Method, device, terminal and storage medium for quantifying flexible regulation potential of air conditioning system

Abstract

The present invention provides a method, device, terminal, and storage medium for quantifying the flexible adjustment potential of an air-conditioning system, which belongs to the field of building energy management technology. The method includes: continuously collecting indoor and outdoor environmental data and air-conditioning system operating data of a target building, and screening out a first training data subset and a second training data subset; constructing a room temperature prediction model; sequentially determining the air-conditioning system thermal resistance, thermal resistance and heat capacity parameter combination, and indoor heating time schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation; obtaining a room temperature prediction curve; and determining the maximum adjustable time as the flexible adjustment potential of the air-conditioning system. The present invention takes into account the air-conditioning system operating data when constructing the room temperature prediction model, and uses the model to solve the maximum response time of the air-conditioning system during demand response, which can effectively improve the quantification accuracy of the flexible adjustment potential of the air-conditioning system and provide accurate data support for demand response strategies.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy management, and in particular to a method, device, terminal and storage medium for quantifying the flexible regulation potential of an air-conditioning system. Background Art

[0002] Currently, the proportion of renewable energy generation, led by wind and photovoltaic power, continues to rise. However, the volatility and intermittent nature of renewable energy generation exacerbates the pressure on the power grid to balance supply and demand, leading to significant imbalances between power supply and demand during peak and off-peak periods. Demand response (DR) technology, by dynamically adjusting user-side loads to achieve a paradigm shift from "source follows load" to "load follows source," has become a key technological path for improving grid resilience and promoting the integration of renewable energy. Public buildings are major energy consumers, with air conditioning systems accounting for up to 70% of summer energy consumption. Because the building's inherent energy storage and the thermal inertia of air conditioning allow for short-term load reductions, minimizing the impact on human thermal comfort, air conditioning systems have become a highly promising flexible regulation resource. By accurately quantifying the flexible regulation potential of air conditioning systems, it is possible to maximize the grid's peak-shaving and off-peak-filling benefits while minimizing user-perceived disturbances, promoting the synergy between building energy conservation and emission reduction and grid stability.

[0003] Existing technologies typically use a thermal resistance and heat capacity model based on equivalent thermal network theory as a room temperature prediction model. Using thermal resistance and heat capacity parameters to describe the heat transfer process within a building system, these models can simulate dynamic characteristics such as indoor and outdoor heat exchange, providing theoretical support for flexible quantification. However, because these models fail to account for the heat storage and enthalpy-humidity dynamics of the air conditioning system itself, the results of determining the flexible regulation potential of the air conditioning system are inaccurate, making it difficult to provide accurate data support for demand response strategies.

[0004] Therefore, the existing technology has defects and needs to be improved and developed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method, device, terminal and storage medium for quantifying the flexible adjustment potential of an air-conditioning system in response to the above-mentioned defects of the prior art, aiming to solve the problem in the prior art that it is difficult to provide accurate data support for demand response strategies.

[0006] The technical solutions adopted by the present invention to solve the technical problems are as follows:

[0007] In a first aspect, an embodiment of the present invention provides a method for quantifying the flexible adjustment potential of an air-conditioning system, the method comprising:

[0008] Utilizing a plurality of preset sensors to continuously collect indoor and outdoor environmental data and air conditioning system operation data of the target building, a first training data subset and a second training data subset are selected from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data;

[0009] Building a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is used to predict changes in temperature in a target building after a demand response is initiated;

[0010] Sequentially calculating the thermal resistance of the air conditioning system, the combination of thermal resistance and heat capacity parameters, and the indoor heating time schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation;

[0011] When a demand response instruction is received, the wall temperature at the time of demand response initiation is calculated, and the wall temperature and the indoor and outdoor environment data and air conditioning system operation data collected last time before the demand response initiation are substituted into the room temperature prediction model with fixed parameters to obtain the room temperature prediction curve;

[0012] Based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, the maximum adjustable time is determined and used as the flexible adjustment potential of the air-conditioning system.

[0013] In one embodiment, the indoor and outdoor environmental data include indoor air temperature and outdoor air dry-bulb temperature, and the air conditioning system operation data include chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity. A first training data subset and a second training data subset are screened from historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, including:

[0014] From the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during a first historical time period are selected as a first training data subset. During the first historical time period, the target building is in a zero-heat state, the air conditioning compressor is turned off, and the water pump and terminal air handling unit are running.

[0015] From the historically collected time-series indoor and outdoor environmental data and the time-series air-conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during the second historical time period are selected as the second training data subset. During the second historical time period, the target building is in an internally heated state and the air-conditioning system operates stably.

[0016] In one embodiment, constructing a room temperature prediction model using the first training data subset and the second training data subset includes:

[0017] Filtering the chilled water outlet temperature, chilled water return temperature, and indoor air temperature at a target time from the first training data subset, wherein the target time is the time when the air conditioning system temperature reaches the dew point temperature when the target building is in a state of no internal heat, the air conditioning compressor is turned off, and the water pump and the terminal air handling unit are running;

[0018] Obtaining a first air-conditioning system temperature based on a chilled water outlet temperature and a chilled water return temperature at a target time, and obtaining a temperature difference coefficient based on the first air-conditioning system temperature and an indoor air temperature at the target time;

[0019] obtaining a return air enthalpy value based on the return air temperature and indoor humidity of the second training data subset, and obtaining a supply air enthalpy value based on the supply air temperature and indoor humidity of the second training data subset;

[0020] Obtaining a preset air specific heat capacity, and obtaining a sensible heat ratio based on the air specific heat capacity, the return air temperature of the second training data subset, the supply air temperature of the second training data subset, the return air enthalpy value, and the supply air enthalpy value;

[0021] Obtaining a latent heat and sensible heat ratio based on the sensible heat ratio, and obtaining a second air-conditioning system temperature based on the chilled water outlet temperature and the chilled water return temperature of the second training data subset;

[0022] Obtaining a proportionality coefficient based on the ratio of latent heat to sensible heat, the indoor air temperature of the second training data subset, the second air conditioning system temperature, and the temperature difference coefficient;

[0023] Based on the temperature difference coefficient and the proportional coefficient, the heat storage characteristics of the target building and the air-conditioning system are coupled to construct a room temperature prediction model.

[0024] In one embodiment, calculating the thermal resistance of the air conditioning system in the room temperature prediction model includes:

[0025] Filtering several groups of target data collected within a target historical time period from the second training data subset, each group of target data including chilled water flow, chilled water return temperature, chilled water outlet temperature, and indoor air temperature;

[0026] Based on the chilled water return temperature and the chilled water outlet temperature in each set of target data, a third air-conditioning system temperature is obtained;

[0027] Based on the chilled water flow rate, chilled water return temperature and chilled water outlet temperature in each set of target data, the heat transfer between the indoor air and the air conditioning system is obtained;

[0028] Obtaining a sensible heat transfer amount between the indoor air and the air conditioning system based on the heat transfer amount and the sensible heat ratio between the indoor air and the air conditioning system;

[0029] The thermal resistance of the air-conditioning system is obtained based on the third air-conditioning system temperature, the sensible heat transfer between the indoor air and the air-conditioning system, and the indoor air temperature in each set of target data.

[0030] In one embodiment, calculating the thermal resistance and heat capacity parameter combination in the room temperature prediction model includes:

[0031] Calculating a first wall temperature corresponding to the first training data subset based on the first training data subset, where the first wall temperature is the wall temperature corresponding to the initial acquisition moment in the first training data subset;

[0032] Substituting the first wall temperature and the first training data subset into a room temperature prediction model with a fixed thermal resistance of the air-conditioning system, using the combination of thermal resistance and heat capacity parameters to be determined in the room temperature prediction model as variables to be optimized, performing a first preset round of optimization using a genetic algorithm, and calculating a value of a first objective function corresponding to each round of optimization;

[0033] Sort the values ​​of the first objective function of all rounds, select a preset number of thermal resistance and heat capacity parameter combinations corresponding to the first objective function from small to large, and calculate the deviation value corresponding to each selected thermal resistance and heat capacity parameter combination;

[0034] Select a set of thermal resistance and heat capacity parameter combinations with the smallest deviation value as the final thermal resistance and heat capacity parameter combination;

[0035] The thermal resistance and heat capacity parameter combination includes: the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, the heat capacity of indoor air, the heat capacity of the air conditioning system and the heat capacity of the building envelope.

