Fine-tuning control method for central air-conditioning based on double-layer optimization

Abstract

The present invention provides a refined control method for central air conditioners based on double-layer optimization. By establishing a building temperature scale model and a building humidity scale model and formulating a comfort range; establishing an upper-layer model for refined control of central air conditioners, including a central air conditioner control model with the best comfort and a central air conditioner control model with the minimum operating cost; using the Shaple value method to allocate the proportion of comfort and economy of the upper-layer model for refined control of central air conditioners, and calculating the total cooling capacity of the central air conditioner; establishing a lower-layer model for refined control of central air conditioners; using the ant colony algorithm to optimize the double-layer optimization model to obtain a refined control scheme for central air conditioners; this method can effectively reduce the operating energy consumption and improve the economy on the premise of ensuring user comfort, can realize the refined control of central air conditioners, reduce the power consumption during peak hours, and is beneficial to alleviating the contradiction of tight power supply and ensuring the safe operation of the power grid.

Description

Technical Field

[0001] The invention relates to a central air-conditioning fine control method based on double-layer optimization, and belongs to the technical field of air-conditioning load control. Background Art

[0002] As total electricity demand continues to climb, the imbalance between power supply and demand is a major challenge facing the power system, particularly during the peak summer season, when power shortages are significant. According to incomplete statistics, in recent years, summer air conditioning loads in some first-tier cities have exceeded 50% of the total social load, significantly widening the peak-to-valley load differential in summer and further exacerbating the power system's supply-demand imbalance. Air conditioning loads have a certain degree of thermal inertia. By regulating air conditioning loads, power can be reduced quickly while indoor temperatures change slowly over time. This allows for rapid response to grid-side dispatch with minimal impact on user comfort.

[0003] Compared to decentralized standalone air conditioners, central air conditioners are primarily used in public buildings. They offer a larger load capacity and greater adjustability, while adjusting central air conditioners in public buildings does not significantly impact public electricity consumption. Central air conditioner loads are primarily concentrated in winter and summer. Simply increasing the capacity of the transmission and distribution network to meet these brief peaks would require significant financial, material, and human resources.

[0004] Existing research on central air conditioning's participation in demand-side response mostly focuses on the top-level design of air conditioning's participation in grid peak regulation and air conditioning control at the grid level. Central air conditioning control is mostly extensive, for example, on-off control and periodic start-stop control, which cannot really pay attention to user comfort, have high operating energy consumption, and pay less attention to the coordinated control between the various subsystems within the central air conditioning. The control accuracy is poor and cannot truly release the regulation potential of central air conditioning.

[0005] The above problems are issues that should be considered and solved in the process of fine-grained control of central air conditioning based on double-layer optimization. Summary of the Invention

[0006] The purpose of the present invention is to provide a central air-conditioning fine control method based on double-layer optimization to solve the problems in the prior art of using extensive control, not really paying attention to the user's comfort, and requiring reduction of operating energy consumption.

[0007] The technical solution of the present invention is:

[0008] A method for fine-tuning central air conditioning based on double-layer optimization includes the following steps:

[0009] S1. Collect user satisfaction data on indoor environment comfort, including the optimal indoor comfort temperature T c and optimal comfortable humidity Wc , establish a building temperature scale model and a building humidity scale model and formulate a comfort range;

[0010] S2. Establish an upper-layer model for refined control of central air-conditioning, including a central air-conditioning control model with the best comfort and a central air-conditioning control model with the minimum operating cost;

[0011] S3. Use the Shapley value method, i.e., the Shapley value method, to allocate the proportion of comfort and economy of the upper-layer model for refined control of central air-conditioning, and calculate the total central air-conditioning cooling capacity;

[0012] S4. Establish a lower-layer model for refined control of central air-conditioning, including a load control model for the terminal fan subsystem, a load control model for the chilled water pump subsystem, a load control model for the chiller subsystem, and a load control model for the cooling water pump subsystem;

[0013] S5. Establish a two-layer optimization model composed of the upper-layer model and the lower-layer model for refined control of central air-conditioning, and use the ant colony algorithm to optimize the two-layer optimization model to obtain a refined control scheme for central air-conditioning.

[0014] Furthermore, in step S1, to establish a building temperature scale model and a building humidity scale model, specifically,

[0015] S11. Establish the building temperature scale model as: λ T_PMV = β1(T in - T c ), where λ T_PMV is the building temperature scale, T in is the indoor temperature, and β1 is the temperature coefficient;

[0016] S12. Establish the building humidity scale model as: λ W_PMV = β2|m*T in |(W in - W c ), where λ W_PMV is the building humidity scale, W in is the indoor humidity, m is the influence factor of temperature on humidity, representing the influence degree of different temperatures on humidity comfort in different building environments, and β2 is the humidity coefficient.

