Self-organizing construction method for energy-saving group control strategy of variable air volume air conditioning system

Abstract

The invention relates to a self-organizing construction method for an energy-saving group control strategy of a variable air volume air conditioning system. The method is suitable for the technical field of building energy saving and intelligent building control. According to the technical scheme, the method comprises the steps that on the basis of equipment composition of a target air conditioning system, intelligent agent individual optimization model classes corresponding to all equipment are called from a pre-constructed intelligent agent class library, and corresponding intelligent agent individual optimization models are generated for all specific equipment in the system; based on the topological structure of the target air conditioning system, the connection relation between the intelligent agent individual optimization models corresponding to all the devices in the system is established; based on the connection relation between the intelligent agent individual optimization models, collaborative optimization rules between the intelligent agent individual optimization models and the adjacent intelligent agent individual optimization models are sequentially added into the intelligent agent individual optimization models, and an energy-saving group control strategy of the target air conditioning system is formed; and real-time operation parameters of the target air conditioning system are obtained and input into the energy-saving group control strategy, optimization of control actions is carried out through a collaborative optimization algorithm, and the optimal control actions of all devices in the system are output.

Description

Technical Field

[0001] The invention relates to a self-organizing construction method of a variable air volume air conditioning system energy-saving group control strategy, which is applicable to the technical fields of building energy-saving and intelligent building control. Background Art

[0002] The variable air volume air conditioning system is the most widely used terminal air conditioning form in large public buildings. It can dynamically change the air supply volume of each terminal to meet the needs of individual load changes at the terminal. Studies have shown that the energy consumption of variable air volume air conditioning systems accounts for a large proportion of the total energy consumption of buildings. Optimizing the control of variable air volume air conditioning systems through reasonable energy-saving group control strategies, thereby reducing the system operation energy consumption while meeting the load requirements of each terminal, is of great significance for achieving building energy conservation.

[0003] The actual variable air volume air conditioning system has a high degree of personalization, and the equipment composition and topology of different systems are different. The existing energy-saving group control strategy of variable air volume air conditioning system generally adopts a top-down development approach, which requires field engineers to develop case by case according to the specific form and characteristics of the target system. This development process is time-consuming and labor-intensive, and requires a high level of professionalism. The existing energy-saving control strategy is difficult to migrate to other variable air volume air conditioning systems at a low cost, which hinders the large-scale promotion and application of the energy-saving group control strategy of variable air volume air conditioning system in this field. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in view of the above-mentioned problems, a self-organizing construction method of an energy-saving group control strategy for a variable air volume air conditioning system is provided.

[0005] The technical solution adopted by the present invention is: a self-organizing construction method of an energy-saving group control strategy for a variable air volume air conditioning system, characterized in that it includes:

[0006] Based on the equipment composition of the target air-conditioning system, the agent individual optimization model class corresponding to each device is retrieved from the pre-built agent class library, and the corresponding agent individual optimization model is generated for each specific device in the system;

[0007] Based on the topological structure of the target air-conditioning system, the connection relationship between the individual optimization models of the intelligent agents corresponding to each device in the system is established;

[0008] Based on the connection relationship between the individual optimization models of intelligent agents, the collaborative optimization rules between the individual optimization models of adjacent intelligent agents are added to the individual optimization models of intelligent agents in turn to form an energy-saving group control strategy for the target air-conditioning system.

[0009] The real-time operating parameters of the target air-conditioning system are obtained and input into the energy-saving group control strategy. The control actions are optimized through the collaborative optimization algorithm, and the optimal control actions of each device in the system are output.

[0010] The intelligent agent class library includes one or more of an air handling unit intelligent agent individual optimization model class, an air duct intelligent agent individual optimization model class and an air conditioning terminal intelligent agent individual optimization model class.