[0036] In one embodiment, calculating the first wall temperature corresponding to the first training data subset includes:

[0037] Using the indoor air temperature at the initial collection time in the first training data subset as a starting point and the outdoor air dry-bulb temperature at the initial collection time as an end point, generating a candidate wall temperature sequence according to a preset fixed step size, wherein the candidate wall temperature sequence includes a plurality of candidate wall temperatures;

[0038] Substituting each candidate wall temperature and the first training data subset into a room temperature prediction model for which the thermal resistance of the air conditioning system has been determined, and performing a second preset round of optimization on the thermal resistance and heat capacity parameter combinations to be determined in the room temperature prediction model using a genetic algorithm to obtain several sets of thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature;

[0039] Using a preset deviation function to evaluate several thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature, and selecting the thermal resistance and heat capacity parameter combination with the smallest deviation as the candidate thermal resistance and heat capacity parameter combination corresponding to the candidate wall temperature;

[0040] Selecting the time-series indoor air temperature and the time-series outdoor dry-bulb air temperature collected during the second target historical time period from the historically collected time-series indoor and outdoor environmental data;

[0041] Substituting the time-series indoor air temperature, the time-series outdoor dry-bulb air temperature, and the candidate thermal resistance and heat capacity parameter combination corresponding to each candidate wall temperature into a room temperature prediction model for which the thermal resistance of the air-conditioning system has been determined, and calculating to obtain several estimated wall temperatures;

[0042] The error between each candidate wall temperature and its corresponding estimated wall temperature is calculated, and the candidate wall temperature with the smallest error is selected as the first wall temperature.

[0043] In one embodiment, calculating an indoor heat gain schedule in a room temperature prediction model includes:

[0044] Substituting the second training data subset into a room temperature prediction model for a determined combination of thermal resistance and thermal resistance / capacity parameters of the air conditioning system, using the hourly indoor heat gain within a preset time range in the room temperature prediction model as a variable to be optimized, and iteratively adjusting its value through a genetic algorithm to minimize the second objective function;

[0045] When the preset end condition is reached, an indoor heating schedule consisting of the indoor heat gain per hour within the preset time range is obtained.

[0046] In a second aspect, an embodiment of the present invention further provides a device for quantifying the flexible adjustment potential of an air-conditioning system, comprising:

[0047] a data acquisition module, configured to continuously collect indoor and outdoor environmental data and air conditioning system operation data of a target building using a plurality of preset sensors, and to select a first training data subset and a second training data subset from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data;

[0048] a model building module, configured to build a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is used to predict changes in the temperature in the target building after the demand response is initiated;

[0049] a parameter determination module for sequentially calculating the thermal resistance of the air conditioning system, the combination of thermal resistance and heat capacity parameters, and the indoor heating time schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation;

[0050] The curve generation module is used to calculate the wall temperature at the time of demand response initiation when a demand response instruction is received, and substitute the wall temperature and the indoor and outdoor environmental data and air conditioning system operation data collected last time before the demand response initiation into the room temperature prediction model with fixed parameters to obtain the room temperature prediction curve;

[0051] The duration determination module is used to determine the maximum adjustable duration based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, and use it as the flexible adjustment potential of the air-conditioning system.

[0052] In a third aspect, an embodiment of the present invention further provides a terminal comprising: a memory, a processor, and an air-conditioning system flexible adjustment potential quantification program stored on the memory and executable on the processor, wherein the air-conditioning system flexible adjustment potential quantification program implements the steps of the air-conditioning system flexible adjustment potential quantification method as described above when executed by the processor.

[0053] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores a program for quantifying the flexible adjustment potential of an air-conditioning system. The program for quantifying the flexible adjustment potential of an air-conditioning system can be executed to implement the steps of the method for quantifying the flexible adjustment potential of an air-conditioning system as described above.

[0054] Beneficial effects of the present invention: The present invention collects indoor and outdoor environmental data and air-conditioning system operating data in the target building, and screens out a first training data subset and a second training data subset; constructs a room temperature prediction model; sequentially determines the air-conditioning system thermal resistance, thermal resistance and heat capacity parameter combination, and indoor heating time schedule in the room temperature prediction model, and uses the calculation results as model fixed parameters after each calculation; obtains a room temperature prediction curve; determines the maximum adjustable time, and uses it as the flexible adjustment potential of the air-conditioning system. The present invention takes the air-conditioning system operating data into consideration when constructing the room temperature prediction model, and uses the model to solve the maximum response time of the air-conditioning system during demand response, which can effectively improve the quantification accuracy of the flexible adjustment potential of the air-conditioning system and provide accurate data support for demand response strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of a preferred embodiment of the method for quantifying the flexible adjustment potential of an air-conditioning system in the present invention.

[0056] Figure 2 It is a structural diagram of the room temperature prediction model in the present invention.

[0057] Figure 3 It is a schematic diagram of the relationship between the actual value and the predicted value of the indoor air temperature in the present invention.

[0058] Figure 4It is a schematic diagram of the relationship between the actual value and the predicted value of the air-conditioning system temperature in the present invention.

[0059] Figure 5 It is a schematic representation of indoor heating time in the present invention.

[0060] Figure 6 It is a schematic diagram of the comparison results between the present invention, the two thermal resistances + two thermal capacitances model and the actual value.

[0061] Figure 7 It is a structural diagram of a preferred embodiment of the device for quantifying the flexible adjustment potential of an air-conditioning system in the present invention.

[0062] Figure 8 It is a block diagram of the terminal principle of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to illustrate the present invention and are not intended to limit the present invention.

[0064] Currently, the proportion of renewable energy generation, led by wind and photovoltaic power, continues to rise. However, the volatility and intermittent nature of renewable energy generation exacerbates the pressure on the power grid to balance supply and demand, leading to significant imbalances between power supply and demand during peak and off-peak periods. Demand response (DR) technology, by dynamically adjusting user-side loads to achieve a paradigm shift from "source follows load" to "load follows source," has become a key technological path for improving grid resilience and promoting the integration of renewable energy. Public buildings are major energy consumers, with air conditioning systems accounting for up to 70% of summer energy consumption. Because the building's inherent energy storage and the thermal inertia of air conditioning allow for short-term load reductions, minimizing the impact on human thermal comfort, air conditioning systems have become a highly promising flexible regulation resource. By accurately quantifying the flexible regulation potential of air conditioning systems, it is possible to maximize the grid's peak-shaving and off-peak-filling benefits while minimizing user-perceived disturbances, promoting the synergy between building energy conservation and emission reduction and grid stability.

[0065] Existing technologies typically construct thermal resistance and heat capacity models based on equivalent thermal network theory. These parameters describe the heat transfer process within a building system, simulating dynamic characteristics such as indoor and outdoor heat exchange, providing theoretical support for flexibility quantification. This model can be used to assess the flexible adjustment potential of air conditioning systems in demand response applications. However, because it fails to consider the system's inherent heat storage and its dynamic enthalpy and humidity characteristics, the quantification of this potential is inaccurate, making it difficult to provide accurate data support for demand response strategies.

[0066] In response to the above-mentioned deficiencies in the prior art, the present invention provides a method, device, terminal, and storage medium for quantifying the flexible adjustment potential of an air-conditioning system. The method comprises: continuously collecting indoor and outdoor environmental data and air-conditioning system operating data within a target building, and screening out a first training data subset and a second training data subset; constructing a room temperature prediction model; sequentially determining the air-conditioning system thermal resistance, thermal resistance and heat capacity parameter combination, and indoor heating time schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation; obtaining a room temperature prediction curve; and determining the maximum adjustable time as the flexible adjustment potential of the air-conditioning system. The present invention takes into account the air-conditioning system operating data when constructing the room temperature prediction model. By using this model to solve the maximum response time of the air-conditioning system during demand response, the accuracy of quantifying the flexible adjustment potential of the air-conditioning system can be effectively improved, and accurate data support can be provided for demand response strategies.

[0067] See Figure 1 The method for quantifying the flexible adjustment potential of an air-conditioning system according to an embodiment of the present invention comprises the following steps:

[0068] Step S100: Utilize a plurality of preset sensors to continuously collect indoor and outdoor environmental data and air conditioning system operation data of a target building, and select a first training data subset and a second training data subset from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data.