[0017] Furthermore, in step S1, the comfort range is formulated as λ T_PMV ∈[-2, 2] and λ W_PMV ∈[-2, 2], λ T_PMV =-2 and λ W_PMV =-2 are the lowest comfort levels of the indoor environment, λ T_PMV =-0 and λ W_PMV =-0 are the best comfort levels of the indoor environment, λT_PMV = 2 and λ W_PMV = 2 represents the highest comfort level in the indoor environment, and λ T_PMV , λ W_PMV are all dimensionless constant values.

[0018] Furthermore, in step S2, an upper-layer model for refined control of the central air conditioner is established, including a central air conditioner control model with the best comfort level and a central air conditioner control model with the minimum operating cost. Specifically,

[0019] Establish an upper-layer model for refined control of the central air conditioner:

[0020] minC = Y1C c + Y2C m

[0021] where C is the upper-layer objective function, Y1 and Y2 are the economic distribution coefficient and the comfort distribution coefficient respectively, C c is the economic objective function, and C m is the comfort objective function;

[0022] The central air conditioner control model with the best comfort level is:

[0023]

[0024] where T set is the set temperature, W set is the set humidity, γ1 and γ2 are the comfort temperature coefficient and the comfort humidity coefficient respectively, e is the exponential function, β1 is the temperature coefficient, m is the temperature influence factor on humidity, and β2 is the humidity coefficient;

[0025] The central air conditioner control target model with the minimum operating cost is:

[0026]

[0027] where γ3 and γ4 are the economic temperature coefficient and the economic humidity coefficient respectively.

[0028] The constraint conditions are:

[0029]

[0030] where T max , T min are the upper and lower limits of the set temperature respectively, and W max , W min are the upper and lower limits of the set humidity respectively.

[0031] Furthermore, in step S3, the Shapley value method is used to allocate the proportion of comfort and economy in the upper-layer model for refined control of the central air conditioner. Specifically,

[0032] The objective function of the upper layer model for the refined control of central air conditioning includes two sub-objective members: the central air conditioning control objective model with the best comfort and the central air conditioning control objective model with the minimum operating cost. Different sub-objectives form different coalitions s. When sub-objective member i participates in the objective function of the upper layer model, there are (|s|-1)! permutation methods. Among them, |s| is the number of sub-objectives included in coalition s, and the remaining (n-|s|) members have (n-|s|)! permutation orders. The distribution coefficient of the sub-objective to the objective function of the upper layer model for the refined control of central air conditioning is:

[0033]

[0034] where N is the number of sets, n is the number of members, y(s) is the characteristic function of coalition s, representing the maximum benefit achieved by coalition s through the mutual cooperation of each objective, and y(s / i) is the benefit obtained after removing the main body i from subset s.

[0035] Furthermore, in step S3, calculate the total central air conditioning cooling capacity, specifically,

[0036] Q th (t) = Q S (t) + Q L (t) = (T out (t) - T set (t))G a + Q m (t) + Q s (t) + k(W in - W set )

[0037] where Q th (t) is the total cooling capacity, Q S (t) is the sensible heat, Q L (t) is the latent heat, T out is the outdoor temperature, G a is the building thermal conductivity, Q m (t) is the heat dissipated by people and equipment, Q s (t) is the solar radiation heat, k is the vaporization heat conductivity coefficient, t represents the time, W in is the indoor humidity, W set is the set temperature.

[0038] Furthermore, in step S4, establish the lower layer model for the refined control of central air conditioning, including the load control model of the terminal fan subsystem, the load control model of the chilled water pump subsystem, the load control model of the chiller subsystem, and the load control model of the cooling water pump subsystem. Specifically,

[0039] The lower layer model for the refined control of central air conditioning is established as:

[0040] minP = min(P s , P q , P l , P d )

[0041] where P is the lower - level objective function, P s is the load of the terminal fan subsystem, P q is the load of the chilled water pump subsystem, P l is the load of the chiller subsystem, P d is the load of the cooling water pump subsystem;

[0042] The load regulation model of the terminal fan subsystem is:

[0043]

[0044] where G n , G s are the rated air supply volume and the actual air supply volume respectively, f n , f s are the rated operating frequency and the actual operating frequency of the terminal fan respectively, P n is the rated operating power of the terminal fan;