[0011] The air handling unit intelligent agent individual optimization model class includes:

[0012]

[0013]

[0014]

[0015]

[0016] Wherein, the subscript ahu represents the air handling unit, the superscript in represents the inlet side parameters, out represents the outlet side parameters, m represents the mass flow rate, T represents the temperature, p represents the pressure, P ahu is the operating energy consumption of the air handling unit, P ahu，rt is the rated power of the air handling unit, f is the operating frequency of the blower, and f rt is the rated operating frequency of the blower, m ahu，rt is the rated flow rate of the air handling unit on the air supply side, Q ahu is the cooling capacity of the air handling unit, c p is the constant pressure specific heat capacity of air, ρ air is the air density, g is the acceleration due to gravity, p is rt is the rated pressure head of the blower, is the inlet flow rate of the chilled water side of the air handling unit, is the inlet temperature of the chilled water side of the air handling unit, a1-a4 and b1-b3 are the performance coefficients of the air handling unit.

[0017] The air duct agent individual optimization model class includes:

[0018] min{0}

[0019]

[0020]

[0021] Where m represents mass flow, superscript in represents inlet side parameters, out represents outlet side parameters, T represents temperature, p represents pressure, subscript ad represents air duct, Ω outis the air supply branch pipe assembly, s i is the pipe impedance of the air supply branch pipe i.

[0022] The air-conditioning terminal agent individual optimization model class includes:

[0023] min C room = -β·(T room -T sp ) 2

[0024]

[0025]

[0026] Where T represents temperature, m represents mass flow, p represents pressure, and the subscript room represents the air conditioning terminal. sp Set the temperature for the terminal room, C room is the thermal comfort of the end user, β is the weight coefficient between the operating energy consumption and the thermal comfort of the end user, Q room is the cooling load of the terminal room, c p is the constant pressure specific heat capacity of air, A dp is the cross-sectional area of ​​the air valve of the variable air volume air conditioning box, is the air valve opening of the variable air volume air conditioning box, exp(·) is the exponential operator, a dp -c dp is the air valve resistance characteristic coefficient.

[0027] Based on the equipment composition of the target air-conditioning system, the intelligent agent individual optimization model class corresponding to each device is retrieved from the pre-built intelligent agent class library, and a corresponding intelligent agent individual optimization model is generated for each specific device in the system, including:

[0028] Set the performance parameters of the individual optimization model of the intelligent agent corresponding to each device, including capacity parameters and undetermined coefficients;

[0029] According to the actual control requirements, set the variables to be optimized of the individual optimization model of the intelligent agent corresponding to each device;

[0030] According to the actual operating capabilities of the equipment, set the upper and lower limits of the optimization of the operating parameters of the individual optimization model of the intelligent agent corresponding to each device.

[0031] The method of sequentially adding collaborative optimization rules between adjacent intelligent agent individual optimization models in the intelligent agent individual optimization models based on the connection relationship between the intelligent agent individual optimization models includes:

[0032] Consistency constraints between the export parameters of the individual optimization model of the upstream intelligent agent and the import parameters of the individual optimization model of the downstream intelligent agent;

[0033] The consistency constraint includes one or more of a temperature equality constraint, a flow equality constraint and a pressure equality constraint.

[0034] The real-time operating parameters of the target air-conditioning system are obtained and input into the energy-saving group control strategy, and the control action is optimized through the collaborative optimization algorithm, and the optimal control action of each device in the system is output, including:

[0035] Initialize the global variable matrix and the Lagrange multiplier matrix for each agent Among them, N cc N represents the total number of consistency constraints of each agent's individual optimization model. cc，i represents the number of consistency constraints of the agent individual optimization model i;

[0036] Set the penalty coefficient ρ, the number of iterations k = 0, and the upper limit of the number of iterations K max ;

[0037] According to the consistency constraints added by the individual optimization model of each agent, the collaborative optimization model of each agent is set as shown in the following formula:

[0038]

[0039] stg i ≤0,h i =0

[0040] Among them, f i represents the objective function of the agent individual optimization model i, g i represents the inequality constraint matrix of the agent individual optimization model i, h i represents the equality constraint matrix of the agent individual optimization model i, ||·||2 represents the 2-norm operator, represents the global variable value related to the agent individual optimization model i in the kth iteration, represents the Lagrange multiplier value associated with the agent individual optimization model i in the kth iteration, The parameter matrix representing the inlet and outlet temperature, flow rate and pressure of the agent individual optimization model i;

[0041] Each agent solves the above collaborative optimization model in parallel to obtain the optimal solution for the k+1th iteration.