[0069] Specifically, indoor and outdoor environmental data include indoor air temperature and outdoor air dry-bulb temperature. Air conditioning system operation data include chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity. A first training data subset and a second training data subset are screened out from the historically collected time-series indoor and outdoor environmental data and the time-series air-conditioning system operation data, including: from the historically collected time-series indoor and outdoor environmental data and the time-series air-conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate and air humidity collected in a first historical time period are screened out as the first training data subset, wherein the target building is in a state without internal heat in the first historical time period, the air-conditioning compressor is turned off, and the water pump and the terminal air handling unit are running; from the historically collected time-series indoor and outdoor environmental data and the time-series air-conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate and air humidity collected in a second historical time period are screened out as the second training data subset, wherein the target building is in a state with internal heat and the air-conditioning system is operating stably in the second historical time period. The demand response involved in the present invention is that when the target building has internal heat, the air-conditioning compressor is shut down and the water pump and the terminal air handling unit are running. At this time, this method can fully release the remaining cooling capacity in the air-conditioning system pipeline and enhance the flexible adjustment capability of the air-conditioning system.

[0070] See Figure 1 The method for quantifying the flexible adjustment potential of an air-conditioning system according to an embodiment of the present invention comprises the following steps:

[0071] Step S200: constructing a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is used to predict changes in temperature in a target building after demand response is initiated.

[0072] Specifically, the coefficients required to construct the room temperature prediction model are gradually solved using the first training data subset and the second training data subset. Then, these coefficients are used to couple the heat storage characteristics of the target building and the air-conditioning system to construct a room temperature prediction model that characterizes the heat exchange process of the target building's envelope structure, air-conditioning system, and indoor air.

[0073] In one implementation, constructing a room temperature prediction model using the first training data subset and the second training data subset includes:

[0074] Filtering the chilled water outlet temperature, chilled water return temperature, and indoor air temperature at a target time from the first training data subset, wherein the target time is the time when the air conditioning system temperature reaches the dew point temperature when the target building is in a state of no internal heat, the air conditioning compressor is turned off, and the water pump and the terminal air handling unit are running;

[0075] Obtaining a first air-conditioning system temperature based on a chilled water outlet temperature and a chilled water return temperature at a target time, and obtaining a temperature difference coefficient based on the first air-conditioning system temperature and an indoor air temperature at the target time;

[0076] obtaining a return air enthalpy value based on the return air temperature and indoor humidity of the second training data subset, and obtaining a supply air enthalpy value based on the supply air temperature and indoor humidity of the second training data subset;

[0077] Obtaining a preset air specific heat capacity, and obtaining a sensible heat ratio based on the air specific heat capacity, the return air temperature of the second training data subset, the supply air temperature of the second training data subset, the return air enthalpy value, and the supply air enthalpy value;

[0078] Obtaining a latent heat and sensible heat ratio based on the sensible heat ratio, and obtaining a second air-conditioning system temperature based on the chilled water outlet temperature and the chilled water return temperature of the second training data subset;

[0079] Obtaining a proportionality coefficient based on the ratio of latent heat to sensible heat, the indoor air temperature of the second training data subset, the second air conditioning system temperature, and the temperature difference coefficient;

[0080] Based on the temperature difference coefficient and the proportional coefficient, the heat storage characteristics of the target building and the air-conditioning system are coupled to construct a room temperature prediction model.

[0081] Specifically, when the air conditioning compressor is turned off and the water pump and terminal air handling unit are running, the evaporator and refrigerant circulation stop, the dehumidification function is interrupted, and the moisture cannot continue to condense, and the latent heat accumulation is reduced. Due to the thermal inertia of the system and the building, the indoor air temperature rises slower than the humidity rises, and the dew point temperature of the indoor air gradually approaches the system chilled water temperature, resulting in the inability of water vapor in the air to condense and the air conditioning system to dehumidify. Calculate the average of the chilled water outlet temperature and the chilled water return temperature at the target time to obtain the first air conditioning system temperature The formula for calculating the temperature difference coefficient is as follows: , where is the indoor air temperature at the target moment.

[0082] In one implementation, the indoor air temperature at the target time is 23.977°C, the first air conditioning system temperature is 19.518°C, and the calculated temperature difference coefficient is The value is 4.459.

[0083] The formula for calculating the sensible heat fraction is as follows: , where is the sensible heat ratio, The preset specific heat capacity of air is 1.007 , is the average return air temperature in the second training data subset, is the average value of the supply air temperature in the second training data subset, is the return air enthalpy, is the supply air enthalpy value. When the sensible heat ratio is obtained, the latent heat ratio is calculated based on the sensible heat ratio, and then the latent heat and sensible heat ratio are obtained based on the sensible heat ratio and the latent heat ratio. The formula for calculating the latent heat ratio is as follows: The formula for calculating the ratio of latent heat to sensible heat in the target building with internal heat is as follows: .

[0084] In one implementation, the ratio of latent heat to sensible heat is 0.565.

[0085] Calculate the scale factor The formula is as follows: Where, is the indoor air temperature of the target building when it is in an internal heat state and the air conditioning system is in a stable operation state, which is obtained by averaging all the indoor air temperatures in the second training data subset. The air conditioning system temperature is calculated by averaging all chilled water outlet and return temperatures in the second training data subset when the target building is experiencing internal heat and the air conditioning system is operating stably. Once the temperature difference coefficient and proportionality coefficient are determined, a room temperature prediction model can be constructed by coupling the thermal storage characteristics of the target building and the air conditioning system. The mathematical expression for the room temperature prediction model is as follows:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] Where, is the ratio of latent heat to sensible heat at any point in time, is the indoor air temperature, is the air conditioning system temperature, is the thermal resistance of the air conditioning system, It represents the sensible heat transfer between indoor air and air conditioning system. Represents the latent heat transfer between indoor air and the air conditioning system, Indicates the amount of heat transferred between indoor air and the air conditioning system. represents the amount of heat transfer between the building envelope and the indoor air, represents the wall temperature, It represents the resistance to heat transfer between the building envelope (i.e. the wall) and the indoor air. represents the heat transfer between the outdoor air and the external envelope, represents the outdoor air dry-bulb temperature, It represents the resistance to heat transfer between the building envelope (i.e. the wall) and the outdoor air. represents the temperature of the air conditioning system at time i+1, Indicates the cooling capacity provided by the air conditioning system. Represents the time step, which is set to one minute in the present invention. represents the heat capacity of the air conditioning system, represents the temperature of the air conditioning system at time i, represents the temperature of the indoor air at time i+1, Indicates indoor heat gain. represents the indoor air heat capacity, represents the temperature of the indoor air at time i, represents the wall temperature at time i+1, represents the thermal capacity of the building envelope, represents the wall temperature at time i. H is a time variable representing the hour of the day and is used as function input. h is a subscript representing a specific hour. represents the indoor heat gain in h hours, is a mapping model that represents the relationship between indoor heat gain (i.e., internal heat gain) and the number of hours in a day. The structural diagram of the room temperature prediction model is shown in the figure below. Figure 2 shown.