[0045] The load regulation model of the chilled water pump subsystem is:

[0046]

[0047] where W q.i is the actual water inflow of the i - th chilled water pump, H q.i is the actual head of the i - th chilled water pump, ρ is the water density, g is the acceleration due to gravity, n q.i is the working point efficiency of the i - th chilled water pump;

[0048] The load regulation model of the chiller subsystem is:

[0049] minP l = a0 + a1(t cwj - t w1 ) + a2(t cwj - t w1 ) 2 + a3(t cwj - t w1 )Q th + a4Q th + a5Q th 2

[0050] where t w1 is the chilled water inlet temperature, t cwjis the inlet water temperature of the cooling water, and a0, a1, a2, a3, a4, a5 are the energy consumption model coefficients respectively, and Q th is the total refrigeration capacity;

[0051] The load regulation model of the cooling water pump subsystem is:

[0052]

[0053] where W d.i is the actual water intake of the i-th cooling water pump, and H d.i is the actual head of the i-th cooling water pump, ρ is the water density, g is the acceleration of gravity, and n d.i is the working point efficiency of the i-th cooling water pump;

[0054] The constraint conditions are:

[0055]

[0056] where G max and G min are the upper and lower limits of the air supply volume respectively, and f i.s is the actual operating frequency of equipment i, f i.max and f i.min are the upper and lower limits of the operating frequency of equipment i respectively, t w2 is the outlet water temperature of the chilled water, are the upper and lower limits of the chilled water inlet and outlet temperatures respectively, are the upper and lower limits of the cooling water inlet temperature respectively, are the upper and lower limits of the head of the i-th chilled water pump respectively, are the upper and lower limits of the head of the i-th cooling water pump respectively, are the upper and lower limits of the chilled water flow rate of the i-th chilled water pump respectively, are the upper and lower limits of the cooling water flow rate of the i-th cooling water pump respectively.

[0057] Furthermore, in step S5, the ant colony algorithm is used to optimize the double-layer optimization model to obtain a refined control scheme for the central air conditioner. Specifically,

[0058] S51. Initialize the population size of the ant colony and initialize relevant parameters, including the heuristic information factor and the pheromone evaporation coefficient;

[0059] S52. Initialize the pheromone matrix. A set of temperature and humidity combinations that meet the comfort interval in the upper layer model of the refined control of the central air conditioner is a path, and a set of equipment load values that meet the equipment operation constraints in the lower layer model of the refined control of the central air conditioner is a path. Assign the initial pheromone values to the paths of each set of temperature and humidity combinations and the paths of each set of equipment load values;

[0060] S53. Randomly initialize the positions of each ant and record the information in their respective tabu lists;

[0061] S54. Iteratively calculate for each ant, and calculate the next visited node according to the state transition probability;

[0062] S55. Update the information in their respective tabu lists;

[0063] S56. Determine whether the tabu list is full. If not, return to step S54; if so, continue to the next step S57;

[0064] S57. Calculate the current upper-layer objective function value and lower-layer objective function value, and update the global pheromone according to the pheromone;

[0065] S58. Determine whether the current situation has fallen into a local optimum. If so, continue to execute step S59; if not, go to step S510;

[0066] S59. Use the variable neighborhood idea to update the local pheromone, and continue with step S510;

[0067] S510. Determine whether the termination condition is met, that is, the maximum number of ant colony iterations is reached. If not, jump to step S53; if so, the program ends and a refined control plan for the central air conditioner is obtained.

[0068] Furthermore, in step S5, the obtained refined control plan for the central air conditioner includes adjusting the air volume of the terminal fan, adjusting the flow rate of the operating chilled water pump, shutting down some chilled water pumps and adjusting the flow rate of the operating chilled water pump, adjusting the chilled water temperature of the chiller, putting some chillers on standby and adjusting the chilled water temperature of the operating chiller, adjusting the flow rate of the operating cooling water pump, and shutting down some cooling water pumps and adjusting the flow rate of the operating cooling water pump.