[0042] Update the Lagrange multiplier matrix and global variable matrix of the k+1th round as shown below:

[0043]

[0044]

[0045] Where, subscript l represents the set of agents adjacent to agent i, and N is the number of agents;

[0046] Determine whether the collaborative optimization process has reached convergence, as shown in the following formula:

[0047]

[0048]

[0049] If the above formula is satisfied or the number of iterations k ≥ K max , this optimization process ends; otherwise, the next round of iteration is carried out.

[0050] A self-organizing construction device for an energy-saving group control strategy of a variable air volume air conditioning system, characterized by comprising:

[0051] The model generation module is used to retrieve the intelligent agent individual optimization model class corresponding to each device from the pre-built intelligent agent class library based on the equipment composition of the target air-conditioning system, and generate the corresponding intelligent agent individual optimization model for each specific device in the system;

[0052] The model connection module is used to establish the connection relationship between the individual optimization models of the intelligent agents corresponding to each device in the system based on the topological structure of the target air-conditioning system;

[0053] A strategy forming module is used to add collaborative optimization rules between the individual optimization models of adjacent intelligent agents in the individual optimization models of intelligent agents in turn based on the connection relationship between the individual optimization models of intelligent agents, so as to form an energy-saving group control strategy for the target air-conditioning system;

[0054] The algorithm optimization module is used to obtain the real-time operating parameters of the target air-conditioning system, input them into the energy-saving group control strategy, optimize the control actions through the collaborative optimization algorithm, and output the optimal control actions for each device in the system.

[0055] A storage medium stores a computer program that can be executed by a processor, characterized in that when the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system are implemented.

[0056] An electronic device has a memory and a processor, wherein the memory stores a computer program that can be executed by the processor, and is characterized in that when the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system are implemented.

[0057] A variable air volume air conditioning system, characterized by comprising:

[0058] The electronic device described above;

[0059] The actuator is set corresponding to each device in the variable air volume air conditioning system and is used to control the corresponding device based on the optimal control action output by the electronic device.

[0060] The beneficial effects of the present invention are as follows: based on the equipment composition of the target air-conditioning system, the present invention retrieves the intelligent agent individual optimization model class corresponding to each device from a pre-built intelligent agent class library, and then quickly generates the corresponding intelligent agent individual optimization model for the equipment in the system, and uses the predefined intelligent agent individual optimization model to perform a standardized description of the thermal characteristics and operating characteristics of different types of equipment in the variable air volume air-conditioning system, thereby realizing the rapid construction of the intelligent agent model.

[0061] The present invention connects individual optimization models of intelligent agents based on the topological structure of the actual system, thereby self-organizing and establishing collaborative optimization rules between intelligent agents to achieve energy-saving optimization of the target system.

[0062] For different variable air volume air conditioning systems, the present invention can quickly build a system-level energy-saving group control strategy according to its equipment composition and topological structure through the reuse and connection of intelligent individual optimization models. It has high development efficiency, can be applied to different forms of variable air volume air conditioning systems, and has good energy-saving effects.

[0063] The individual optimization model class of the air-conditioning terminal intelligent agent in the present invention takes into account the thermal comfort of the end user, so that the present invention can effectively improve the operating efficiency of the variable air volume air-conditioning system while ensuring the thermal comfort of the end user. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 Schematic diagram of the structure of the variable air volume air conditioning system in Example 1.

[0065] Figure 2 This is a flow chart of the self-organizing construction method of the energy-saving group control strategy in Example 1.