[0097] Currently, the assessment of the flexible regulation potential of building air conditioning systems relies primarily on three types of room temperature prediction models: white-box models based on physical mechanisms (such as EnergyPlus), data-driven black-box models (such as Long Short-Term Memory (LSTM)), and gray-box models, exemplified by the 2Resistance+2Capacitance (2R+2C) model. White-box models based on physical mechanisms require complete building geometry and have a long modeling cycle, making them incapable of quickly responding to regulation needs. While data-driven black-box models avoid complex modeling processes, they suffer from poor adaptability to operating conditions, weak interpretability, and a reliance on high computing power, making them prone to significant assessment bias during sudden demand response events. The gray-box model simplifies thermodynamic representation by using thermal resistance (R) and heat capacity (C) parameters, achieving a balance between computational efficiency and physical interpretability. However, it has the following limitations: First, the 2Resistance+2Capacitance (2R+2C) model only considers the thermal inertia of the building envelope and fails to establish a mathematical representation of the thermal resistance and heat capacity of the air-conditioning system. This makes it impossible to quantify the storage and transfer of cold energy in the system, hindering the analysis of the mechanism affecting room temperature. Second, the 2Resistance+2Capacitance (2R+2C) model does not consider the dynamic enthalpy and humidity characteristics of the air-conditioning system and ignores the influence of the system's latent heat on room temperature, resulting in limited model prediction accuracy. Third, the 2Resistance+2Capacitance (2R+2C) model uses a single-stage global optimization algorithm to jointly train the thermal resistance and heat capacity (RC) parameters, resulting in strong coupling between the model parameters, making it difficult to accurately fit the thermal resistance and heat capacity (RC) parameters, thereby causing inaccurate flexible quantification. These limitations restrict the precise quantification of the flexible regulation potential of building air conditioning systems, leading to issues such as insufficient matching between virtual power plant dispatch instructions and actual response times. This paper constructs a room temperature prediction model that incorporates dynamic adjustment of the air conditioning system's enthalpy and humidity. This model, known as the 3R+3C (3Resistance+3Capacitance) model, incorporates the air conditioning system's heat capacity and the thermal resistance of heat transfer between the air conditioning system and the indoor air temperature, building upon the 2R+3C model. This model, referred to as the 3R+3C model, allows for more accurate calculation of the air conditioning system's maximum response time.

[0098] By constructing a room temperature prediction model, the present invention can predict the indoor air temperature after demand response, thereby obtaining the maximum adjustable time, that is, the time it takes for the indoor air temperature of the target building to reach a preset temperature threshold after demand response. This maximum adjustable time, as the flexible adjustment potential of the air conditioning system, can be fed back to the power grid for better coordination and scheduling of power resources. Demand response in this context refers to the air conditioning compressor being turned off and the water pump and terminal air handling unit being operated. Because the building has a certain heat storage capacity, even in this case, it will take a certain amount of time for the indoor air temperature to reach the preset temperature threshold (i.e., 26 degrees Celsius). By calculating the time required to reach 26 degrees Celsius, i.e., the maximum response time, it can be fed back to the power grid as a basis for power regulation and scheduling.

[0099] See Figure 1 The method for quantifying the flexible adjustment potential of an air-conditioning system according to an embodiment of the present invention further includes the following steps:

[0100] Step S300: sequentially calculate the thermal resistance of the air conditioning system, the thermal resistance and heat capacity parameter combination, and the indoor heat gain schedule in the room temperature prediction model, and use the calculation results as fixed parameters of the model after each calculation.

[0101] Specifically, the present invention sequentially determines the unknown parameters in the room temperature prediction model, fixing each unknown parameter as a fixed parameter in the model. This stepwise solution reduces computational complexity. The present invention employs a two-step identification framework. In the first step, the thermal resistance of the air conditioning system is fixed. The thermal resistance and heat capacity parameter combinations are trained using a first training data subset (i.e., without internal heat data). In the second step, the indoor heating time parameters are optimized and fitted using a second training data subset (i.e., with internal heat data), achieving efficient and accurate identification of model parameters.

[0102] In one implementation, calculating the thermal resistance of the air conditioning system in the room temperature prediction model includes:

[0103] Filtering several groups of target data collected within a target historical time period from the second training data subset, each group of target data including chilled water flow, chilled water return temperature, chilled water outlet temperature, and indoor air temperature;

[0104] Based on the chilled water return temperature and the chilled water outlet temperature in each set of target data, a third air-conditioning system temperature is obtained;

[0105] Based on the chilled water flow rate, chilled water return temperature and chilled water outlet temperature in each set of target data, the heat transfer between the indoor air and the air conditioning system is obtained;

[0106] Obtaining a sensible heat transfer amount between the indoor air and the air conditioning system based on the heat transfer amount and the sensible heat ratio between the indoor air and the air conditioning system;

[0107] The thermal resistance of the air-conditioning system is obtained based on the third air-conditioning system temperature, the sensible heat transfer between the indoor air and the air-conditioning system, and the indoor air temperature in each set of target data.

[0108] Specifically, after building the room temperature prediction model, it is necessary to solve the unknown parameters in the model one by one. The first thing solved in this invention is the thermal resistance of the air conditioning system. Under normal conditions of large central air-conditioning systems, the chilled water flow rate and inlet and outlet water temperatures fluctuate slightly, and the system average temperature is approximately constant. Therefore, the cooling load stored in the system is relatively stable and easy to calculate based on the chilled water flow rate and inlet and outlet water temperature difference. When the air-conditioning system is in an internal heat state for the target building, The cooling load stored in the system is calculated by selecting any hour within the data collection time range of the second training data subset as the target historical time period, and obtaining the chilled water flow rate, chilled water return temperature, chilled water outlet temperature and indoor air temperature in this time period as the target data. Then the heat transfer between the indoor air and the air conditioning system is calculated. , the calculation formula is as follows: , where Represents the specific heat capacity of water, which is 4.18 ; Indicates the average chilled water flow rate during the target historical time period; Indicates the average value of the chilled water return temperature during the target historical time period; The sensible heat transfer between indoor air and the air conditioning system is calculated using the following formula: , Among them, sensible heat accounts for The value has been obtained when building the room temperature prediction model and can be used directly here. Then the thermal resistance of the air conditioning system is calculated using the following formula: Where, Indicates averaging the data of the target historical time period, and substituting the average indoor air temperature in the target historical time period into , substitute the average value of the air temperature of the third air conditioning system during the target historical period into The third air conditioning system temperature is obtained by averaging the chilled water return temperature and chilled water outlet temperature in each set of target data. After that, it is used as a fixed parameter of the room temperature prediction model and subsequently used to solve other unknown parameters.

[0109] In one implementation, calculating the thermal resistance and heat capacity parameter combination in the room temperature prediction model includes:

[0110] Calculating a first wall temperature corresponding to the first training data subset based on the first training data subset, where the first wall temperature is the wall temperature corresponding to the initial acquisition moment in the first training data subset;

[0111] Substituting the first wall temperature and the first training data subset into a room temperature prediction model with a fixed thermal resistance of the air-conditioning system, using the combination of thermal resistance and heat capacity parameters to be determined in the room temperature prediction model as variables to be optimized, performing a first preset round of optimization using a genetic algorithm, and calculating a value of a first objective function corresponding to each round of optimization;

[0112] Sort the values ​​of the first objective function of all rounds, select a preset number of thermal resistance and heat capacity parameter combinations corresponding to the first objective function from small to large, and calculate the deviation value corresponding to each selected thermal resistance and heat capacity parameter combination;

[0113] Select a set of thermal resistance and heat capacity parameter combinations with the smallest deviation value as the final thermal resistance and heat capacity parameter combination;

[0114] The thermal resistance and heat capacity parameter combination includes: the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, the heat capacity of indoor air, the heat capacity of the air conditioning system and the heat capacity of the building envelope.

[0115] Specifically, when the thermal resistance of the air conditioning system is solved Then, it is used as a fixed parameter of the model. On this basis, in order to calculate the thermal resistance and heat capacity parameter combination, it is necessary to first calculate the wall temperature corresponding to the initial acquisition time in the first training data subset (i.e., the first wall temperature), and then substitute the first wall temperature into the room temperature prediction model to determine the final thermal resistance and heat capacity parameter combination. The thermal resistance and heat capacity parameter combination of the present invention includes the resistance value of heat transfer between the building envelope and the indoor air. , the resistance to heat transfer between the building envelope and the outdoor air , indoor air heat capacity , air conditioning system heat capacity and building envelope thermal capacity . The genetic algorithm uses the above five parameters as chromosomes in the optimization process, randomly initializes them within a preset order of magnitude, and generates an initial population. The genetic algorithm then optimizes the population through operations such as selection, crossover, and mutation. After substituting the first training data subset into the room temperature prediction model, the genetic algorithm is used to optimize the first prediction round, and the first preset round can be 5000 times. When the number of iterations reaches the preset number, the genetic algorithm optimization is stopped. Each round of optimization calculates the value of the corresponding first objective function. The expression of the first objective function is:

[0116] ;

[0117] Where, is the weighted sum of the indoor air temperature and the air conditioning system temperature obtained from the first training data subset, that is, the actual value. is the weighted sum of the predicted values ​​of the indoor air temperature and the air conditioning system temperature obtained by the room temperature prediction model, and n is the number of time steps for collecting the first data training subset. The calculation formula is as follows: . In the formula, r represents the ratio of the indoor air temperature variation range to the air conditioning system temperature fluctuation range, and the value is between 0 and 1. The indoor air temperature variation range is the difference between the average indoor temperature when the air conditioner is stopped and the average temperature when the air conditioner is running stably, and the air conditioning system temperature fluctuation range is the difference between the average indoor temperature when the air conditioner is stopped and the average temperature when the air conditioner is running stably. By calculating the historical data, the ratio r of the indoor air temperature variation range to the air conditioning system temperature fluctuation range can be obtained. After 5000 iterations, due to the uncertainty of the genetic algorithm, 500 sets of thermal resistance and heat capacity parameter combinations with the smallest first objective function value are selected, and then the deviation values ​​of these 500 sets of thermal resistance and heat capacity parameters are calculated.

