (HVAC) system; determining the predicted number of people and the room occupancy state at a target time point according to the initial number of people; determining the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people; inputting the room occupancy state, the predicted indoor temperature and the predicted indoor carbon dioxide concentration into a pre-trained model optimizer to obtain the indoor air temperature set value and the fresh air flow of the HVAC system at the target time point, wherein the optimization objectives of the model optimizer include a thermal comfort penalty, an indoor air quality penalty and an energy consumption penalty, and wherein the determination of the predicted number of people and the room occupancy state at the target time point according to the initial number of people comprises: dynamically estimating transition probability parameters by using an adaptive B-spline method, wherein a plurality of the transition probability parameters jointly constitute a transition probability matrix; determining an initial state probability distribution according to the initial number of people; determining the predicted number of people at the target time point based on the transition probability matrix and the initial state probability distribution; determining the room occupancy state at the target time point based on the predicted number of people at the target time point and a preset threshold, the determination of the predicted indoor temperature and the predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people comprises: obtaining outdoor weather forecast data of the building served by the HVAC system at the target time point, wherein the outdoor weather forecast data includes solar radiation and outdoor temperature; determining number-related disturbance prediction data at the target time point based on the predicted number of people, wherein the number-related disturbance prediction data includes indoor thermal disturbance data and carbon dioxide generation data; constructing a building dynamic model based on stochastic differential equations according to the outdoor weather forecast data and the number-related disturbance prediction data, wherein the building dynamic model includes a first model describing the change of indoor temperature and a second model describing the change of indoor carbon dioxide concentration; determining the predicted indoor temperature at the target time point based on the first model and the initial indoor temperature; determining the predicted indoor carbon dioxide concentration at the target time point based on the second model and the initial indoor carbon dioxide concentration.

2. The method of claim 1, wherein, wherein the first model includes a first control variable for describing the cooling supply of a room, and the first control variable is less than or equal to the maximum cooling capacity of a fan-coil unit in the HVAC system; the second model includes a second control variable for describing the supply of fresh air flow, and the value of the second control variable belongs to a predefined air flow value range.

3. The method of claim 1, wherein, The model optimizer performs multi-objective optimization control in the following manner: constructing a first penalty function corresponding to the thermal comfort penalty, a second penalty function corresponding to the indoor air quality penalty, and a third penalty function corresponding to the energy consumption penalty; determining a first importance description value corresponding to the thermal comfort penalty, a second importance description value corresponding to the indoor air quality penalty, and a third importance description value corresponding to the energy consumption penalty; determining a first product value of the first penalty function and the first importance description value, a second product value of the second penalty function and the second importance description value, and a third product value of the third penalty function and the third importance description value; determining a first sum value of the first product value and the second product value, and determining a fourth product value of the first sum value and the room occupancy state; determining a second sum value of the fourth product value and the third product value; performing a minimization process on a cumulative sum of the second sum value corresponding to a plurality of target time points within a prediction time range to obtain the indoor air temperature set value corresponding to each target time point and the fresh air flow.

4. The method of claim 3, wherein, wherein, the first penalty function is used to evaluate the deviation between the indoor air temperature set value and the expected reference temperature; the second penalty function is used to indicate the case that the indoor carbon dioxide concentration exceeds the preset control threshold, wherein the second penalty function takes a value of zero when the indoor carbon dioxide concentration is less than the preset control threshold; the third penalty function is used to indicate the energy consumption required for providing cooling for the room and processing and supplying fresh air.

5. The method of claim 4, wherein, The method further comprises: obtaining comfort feedback data of indoor temperature corresponding to users in a building served by the heating ventilation and air conditioning system; updating the expected reference temperature according to the comfort feedback data; resetting the expected reference temperature to a preset default value based on a preset time interval.

6. A heating, ventilation, and air conditioning system control device, comprising: comprise: a first determination module configured to determine an initial number of people, an initial indoor temperature, and an initial indoor carbon dioxide concentration of a building served by a heating ventilation and air conditioning system; a second determination module configured to determine a predicted number of people and a room occupancy state at a target time point according to the initial number of people; a third determination module configured to determine a predicted indoor temperature and a predicted indoor carbon dioxide concentration at the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration, and the predicted number of people; a processing module configured to input the room occupancy state, the predicted indoor temperature, and the predicted indoor carbon dioxide concentration into a pre-trained model optimizer to obtain an indoor air temperature set value and a fresh air flow of the heating ventilation and air conditioning system at the target time point, wherein the optimization objective of the model optimizer comprises a thermal comfort penalty, an indoor air quality penalty, and an energy consumption penalty; wherein the third determination module is specifically configured to: dynamically estimate transition probability parameters using an adaptive B-spline method, wherein a plurality of the transition probability parameters jointly constitute a transition probability matrix; determine an initial state probability distribution according to the initial number of people; determine the predicted number of people at the target time point based on the transition probability matrix and the initial state probability distribution; determining the room occupancy state of the target time point based on the predicted number of people of the target time point and a preset threshold, determining the predicted indoor temperature and the predicted indoor carbon dioxide concentration of the target time point according to the initial indoor temperature, the initial indoor carbon dioxide concentration and the predicted number of people, comprises: obtaining outdoor weather forecast data of a building served by the heating ventilation air conditioning system at the target time point, wherein the outdoor weather forecast data comprises solar radiation and outdoor temperature; determining number-of-people related disturbance prediction data at the target time point based on the predicted number of people, wherein the number-of-people related disturbance prediction data comprises indoor heat disturbance data and carbon dioxide generation data; constructing a building dynamic model based on a random differential equation according to the outdoor weather forecast data and the number-of-people related disturbance prediction data, wherein the building dynamic model comprises a first model describing changes in indoor temperature and a second model describing changes in indoor carbon dioxide concentration; determining the predicted indoor temperature of the target time point based on the first model and the initial indoor temperature; determining the predicted indoor carbon dioxide concentration of the target time point based on the second model and the initial indoor carbon dioxide concentration.

7. An electronic device, comprising: a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the method of any one of claims 1 to 5.

8. A storage medium, characterized by The storage medium is used to store a computer program, and the computer program is loaded by a processor to execute the method of any one of claims 1 to 5.

Images