[0066] Figure 3 is the cooling load demand on a typical summer day in Example 1.

[0067] Figure 4 The results of the system operation energy consumption in Example 1 are compared.

[0068] Figure 5 The results of the indoor temperature of the terminal room in Example 1 are compared. DETAILED DESCRIPTION

[0069] Example 1: Figure 1As shown in the figure, the variable air volume air conditioning system in this embodiment consists of one air handling unit and five variable air volume air conditioning boxes, which respectively adjust the temperature of the corresponding air-conditioned terminal rooms. The rated power and rated air volume of the air handling unit are 8.5kW and 417m 3 / min, with a rated cooling capacity of 160kW. The size of each room is 12m×16m×4m, and they are all office-type rooms. The office hours are from 8am to 19pm.

[0070] like Figure 2 As shown, the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system in this embodiment includes the following steps:

[0071] S1. Based on the equipment composition of the target air-conditioning system, the intelligent agent individual optimization model class corresponding to each device is retrieved from the pre-built intelligent agent class library, and the corresponding intelligent agent individual optimization model is generated for each specific device in the system.

[0072] S1-1. According to the types of equipment that may be included in the variable air volume air conditioning system, a corresponding intelligent agent individual optimization model class is established for each type of equipment, and an intelligent agent class library is constructed.

[0073] The types of equipment that a variable air volume air conditioning system may contain include air handling units, air ducts, and air conditioning terminals. The intelligent agent individual optimization model class for each type of equipment is constructed as follows:

[0074] A. The individual optimization model of the air handling unit agent is as follows:

[0075]

[0076]

[0077]

[0078] Wherein, the subscript ahu represents the air handling unit, the superscript in represents the inlet side parameters, out represents the outlet side parameters, m represents the mass flow rate, T represents the temperature, p represents the pressure, P ahu is the operating energy consumption of the air handling unit, P ahu，rt is the rated power of the air handling unit, f is the operating frequency of the blower, and f rt is the rated operating frequency of the blower, m ahu,rt is the rated flow rate of the air handling unit on the air supply side, Q ahu is the cooling capacity of the air handling unit, c p is the constant pressure specific heat capacity of air, ρ air is the air density, g is the acceleration due to gravity, p is rt is the rated pressure head of the blower, is the inlet flow rate of the chilled water side of the air handling unit, is the inlet temperature of the chilled water side of the air handling unit, a1-a4 and b1-b3 are the performance coefficients of the air handling unit.

[0079] B. The individual optimization model of the air duct agent is shown as follows:

[0080] min{0}

[0081]

[0082]

[0083] Among them, the subscript ad represents the air duct, Ω out is the air supply branch pipe assembly, s i is the pipe impedance of the air supply branch pipe i.

[0084] C. The individual optimization model of the air-conditioning terminal agent is as follows:

[0085] min C room = -β·(T room -T sp ) 2

[0086]

[0087]

[0088] The subscript room indicates the air conditioner terminal, T sp Set the temperature for the terminal room, C room is the thermal comfort of the end user, β is the weight coefficient between the operating energy consumption and the thermal comfort of the end user, Q room is the cooling load of the terminal room, A dp is the cross-sectional area of ​​the air valve of the variable air volume air conditioning box, is the air valve opening of the variable air volume air conditioning box, exp(·) is the exponential operator, a dp -c dp is the air valve resistance characteristic coefficient.

[0089] S1-2. According to the equipment composition of the target air-conditioning system, the intelligent agent individual optimization model class corresponding to each device is retrieved from the intelligent agent class library. Through the instantiation of the intelligent agent class, the corresponding intelligent agent individual optimization model is generated for each specific device to complete the initialization of the intelligent agent model.

[0090] Set the performance parameters of the individual optimization model of each equipment agent, including capacity parameters and undetermined coefficients. Capacity parameters include the rated power, rated flow, and rated cooling capacity of the air handling unit. The undetermined coefficients include the performance coefficient of the air handling unit, the pipe impedance of each branch of the air duct, and the cross-sectional area and resistance characteristic coefficient of the air valve of the variable air volume air conditioning box.