[0118] The calculation formula of the deviation value is as follows:

[0119] ;

[0120] Where, Indicates the deviation value of the thermal resistance and heat capacity parameter combination, Represents the average value of the resistance to heat transfer between the building envelope and the indoor air in all thermal resistance and heat capacity combinations. Represents the average value of the resistance to heat transfer between the building envelope and the outdoor air in all thermal resistance and heat capacity combinations. Represents the average value of the heat capacity of the air conditioning system in all thermal resistance and heat capacity combinations, Represents the average value of indoor air heat capacity in all thermal resistance and heat capacity combinations, Represents the average thermal capacity of the building envelope in all thermal resistance and thermal capacity combinations.

[0121] The mathematical expression for selecting the thermal resistance and heat capacity parameter combination with the smallest deviation value is as follows:

[0122] ;

[0123] Where, is the final thermal resistance and heat capacity parameter combination of the room temperature prediction model, It is the thermal resistance and heat capacity parameter combination with the smallest deviation.

[0124] In one implementation, .

[0125] In one implementation, the specific value of the thermal resistance and thermal capacitance parameter combination finally determined is: , , , , .

[0126] After the thermal resistance and heat capacity parameters are determined through training, they are fixed in the room temperature prediction model as model fixed parameters.

[0127] After the first preset round of training, the relationship between the actual value and the predicted value of indoor air temperature is shown in the figure below: Figure 3 As shown in the figure, the relationship between the actual value and the predicted value of the air conditioning system temperature is shown in the figure. Figure 4 The root mean square error (RMSE) between the actual and predicted indoor air temperature is 0.036°C, and the root mean square error (RMSE) between the actual and predicted air conditioning system temperature is 0.416°C.

[0128] In one implementation, calculating the first wall temperature corresponding to the first training data subset based on the first training data subset includes:

[0129] Using the indoor air temperature at the initial collection time in the first training data subset as a starting point and the outdoor air dry-bulb temperature at the initial collection time as an end point, generating a candidate wall temperature sequence according to a preset fixed step size, wherein the candidate wall temperature sequence includes a plurality of candidate wall temperatures;

[0130] Substituting each candidate wall temperature and the first training data subset into a room temperature prediction model for which the thermal resistance of the air conditioning system has been determined, and performing a second preset round of optimization on the thermal resistance and heat capacity parameter combinations to be determined in the room temperature prediction model using a genetic algorithm to obtain several sets of thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature;

[0131] Using a preset deviation function to evaluate several thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature, and selecting the thermal resistance and heat capacity parameter combination with the smallest deviation as the candidate thermal resistance and heat capacity parameter combination corresponding to the candidate wall temperature;

[0132] Selecting the time-series indoor air temperature and the time-series outdoor dry-bulb air temperature collected during the second target historical time period from the historically collected time-series indoor and outdoor environmental data;

[0133] Substituting the time-series indoor air temperature, the time-series outdoor dry-bulb air temperature, and the candidate thermal resistance and heat capacity parameter combination corresponding to each candidate wall temperature into a room temperature prediction model for which the thermal resistance of the air-conditioning system has been determined, and calculating to obtain several estimated wall temperatures;

[0134] The difference between each candidate wall temperature and its corresponding estimated wall temperature is calculated, and the candidate wall temperature with the smallest difference is selected as the first wall temperature.

[0135] Specifically, while both indoor air temperature and outdoor air dry-bulb temperature can be collected via sensors, wall temperature is difficult to measure. The setting of wall temperature in the model significantly impacts the subsequent identification of other unknown parameters. Therefore, the present invention utilizes the first training data subset for multiple rounds of iterative training to optimize the calculation of the wall temperature corresponding to the initial collection time in the first training data subset.

[0136] Using the indoor air temperature at the initial collection time in the first training data subset as the starting point and the outdoor air dry-bulb temperature at the initial collection time as the ending point, an exhaustive method is used to select values ​​in increments of 0.1 degrees Celsius to obtain a candidate wall temperature sequence. For example, if the first training data subset collects indoor air temperature and outdoor air dry-bulb temperature starting at 21:00 on August 1, 2024, the indoor air temperature is 26.5 degrees Celsius, and the outdoor air dry-bulb temperature is 27.5 degrees Celsius, then a candidate wall temperature sequence of the form (26.5, 26.6, 26.7, 26.8, 26.9, 27.0, 27.1, 27.2, 27.3, 27.4, 27.5) is generated.

[0137] Substitute each candidate wall temperature and the first training data subset into the room temperature prediction model. The thermal resistance and heat capacity parameter combinations to be determined in the room temperature prediction model (i.e., the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, the heat capacity of indoor air, the heat capacity of the air conditioning system, and the heat capacity of the building envelope) are used as variables to be optimized. A genetic algorithm is used to perform a second preset round of optimization, which can be 100. After the above optimization, several groups of thermal resistance and heat capacity combinations are obtained. These combinations are then evaluated using a preset deviation function, and the thermal resistance and heat capacity parameter combination with the smallest deviation is selected as the candidate thermal resistance and heat capacity parameter combination. The formula for the deviation function is: ;

[0138] The mathematical expression for selecting the thermal resistance and heat capacity parameter combination with the smallest deviation as the candidate thermal resistance and heat capacity parameter combination is:

[0139] , where is the candidate thermal resistance and heat capacity parameter combination corresponding to a candidate wall temperature, This is the thermal resistance and heat capacity parameter combination with the smallest deviation. Please note that the thermal resistance and heat capacity parameter combination with the smallest deviation selected this time is only used for the subsequent calculation of the first wall temperature and is not the final thermal resistance and heat capacity parameter combination of the temperature prediction model.

[0140] The second target historical time period is the period starting from the initial collection time of the first training data subset and going back 30 hours. For example, if the initial collection time of the first training data set is 21:00 on August 1, 2024, the target historical time period is from 15:00 on July 31, 2024 to 21:00 on August 1, 2024.

[0141] At this point, the first indoor air temperature in the time series and the first outdoor dry-bulb air temperature in the time series are averaged to obtain the wall temperature corresponding to the initial collection time in the second target historical time period. Then, the wall temperature corresponding to the initial collection time in the second target historical time period, the first indoor air temperature, and the candidate thermal resistance and heat capacity parameter combination corresponding to each candidate wall temperature are substituted into the wall temperature calculation formula in the temperature prediction model for the determined air conditioning system thermal resistance and thermal resistance and heat capacity parameter combination. The wall temperature calculation formula is as follows:

[0142] ;

[0143] ;

[0144] ;

[0145] Specifically, when i=0, substitute the wall temperature corresponding to the initial acquisition time in the second target historical time period into and , substitute the first indoor air temperature in the time series into , substitute the outdoor dry-bulb air temperature that ranks first in the time series into , the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, and the heat capacity of the building envelope in the candidate thermal resistance and heat capacity parameter combination are also substituted into the formula , the wall temperature corresponding to the next acquisition moment can be obtained by calculation In the next calculation, substitute the second indoor air temperature in the time series into , substitute the second outdoor dry-bulb air temperature in the time series into ,Will As the wall temperature at the second collection moment in the second target historical time period, Maintaining this constant, the calculation is repeated to obtain the wall temperature at the third data collection moment in the second target historical time period. This calculation process is repeated until the estimated wall temperature at the last data collection moment in the second target historical time period (i.e., the initial collection moment of the first training dataset) is determined. It will be appreciated that the above calculation process uses a fixed candidate thermal resistance and heat capacity parameter combination corresponding to a candidate wall temperature. The above calculation is performed for each candidate thermal resistance and heat capacity parameter combination corresponding to each candidate wall temperature. After calculating all estimated wall temperatures, the difference between each candidate wall temperature and its corresponding estimated wall temperature is calculated, and the candidate wall temperature with the smallest difference is selected as the first wall temperature.

[0146] In one implementation, calculating an indoor heat gain schedule in a room temperature prediction model includes:

[0147] Substituting the second training data subset into a room temperature prediction model for a determined combination of thermal resistance and thermal resistance / capacity parameters of the air conditioning system, using the hourly indoor heat gain within a preset time range in the room temperature prediction model as a variable to be optimized, and iteratively adjusting its value through a genetic algorithm to minimize the second objective function;

[0148] When the preset end condition is reached, an indoor heating schedule consisting of the indoor heat gain per hour within the preset time range is obtained.

[0149] Specifically, when the thermal resistance and thermal capacity combination of the air conditioning system are determined, these two are used as fixed parameters of the model to solve the indoor heating schedule. , that is, the indoor heat gain corresponding to H in each hour of the day. The expression of the second objective function is: Where, represents the indoor heat gain at time i, is the indoor heat gain at time i predicted by the room temperature prediction model. Genetic algorithm is used to randomly generate multiple groups of , then calculate the second objective function of each set of solutions, retain the result with smaller error, and generate new solutions through combination and random perturbation until the preset number of iterations is reached or the second objective function converges to the preset value, and select the solution with the smallest second objective function value. The indoor heating timetable generated by the present invention is as follows: Figure 5 shown.

[0150] While considering the cooling capacity of buildings and air conditioning systems, the present invention focuses on the close relationship between the cooling capacity of the system and the changes in air temperature and humidity. The proposed model uses a two-objective penalty function to balance the relationship between the air conditioning system temperature and the indoor air temperature, thereby obtaining a stable combination of thermal resistance and heat capacity parameters and an indoor heating schedule. The room temperature prediction model of the present invention is used to predict the changes in the indoor air temperature of the target building during demand response (i.e., when the air conditioning compressor is turned off and the water pump and terminal air handling unit are operating in the internal heating state). By constructing an experiment with two days of different outdoor temperature scenarios, the root mean square error (RMSE) of the indoor air temperature was 0.061°C and 0.174°C, respectively, and the root mean square error (RMSE) of the air conditioning system temperature was 0.347°C and 0.697°C, respectively. Based on the prediction results, the maximum allowable adjustment time was further quantified to achieve an accurate assessment of the flexible adjustment potential of the air conditioning system, providing a basis for optimizing the demand response strategy.

[0151] See Figure 1 The method for quantifying the flexible adjustment potential of an air-conditioning system according to an embodiment of the present invention further includes the following steps:

[0152] Step S400: When a demand response instruction is received, the wall temperature at the time of demand response initiation is calculated, and the wall temperature and the indoor and outdoor environmental data and air-conditioning system operation data collected last time before the demand response initiation are substituted into a room temperature prediction model with fixed parameters to obtain a room temperature prediction curve.

[0153] Specifically, when a demand response instruction is received from the power grid, it typically includes the demand response initiation time. The wall temperature at the demand response initiation time must be calculated to determine how long it will take for the target building to reach the preset temperature threshold when the demand response is actually initiated. Indoor and outdoor environmental data and air conditioning system operating data collected 30 hours prior to the demand response initiation time are obtained as relevant data for the calculation. This data is then substituted into a room temperature prediction model that has fixed parameters for the air conditioning system's thermal resistance, thermal resistance and heat capacity parameter combination, and indoor heat gain schedule. The temperature inference formula in the room temperature prediction model can be used to calculate the wall temperature at the demand response initiation time. This data, along with the indoor and outdoor environmental data and air conditioning system operating data collected last time before the demand response initiation time, are then substituted into a room temperature prediction model that has fixed parameters (i.e., the air conditioning system's thermal resistance, thermal resistance and heat capacity parameter combination, and indoor heat gain schedule) to produce a room temperature prediction curve. This invention innovatively proposes a wall temperature calculation method that can improve the accuracy of parameter magnitude estimation.

[0154] See Figure 1 The method for quantifying the flexible adjustment potential of an air-conditioning system according to an embodiment of the present invention further includes the following steps:

[0155] Step S500: Based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, a maximum adjustable time is determined and used as the flexible adjustment potential of the air-conditioning system.

[0156] Specifically, determine the arrival time when the room temperature prediction curve reaches the preset temperature threshold for the first time , subtract the demand response start time from the arrival time , get the maximum adjustable time This duration reflects the flexible adjustment potential of the air conditioning system. This maximum adjustable duration can be reported to the power grid to facilitate power allocation decisions.

[0157] To verify the effectiveness of the room temperature prediction model of the present invention, data from a demand response event in which the compressor was shut down while the water pump and terminal air handling unit were kept running was used for accuracy verification. The root mean square error (RMS) between the predicted room temperature and the true value calculated by the room temperature prediction model of the present invention was 0.174°C, and the RMS error between the predicted air conditioning system temperature and the true value was 0.697°C, demonstrating the model's excellent prediction performance.

[0158] In addition, the room temperature prediction model of the present invention was compared with the prior art 2Resistance+2Capacitance (2R+2C) model, which only considers the building thermal process. The training and validation of the 2Resistance+2Capacitance model used the same data and genetic algorithm operating parameters as the present invention. The room temperature prediction model of the present invention and the 2Resistance+2Capacitance model were validated using the same validation set. A schematic diagram comparing the room temperature prediction model of the present invention, the 2Resistance+2Capacitance model (i.e., the 2R+2C model), and the actual values ​​is shown in the attached figure. Figure 6 The calculated root mean square error between the indoor air temperature prediction value and the true value of the two-thermal resistance + two-heat capacity model is 0.453°C, while the root mean square error between the indoor air temperature prediction value and the true value of the room temperature prediction model of the present invention is 0.174°C, a reduction of 0.279°C. This shows that the room temperature prediction model of the present invention effectively captures the heat transfer process of the air conditioning system during demand response events and its impact on indoor temperature and humidity, thereby significantly improving the model's room temperature prediction accuracy.

[0159] In one implementation, the room temperature prediction model of the present invention calculates a flexible regulation potential of 76 minutes, while the actual regulation potential calculated based on data collected by the collector is 80 minutes. In comparison, the flexible regulation potential calculated using the 2Resistance+2Capacitance (2R+2C) model is 63 minutes. This demonstrates that the flexibility quantification accuracy of the temperature prediction results based on the present invention is improved by over 70% compared to traditional methods. This improved method uses physical modeling to accurately assess the flexible regulation capability of building air conditioning systems during demand response, providing strong support for the optimized scheduling of building virtual power plants.

[0160] In summary, this paper proposes a room temperature prediction (3Resistance+3Capacitance, 3R+3C) model. This model characterizes the cooling capacity of the air conditioning system and its heat exchange path with the indoor air by incorporating the heat capacity and thermal resistance of the air conditioning system. Incorporating the dynamic coupling mechanism of sensible and latent heat, the system coefficients α and β quantify the latent heat contribution, optimizing the mathematical representation of the heat transfer process. Compared with the existing two-resistance+two-capacity model, which only considers the thermal storage characteristics of the building, this model significantly improves the accuracy of room temperature prediction during demand response events, reducing the root mean square error (RMSE) by over 60%. This paper also proposes a method for directly calculating and fixing the thermal resistance of the air conditioning system based on the chilled water flow rate and the difference between the return and outlet temperatures of the chilled water. Unlike the traditional two-resistance+two-capacity model, which involves full parameter optimization, this method simplifies parameter identification by fixing the thermal resistance of the air conditioning system, shortens model training time, and avoids error accumulation caused by parameter coupling, thereby improving model robustness and prediction accuracy. To solve the problem of unmeasurable wall temperature, this paper proposes an initial wall temperature optimization method that combines exhaustive method with historical data. The initial temperature sequence is generated by exhaustive method, and the deviation function ( ) to screen the optimal parameter combination and calculate the initial wall temperature based on historical data. This method effectively addresses the prediction inaccuracy problem of traditional models caused by initial wall temperature setting deviations. Experimental verification shows that the prediction error can be controlled within 0.2°C, significantly improving the model's prediction accuracy and stability. This paper proposes a flexible quantification method based on accurate room temperature prediction. By integrating the heat storage and enthalpy-humidity dynamic characteristics of the air conditioning system, it improves room temperature prediction accuracy and thus accurately assesses the flexible adjustment potential of the building's air conditioning system. Traditional flexible quantification methods primarily rely on simulation software or simplified thermal resistance and heat capacitance (RC) models, which fail to accurately reflect the thermal inertia of the air conditioning system and its dynamic impact on room temperature, resulting in large errors in the assessment of flexible adjustment duration. This method optimizes the heat transfer path of the RC model to accurately predict room temperature changes during the demand response period. Combined with user thermal comfort thresholds, it calculates the maximum adjustable potential, providing a more accurate basis for flexible quantification. Experimental verification shows that this method improves the flexibility quantification accuracy by over 70% compared to traditional methods, providing a refined control basis for building virtual power plant scheduling and grid demand response optimization.