[0091] According to the actual control requirements, the variables to be optimized of the individual optimization model of each device agent are set. The variables to be optimized for the air handling unit include the supply air temperature set value and the supply fan operating frequency, and the variables to be optimized for the air conditioning terminal include the variable air volume air conditioning box air valve opening and the terminal room supply air volume.

[0092] According to the actual equipment operation capacity, set the upper and lower limits of the optimization of the operation parameters of the individual optimization model of each equipment agent. The operation parameters of the air handling unit agent include the flow rate, temperature and pressure of the inlet and outlet, the operation frequency of the blower and the cooling capacity; the operation parameters of the air duct agent include the flow rate, temperature and pressure of the inlet and outlet; the operation parameters of the air conditioning terminal agent include the indoor temperature and the flow rate, temperature and pressure of the inlet.

[0093] S2. Based on the topological structure of the target air-conditioning system, establish the connection relationship between the individual optimization models of the intelligent agents corresponding to each device in the system.

[0094] In this embodiment, the air handling unit intelligent body is first connected to the air duct intelligent body, and the air duct intelligent body is then connected to five air conditioning terminal (variable air volume air conditioning box) intelligent bodies respectively.

[0095] S3. Based on the connection relationship between the individual optimization models of intelligent agents, collaborative optimization rules between the individual optimization models of adjacent intelligent agents are added to the individual optimization models of intelligent agents in turn to form an energy-saving group control strategy for the target air-conditioning system.

[0096] The collaborative optimization rules added to each agent's individual optimization model include consistency constraints with the export parameters of its upstream agent model and consistency constraints with the import parameters of its downstream agent model. The two consistency constraints are further divided into temperature equality constraints, flow equality constraints, and pressure equality constraints, as shown in the following formula:

[0097]

[0098]

[0099]

[0100] Among them, j represents the upstream agent adjacent to agent i, and k represents the downstream agent adjacent to agent i.

[0101] S4. Obtain the real-time operating parameters of the target air-conditioning system, input them into the energy-saving group control strategy, optimize the control actions through the collaborative optimization algorithm, and output the optimal control actions for each device in the system.

[0102] S4-1: Initialize the global variable matrix and the Lagrange multiplier matrix for each agent Among them, N cc represents the total number of consistency constraints of each agent, N cc，i Represents the number of consistency constraints that agent i has.

[0103] In this embodiment, N cc =6, air handling unit agent N cc，1 =1, air duct agent N cc，2 =6, air conditioning terminal intelligent agent 1-5N cc，3 -N cc，7 =1.

[0104] S4-2: Set the penalty coefficient ρ, the number of iterations k = 0, and the upper limit of the number of iterations K max In this example, the penalty coefficient ρ is set to 0.5, the number of iterations k is set to 0, and the upper limit of the number of iterations K is set to max =150.

[0105] S4-3: According to the consistency constraints added by each agent, the collaborative optimization model of each agent is set as shown in the following formula:

[0106]

[0107] stg i ≤0,h i =0

[0108] Among them, f i represents the objective function of agent i, g i represents the inequality constraint matrix of agent i, h i represents the equality constraint matrix of agent i, ||·||2 represents the 2-norm operator, represents the value of the global variable related to agent i in the kth iteration, represents the Lagrange multiplier value associated with agent i in the kth iteration, The parameter matrix representing the inlet and outlet temperature, flow rate, and pressure of agent i.

[0109] S4-4: Each agent solves the above collaborative optimization model in parallel to obtain the optimal solution for the k+1th iteration

[0110] S4-5: Update the Lagrange multiplier matrix and global variable matrix of the k+1th round, as shown in the following formula:

[0111]

[0112]

[0113] Wherein, subscript l represents the set of agents adjacent to agent i, and N is the number of agents, N = 7 in this example.