[0161] In one embodiment, if Figure 7 As shown, based on the above-mentioned method for quantifying the flexible adjustment potential of an air-conditioning system, the present invention also provides a device for quantifying the flexible adjustment potential of an air-conditioning system, comprising:

[0162] The data acquisition module 100 is configured to continuously collect indoor and outdoor environmental data and air conditioning system operation data of the target building using a plurality of preset sensors, and to select a first training data subset and a second training data subset from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data;

[0163] A model building module 200 is configured to build a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is configured to predict changes in the temperature in the target building after the demand response is initiated;

[0164] a parameter determination module 300 for sequentially calculating the thermal resistance of the air conditioning system, the combination of thermal resistance and heat capacity parameters, and the indoor heat gain schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation;

[0165] Curve generation module 400 is used to calculate the wall temperature at the time of demand response initiation upon receiving a demand response instruction, and substitute the wall temperature and the indoor and outdoor environmental data and air conditioning system operation data collected last time before the demand response initiation into a room temperature prediction model with fixed parameters to obtain a room temperature prediction curve;

[0166] The duration determination module 500 is used to determine the maximum adjustable duration based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, and use it as the flexible adjustment potential of the air-conditioning system.

[0167] It should be noted that the above explanation of the embodiment of the method for quantifying the flexible adjustment potential of an air-conditioning system is also applicable to the device for quantifying the flexible adjustment potential of an air-conditioning system of this embodiment, and will not be repeated here.

[0168] Based on the above embodiment, the present invention further provides a terminal, whose principle block diagram can be shown as follows: Figure 8 As shown. The terminal includes a processor, a memory, a network interface and a display screen connected via a device bus. The processor of the terminal is used to provide computing and control capabilities. The memory of the terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating device and an air-conditioning system flexible adjustment potential quantification program. The internal memory provides an environment for the operation of the operating device and the air-conditioning system flexible adjustment potential quantification program in the non-volatile storage medium. The network interface of the terminal is used to communicate with an external terminal via a network connection. When the air-conditioning system flexible adjustment potential quantification program is executed by the processor, the steps of any one of the above-mentioned air-conditioning system flexible adjustment potential quantification methods are implemented. The display screen of the terminal can be a liquid crystal display or an electronic ink display.

[0169] Those skilled in the art will understand that Figure 8The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0170] In one embodiment, a terminal is provided, which includes a memory, a processor, and an air-conditioning system flexible adjustment potential quantification program stored in the memory and executable on the processor. When the air-conditioning system flexible adjustment potential quantification program is executed by the processor, the steps of any one of the air-conditioning system flexible adjustment potential quantification methods provided in the embodiments of the present invention are implemented.

[0171] An embodiment of the present invention also provides a computer-readable storage medium, on which a program for quantifying the flexible adjustment potential of an air-conditioning system is stored. When the program for quantifying the flexible adjustment potential of an air-conditioning system is executed by a processor, the steps of any one of the methods for quantifying the flexible adjustment potential of an air-conditioning system provided in an embodiment of the present invention are implemented.

[0172] It should be understood that the sequence numbers of the steps in the above embodiments do not imply a specific order of execution; the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0173] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0174] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0175] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0176] In the embodiments provided by the present invention, it should be understood that the disclosed apparatus / terminal device and method can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For example, the division of the modules or units described above is merely a logical functional division. In actual implementation, other division methods may be used. For example, multiple units or components may be combined or integrated into another device, or some features may be omitted or not implemented.

[0177] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A method for quantifying the flexible regulation potential of an air conditioning system, characterized in that: The method comprises: Utilizing a plurality of preset sensors to continuously collect indoor and outdoor environmental data and air conditioning system operation data of the target building, a first training data subset and a second training data subset are selected from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data; Constructing a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is used to predict a change in the temperature in the target building after a demand response is initiated, wherein the demand response is that the target building has internal heat and the air conditioning compressor is turned off and the water pump and the terminal air handling unit are running; Sequentially calculating the air conditioning system thermal resistance, the thermal resistance and heat capacity parameter combination, and the indoor heat gain schedule in the room temperature prediction model, and using the calculation results as fixed parameters of the model after each calculation, wherein the thermal resistance and heat capacity parameter combination includes: the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, the indoor air heat capacity, the air conditioning system heat capacity, and the building envelope heat capacity; When a demand response instruction is received, the wall temperature at the time of demand response initiation is calculated, and the wall temperature and the indoor and outdoor environment data and air conditioning system operation data collected last time before the demand response initiation are substituted into the room temperature prediction model with fixed parameters to obtain the room temperature prediction curve; Based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, a maximum adjustable time is determined and used as the flexible adjustment potential of the air conditioning system; The indoor and outdoor environmental data include indoor air temperature and outdoor air dry-bulb temperature, and the air conditioning system operation data include chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity. A first training data subset and a second training data subset are screened from the historically collected time-series indoor and outdoor environmental data and the time-series air conditioning system operation data, including: From the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during a first historical time period are selected as a first training data subset. During the first historical time period, the target building is in a zero-heat state, the air conditioning compressor is turned off, and the water pump and terminal air handling unit are running. From the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during a second historical time period are selected as a second training data subset. During the second historical time period, the target building is in an internally heated state and the air conditioning system is operating stably. Constructing a room temperature prediction model using the first training data subset and the second training data subset includes: Filtering the chilled water outlet temperature, chilled water return temperature, and indoor air temperature at a target time from the first training data subset, wherein the target time is the time when the air conditioning system temperature reaches the dew point temperature when the target building is in a state of no internal heat, the air conditioning compressor is turned off, and the water pump and the terminal air handling unit are running; Obtaining a first air-conditioning system temperature based on a chilled water outlet temperature and a chilled water return temperature at a target time, and obtaining a temperature difference coefficient based on the first air-conditioning system temperature and an indoor air temperature at the target time; obtaining a return air enthalpy value based on the return air temperature and indoor humidity of the second training data subset, and obtaining a supply air enthalpy value based on the supply air temperature and indoor humidity of the second training data subset; Obtaining a preset air specific heat capacity, and obtaining a sensible heat ratio based on the air specific heat capacity, the return air temperature of the second training data subset, the supply air temperature of the second training data subset, the return air enthalpy value, and the supply air enthalpy value; Obtaining a latent heat and sensible heat ratio based on the sensible heat ratio, and obtaining a second air-conditioning system temperature based on the chilled water outlet temperature and the chilled water return temperature of the second training data subset; Obtaining a proportionality coefficient based on the ratio of latent heat to sensible heat, the indoor air temperature of the second training data subset, the second air conditioning system temperature, and the temperature difference coefficient; Based on the temperature difference coefficient and the proportional coefficient, the heat storage characteristics of the target building and the air-conditioning system are coupled to construct a temperature prediction model.

2. The method for quantifying the flexible adjustment potential of an air conditioning system according to claim 1, characterized in that: Calculate the thermal resistance of the air conditioning system in the room temperature prediction model, including: Filtering several groups of target data collected within a target historical time period from the second training data subset, each group of target data including chilled water flow, chilled water return temperature, chilled water outlet temperature, and indoor air temperature; Based on the chilled water return temperature and the chilled water outlet temperature in each set of target data, a third air-conditioning system temperature is obtained; Based on the chilled water flow rate, chilled water return temperature and chilled water outlet temperature in each set of target data, the heat transfer between the indoor air and the air conditioning system is obtained; Obtaining a sensible heat transfer amount between the indoor air and the air conditioning system based on the heat transfer amount and the sensible heat ratio between the indoor air and the air conditioning system; The thermal resistance of the air-conditioning system is obtained based on the third air-conditioning system temperature, the sensible heat transfer between the indoor air and the air-conditioning system, and the indoor air temperature in each set of target data.

3. The method for quantifying the flexible adjustment potential of an air conditioning system according to claim 1, characterized in that: Calculate the thermal resistance and heat capacity parameter combinations in the room temperature prediction model, including: Calculating a first wall temperature corresponding to the first training data subset based on the first training data subset, where the first wall temperature is the wall temperature corresponding to the initial acquisition moment in the first training data subset; Substituting the first wall temperature and the first training data subset into a room temperature prediction model with a fixed thermal resistance of the air-conditioning system, using the combination of thermal resistance and heat capacity parameters to be determined in the room temperature prediction model as variables to be optimized, performing a first preset round of optimization using a genetic algorithm, and calculating a value of a first objective function corresponding to each round of optimization; Sort the values ​​of the first objective function of all rounds, select a preset number of thermal resistance and heat capacity parameter combinations corresponding to the first objective function from small to large, and calculate the deviation value corresponding to each selected thermal resistance and heat capacity parameter combination; A set of thermal resistance and heat capacity parameter combinations with the smallest deviation value is selected as the final thermal resistance and heat capacity parameter combination.

4. The method for quantifying the flexible adjustment potential of an air conditioning system according to claim 3, characterized in that: Calculating a first wall temperature corresponding to the first training data subset based on the first training data subset includes: Using the indoor air temperature at the initial collection time in the first training data subset as a starting point and the outdoor air dry-bulb temperature at the initial collection time as an end point, generating a candidate wall temperature sequence according to a preset fixed step size, wherein the candidate wall temperature sequence includes a plurality of candidate wall temperatures; Substituting each candidate wall temperature and the first training data subset into a room temperature prediction model for which the thermal resistance of the air conditioning system has been determined, and performing a second preset round of optimization on the thermal resistance and heat capacity parameter combinations to be determined in the room temperature prediction model using a genetic algorithm to obtain a plurality of thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature; Using a preset deviation function to evaluate several thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature, and selecting the thermal resistance and heat capacity parameter combination with the smallest deviation as the candidate thermal resistance and heat capacity parameter combination corresponding to the candidate wall temperature; Selecting the time-series indoor air temperature and the time-series outdoor dry-bulb air temperature collected during the second target historical time period from the historically collected time-series indoor and outdoor environmental data; Obtaining a plurality of estimated wall temperatures based on the time-series indoor air temperature, the time-series outdoor dry-bulb air temperature, and the candidate thermal resistance and heat capacity parameter combinations corresponding to each candidate wall temperature; The difference between each candidate wall temperature and its corresponding estimated wall temperature is calculated, and the candidate wall temperature with the smallest difference is selected as the first wall temperature.

5. The method for quantifying the flexible adjustment potential of an air conditioning system according to claim 1, characterized in that: Calculates the indoor heat gain schedule for the room temperature prediction model, including: Substituting the second training data subset into a room temperature prediction model for a determined combination of thermal resistance and thermal resistance / capacity parameters of the air conditioning system, using the hourly indoor heat gain within a preset time range in the room temperature prediction model as a variable to be optimized, and iteratively adjusting its value through a genetic algorithm to minimize the second objective function; When the preset end condition is reached, an indoor heating schedule consisting of the indoor heat gain per hour within the preset time range is obtained.

6. A device for quantifying the flexible adjustment potential of an air conditioning system, characterized in that: include: a data acquisition module, configured to continuously collect indoor and outdoor environmental data and air conditioning system operation data of a target building using a plurality of preset sensors, and to select a first training data subset and a second training data subset from the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data; The indoor and outdoor environmental data include indoor air temperature and outdoor air dry-bulb temperature, and the air conditioning system operation data include chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity. A first training data subset and a second training data subset are screened from the historically collected time-series indoor and outdoor environmental data and the time-series air conditioning system operation data, including: From the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during a first historical time period are selected as a first training data subset. During the first historical time period, the target building is in a zero-heat state, the air conditioning compressor is turned off, and the water pump and terminal air handling unit are running. From the historically collected time-series indoor and outdoor environmental data and time-series air conditioning system operation data, the indoor air temperature, outdoor air dry-bulb temperature, chilled water outlet temperature, chilled water return temperature, return air temperature, supply air temperature, chilled water flow rate, and air humidity collected during a second historical time period are selected as a second training data subset. During the second historical time period, the target building is in an internally heated state and the air conditioning system is operating stably. a model construction module, configured to construct a room temperature prediction model using the first training data subset and the second training data subset, wherein the room temperature prediction model is configured to predict a change in the temperature within a target building after a demand response is initiated, wherein the demand response is that the target building has internal heat and the air conditioning compressor is turned off and the water pump and terminal air handling unit are operated; Constructing a room temperature prediction model using the first training data subset and the second training data subset includes: Filtering the chilled water outlet temperature, chilled water return temperature, and indoor air temperature at a target time from the first training data subset, wherein the target time is the time when the air conditioning system temperature reaches the dew point temperature when the target building is in a state of no internal heat, the air conditioning compressor is turned off, and the water pump and the terminal air handling unit are running; Obtaining a first air-conditioning system temperature based on a chilled water outlet temperature and a chilled water return temperature at a target time, and obtaining a temperature difference coefficient based on the first air-conditioning system temperature and an indoor air temperature at the target time; obtaining a return air enthalpy value based on the return air temperature and indoor humidity of the second training data subset, and obtaining a supply air enthalpy value based on the supply air temperature and indoor humidity of the second training data subset; Obtaining a preset air specific heat capacity, and obtaining a sensible heat ratio based on the air specific heat capacity, the return air temperature of the second training data subset, the supply air temperature of the second training data subset, the return air enthalpy value, and the supply air enthalpy value; Obtaining a latent heat and sensible heat ratio based on the sensible heat ratio, and obtaining a second air-conditioning system temperature based on the chilled water outlet temperature and the chilled water return temperature of the second training data subset; Obtaining a proportionality coefficient based on the ratio of latent heat to sensible heat, the indoor air temperature of the second training data subset, the second air conditioning system temperature, and the temperature difference coefficient; Based on the temperature difference coefficient and the proportional coefficient, a temperature prediction model is constructed by coupling the heat storage characteristics of the target building and the air conditioning system; a parameter determination module, configured to sequentially calculate the air conditioning system thermal resistance, the thermal resistance and heat capacity parameter combination, and the indoor heat gain schedule in the room temperature prediction model, and use the calculation results as fixed parameters of the model after each calculation, wherein the thermal resistance and heat capacity parameter combination includes: the resistance of heat transfer between the building envelope and indoor air, the resistance of heat transfer between the building envelope and outdoor air, the indoor air heat capacity, the air conditioning system heat capacity, and the building envelope heat capacity; The curve generation module is used to calculate the wall temperature at the time of demand response initiation when a demand response instruction is received, and substitute the wall temperature and the indoor and outdoor environmental data and air conditioning system operation data collected last time before the demand response initiation into the room temperature prediction model with fixed parameters to obtain the room temperature prediction curve; The duration determination module is used to determine the maximum adjustable duration based on the demand response start time, the preset temperature threshold and the room temperature prediction curve, and use it as the flexible adjustment potential of the air-conditioning system.

7. A terminal, characterized in that: The terminal includes a memory, a processor, and an air-conditioning system flexible adjustment potential quantification program stored in the memory and runnable on the processor. When the air-conditioning system flexible adjustment potential quantification program is executed by the processor, the steps of the air-conditioning system flexible adjustment potential quantification method as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an air conditioning system flexible adjustment potential quantification program. When the air conditioning system flexible adjustment potential quantification program is executed by the processor, the steps of the air conditioning system flexible adjustment potential quantification method according to any one of claims 1 to 5 are implemented.

Images