[0114] S4-6: Determine whether the collaborative optimization process has reached convergence, as shown in the following formula:

[0115]

[0116]

[0117] In this embodiment, ε1=ε2=0.01. If the above formula is satisfied or the number of iterations k≥K max , the optimization process ends and goes to step S4-7. Otherwise, repeat steps S4-4 to S4-6.

[0118] S4-7: Output the optimal variable result to be optimized to the actuator for execution of the control action.

[0119] The operating data of a typical summer day are selected to verify the energy-saving effect of the group control strategy. The cooling load demand on that day is as follows: Figure 3 The traditional rule-based group control strategy is set for performance comparison. Under this strategy, the air handling unit fan always runs at the industrial frequency, and the valve opening of the terminal variable air volume air conditioning box is always kept fully open. The indoor temperature setting value T of each terminal room sp At 26°C, the air handling unit inlet pressure is -50Pa (gauge pressure), room pressure p room The air supply temperature of the air handling unit is 25Pa (gauge pressure). The control interval is uniformly set to 1h.

[0120] Comparison of system energy consumption under two group control strategies Figure 4 As shown. Since the air handling unit blower always runs at the industrial frequency under the traditional rule group control strategy, its operating energy consumption remains high at the rated power. In contrast, the energy-saving group control strategy constructed in this embodiment adopts the model optimization method to calculate the optimal control scheme under different working conditions. The total operating energy consumption of the system is reduced from 107.5kWh to 72.3kWh, a reduction of 32.8%.

[0121] The results of indoor temperature of each terminal room under the two group control strategies are compared. Figure 5As shown. Because the air handling unit blower always supplies air to the terminal at the rated air supply under the rule control strategy, there is a lack of effective regulation of the terminal temperature, resulting in the indoor temperature in most time periods under this strategy being significantly deviated from the set value (26°C), reducing the thermal comfort of the end users. In contrast, the energy-saving group control strategy constructed in this embodiment adjusts the air supply volume of each room through coordination between intelligent agents, so that the indoor temperature of each room is maintained near the set value, with a maximum deviation of less than 0.5K, achieving a balance between operating energy consumption and terminal thermal comfort.

[0122] In summary, this embodiment can quickly construct an energy-saving group control strategy adapted to local conditions according to the equipment composition and topological structure of the actual variable air volume air conditioning system, has high development efficiency and versatility, and is applicable to different forms of variable air volume air conditioning systems. At the same time, the energy-saving group control strategy constructed by this embodiment has high energy-saving and temperature control capabilities, and can effectively improve the operating efficiency of the variable air volume air conditioning system while ensuring the thermal comfort of the end users.

[0123] Embodiment 2: This embodiment is a self-organizing construction device for a variable air volume air conditioning system energy-saving group control strategy, including: a model generation module, a model connection module, a strategy formation module, an algorithm optimization module, etc.

[0124] In this example, the model generation module is used to retrieve the intelligent agent individual optimization model class corresponding to each device from the pre-built intelligent agent class library based on the equipment composition of the target air-conditioning system, and generate the corresponding intelligent agent individual optimization model for each specific device in the system.

[0125] In this embodiment, the model connection module is used to establish the connection relationship between the individual optimization models of the intelligent agents corresponding to the devices in the system based on the topological structure of the target air-conditioning system.

[0126] In this example, the strategy formation module is used to add collaborative optimization rules between adjacent intelligent agent individual optimization models in the intelligent agent individual optimization model based on the connection relationship between the intelligent agent individual optimization models, so as to form an energy-saving group control strategy for the target air-conditioning system.

[0127] In this embodiment, the algorithm optimization module is used to obtain the real-time operating parameters of the target air-conditioning system, input them into the energy-saving group control strategy, optimize the control action through the collaborative optimization algorithm, and output the optimal control action for each device in the system.

[0128] Embodiment 3: This embodiment is a storage medium on which a computer program that can be executed by a processor is stored. When the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system in Embodiment 1 are implemented.

[0129] Embodiment 4: This embodiment is an electronic device having a memory and a processor. The memory stores a computer program that can be executed by the processor. When the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system in Embodiment 1 are implemented.

[0130] Embodiment 5: This embodiment is a variable air volume air conditioning system, in which each device is configured with an actuator to control its action. The system also includes the electronic device in Embodiment 4, which outputs the optimal result of the variable to be optimized to the actuator to execute the control action.

Claims

1. A self-organizing construction method for energy-saving group control strategy of variable air volume air conditioning system, characterized in that: include: Based on the equipment composition of the target air-conditioning system, the agent individual optimization model class corresponding to each device is retrieved from the pre-built agent class library, and the corresponding agent individual optimization model is generated for each specific device in the system; Based on the topological structure of the target air-conditioning system, the connection relationship between the individual optimization models of the intelligent agents corresponding to each device in the system is established; Based on the connection relationship between the individual optimization models of intelligent agents, the collaborative optimization rules between the individual optimization models of adjacent intelligent agents are added to the individual optimization models of intelligent agents in turn to form an energy-saving group control strategy for the target air-conditioning system. The real-time operating parameters of the target air-conditioning system are obtained and input into the energy-saving group control strategy. The control actions are optimized through the collaborative optimization algorithm, and the optimal control actions of each device in the system are output.

2. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 1 is characterized by: The intelligent agent class library includes one or more of an air handling unit intelligent agent individual optimization model class, an air duct intelligent agent individual optimization model class and an air conditioning terminal intelligent agent individual optimization model class.

3. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 2 is characterized in that: The air handling unit intelligent agent individual optimization model class includes: Wherein, the subscript ahu represents the air handling unit, the superscript in represents the inlet side parameters, out represents the outlet side parameters, m represents the mass flow rate, T represents the temperature, p represents the pressure, P ahu is the operating energy consumption of the air handling unit, P ahu,rt is the rated power of the air handling unit, f is the operating frequency of the blower, and f rt is the rated operating frequency of the blower, m ahu,rt is the rated flow rate of the air handling unit on the air supply side, Q ahu is the cooling capacity of the air handling unit, c p is the constant pressure specific heat capacity of air, ρ air is the air density, g is the acceleration due to gravity, p rt is the rated pressure head of the blower, is the inlet flow rate of the chilled water side of the air handling unit, is the inlet temperature of the chilled water side of the air handling unit, a1-a4 and b1-b3 are the performance coefficients of the air handling unit.

4. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 2 is characterized in that: The air duct agent individual optimization model class includes: min{0} Where m represents mass flow, superscript in represents inlet side parameters, out represents outlet side parameters, T represents temperature, p represents pressure, subscript ad represents air duct, Ω out is the air supply branch pipe assembly, s i is the pipe impedance of the air supply branch pipe i.

5. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 2 is characterized in that: The air-conditioning terminal agent individual optimization model class includes: min C room =-β·(T room -T sp ) 2 Where T represents temperature, m represents mass flow, p represents pressure, and the subscript room represents the air conditioning terminal. sp Set the temperature for the terminal room, C room is the thermal comfort of the end user, β is the weight coefficient between the operating energy consumption and the thermal comfort of the end user, Q room is the cooling load of the terminal room, c p is the constant pressure specific heat capacity of air, A dp is the cross-sectional area of ​​the air valve of the variable air volume air conditioning box, θ cdp is the air valve opening of the variable air volume air conditioning box, exp(·) is the exponential operator, a dp -c dp is the air valve resistance characteristic coefficient.

6. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 1 is characterized in that: Based on the equipment composition of the target air-conditioning system, the intelligent agent individual optimization model class corresponding to each device is retrieved from the pre-built intelligent agent class library, and a corresponding intelligent agent individual optimization model is generated for each specific device in the system, including: Set the performance parameters of the individual optimization model of the intelligent agent corresponding to each device, including capacity parameters and undetermined coefficients; According to the actual control requirements, set the variables to be optimized of the individual optimization model of the intelligent agent corresponding to each device; According to the actual operating capabilities of the equipment, set the upper and lower limits of the optimization of the operating parameters of the individual optimization model of the intelligent agent corresponding to each device.

7. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 1 is characterized by: The method of sequentially adding collaborative optimization rules between adjacent intelligent agent individual optimization models in the intelligent agent individual optimization models based on the connection relationship between the intelligent agent individual optimization models includes: Consistency constraints between the export parameters of the individual optimization model of the upstream intelligent agent and the import parameters of the individual optimization model of the downstream intelligent agent; The consistency constraint includes one or more of a temperature equality constraint, a flow equality constraint and a pressure equality constraint.

8. The self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to claim 1 is characterized in that: The real-time operating parameters of the target air-conditioning system are obtained and input into the energy-saving group control strategy, and the control action is optimized through the collaborative optimization algorithm, and the optimal control action of each device in the system is output, including: Initialize the global variable matrix and the Lagrange multiplier matrix for each agent Among them, N cc N represents the total number of consistency constraints of each agent's individual optimization model. cc,i represents the number of consistency constraints of the agent individual optimization model i; Set the penalty coefficient ρ, the number of iterations k = 0, and the upper limit of the number of iterations K max ; According to the consistency constraints added by the individual optimization model of each agent, the collaborative optimization model of each agent is set as shown in the following formula: s.t.g i ≤0,h i =0 Among them, f i represents the objective function of the agent individual optimization model i, g i represents the inequality constraint matrix of the agent individual optimization model i, h i represents the equality constraint matrix of the agent individual optimization model i, ‖·‖2 represents the 2-norm operator, represents the global variable value related to the agent individual optimization model i in the kth iteration, represents the Lagrange multiplier value associated with the agent individual optimization model i in the kth iteration, The parameter matrix representing the inlet and outlet temperature, flow rate and pressure of the agent individual optimization model i; Each agent solves the above collaborative optimization model in parallel to obtain the optimal solution for the k+1th iteration. Update the Lagrange multiplier matrix and global variable matrix of the k+1th round as shown below: Where, subscript l represents the set of agents adjacent to agent i, and N is the number of agents; Determine whether the collaborative optimization process has reached convergence, as shown in the following formula: If the above formula is satisfied or the number of iterations k ≥ K max , this optimization process ends; otherwise, the next round of iteration is carried out.

9. A self-organizing construction device for energy-saving group control strategy of variable air volume air conditioning system, characterized in that: include: The model generation module is used to retrieve the intelligent agent individual optimization model class corresponding to each device from the pre-built intelligent agent class library based on the equipment composition of the target air-conditioning system, and generate the corresponding intelligent agent individual optimization model for each specific device in the system; The model connection module is used to establish the connection relationship between the individual optimization models of the intelligent agents corresponding to each device in the system based on the topological structure of the target air-conditioning system; A strategy forming module is used to add collaborative optimization rules between the individual optimization models of adjacent intelligent agents in the individual optimization models of intelligent agents in turn based on the connection relationship between the individual optimization models of intelligent agents, so as to form an energy-saving group control strategy for the target air-conditioning system; The algorithm optimization module is used to obtain the real-time operating parameters of the target air-conditioning system, input them into the energy-saving group control strategy, optimize the control actions through the collaborative optimization algorithm, and output the optimal control actions for each device in the system.

10. A storage medium having stored thereon a computer program executable by a processor, characterized in that: When the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to any one of claims 1 to 8 are implemented.

11. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable by the processor, wherein: When the computer program is executed, the steps of the self-organizing construction method of the energy-saving group control strategy of the variable air volume air conditioning system according to any one of claims 1 to 8 are implemented.

12. A variable air volume air conditioning system, characterized in that: include: The electronic device according to claim 11; The actuator is set corresponding to each device in the variable air volume air conditioning system and is used to control the corresponding device based on the optimal control action output by the electronic device.

Images