[0069] The beneficial effects of the present invention are as follows: This refined control method for central air conditioners based on double-layer optimization can effectively reduce the operating energy consumption and improve the economy through a double-layer optimization model on the premise of ensuring user comfort. It can achieve refined control of central air conditioners, reduce power consumption during peak hours, and is conducive to alleviating the contradiction of tight power supply and ensuring the safe operation of the power grid. Brief Description of the Drawings

[0070] Figure 1 is a flowchart of the refined control method for central air conditioners based on double-layer optimization according to an embodiment of the present invention;

[0071] Figure 2 is an explanatory schematic diagram of the double-layer optimization model in the embodiment;

[0072] Figure 3 [[ID=4.]]is a flowchart of optimizing the double-layer optimization model by using the ant colony algorithm in the embodiment;

[0073] Figure 4 It is a schematic diagram for explaining the refined control scheme of the central air conditioner in the embodiment. Specific implementation manner

[0074] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0075] Embodiment

[0076] A refined control method for a central air conditioner based on double-layer optimization, as Figure 1 , includes the following steps,

[0077] S1. Collect data on the user's satisfaction with the indoor environment comfort, including the indoor optimal comfort temperature T c and the optimal comfort humidity W c , establish a building temperature scale model and a building humidity scale model and formulate a comfort interval;

[0078] In step S1, establishing a building temperature scale model and a building temperature scale model, specifically,

[0079] S11. Establish the building temperature scale model as: λ T_PMV =β1(T in -T c ), where λ T_PMV is the building temperature scale, T in is the indoor temperature, and β1 is the temperature coefficient;

[0080] S12. Establish the building humidity scale model as: λ W_PMV =β2|m*T in |(W in -W c ), where λ W_PMV is the building humidity scale, W in is the indoor humidity, m is the temperature influence factor on humidity, representing the influence degree of different temperatures on humidity comfort in different building environments, and β2 is the humidity coefficient.

[0081] In step S1, the comfort interval is formulated as λ T_PMV ∈[-2, 2] and λ W_PMV ∈[-2, 2], λ T_PMV =-2 and λ W_PMV =-2 are the lowest comfort degrees of the indoor environment, λ T_PMV =-0 and λ W_PMV =-0 are the best comfort degrees of the indoor environment, λ T_PMV =2 and λ W_PMV =2 are the highest comfort degrees of the indoor environment, λ T_PMV , λ W_PMV are all dimensionless constant values.

[0082] S2. Establish the upper-layer model for the refined control of central air-conditioning, including the central air-conditioning control model with the best comfort and the central air-conditioning control model with the minimum operating cost. Specifically,

[0083] Establish the upper-layer model for the refined control of central air-conditioning:

[0084] minC = Y1C c +Y2C m

[0085] where C is the upper-layer objective function, Y1 and Y2 are the economic distribution coefficient and the comfort distribution coefficient respectively, C c is the economic objective function, and C m is the comfort objective function;

[0086] The central air-conditioning control model with the best comfort is:

[0087]

[0088] where T set is the set temperature, W set is the set humidity, γ1 and γ2 are the comfort temperature coefficient and the comfort humidity coefficient respectively, e is the exponential function, β1 is the temperature coefficient, m is the influence factor of temperature on humidity, and β2 is the humidity coefficient;

[0089] The central air-conditioning control objective model with the minimum operating cost is:

[0090]

[0091] where γ3 and γ4 are the economic temperature coefficient and the economic humidity coefficient respectively.

[0092] The constraint conditions are:

[0093]

[0094] where T max and T min are the upper and lower limits of the set temperature respectively, and W max and W min are the upper and lower limits of the set humidity respectively.

[0095] S3. Use the Shapley value method, i.e., the Shapley value method, to allocate the proportion of comfort and economy in the upper-layer model of the refined control of central air-conditioning. Specifically,

[0096] The objective function of the upper layer model for the refined control of the central air conditioning includes two sub-objective members: the central air conditioning control objective model with the best comfort and the central air conditioning control objective model with the minimum operating cost. Different sub-objectives form different coalitions s. When the sub-objective member i participates in the objective function of the upper layer model, there are (|s|-1)! permutation methods. Among them, |s| is the number of sub-objectives included in the coalition s, and the remaining (n-|s|) members have (n-|s|)! permutation orders. The distribution coefficient of the sub-objective to the objective function of the upper layer model for the refined control of the central air conditioning is:

[0097]

[0098] Among them, N is the number of sets, n is the number of members, y(s) is the characteristic function of the coalition s, representing the maximum benefit achieved by the coalition s through the mutual cooperation of each objective, and y(s / i) is the benefit obtained after removing the main body i from the subset s.

[0099] In step S3, calculate the total central air conditioning cooling capacity. Specifically,

[0100] Q th (t) = Q S (t) + Q L (t) = (T out (t) - T set (t))G a + Q m (t) + Q s (t) + k(W in - W set )

[0101] Among them, Q th (t) is the total cooling capacity, Q S (t) is the sensible heat, Q L (t) is the latent heat, T out is the outdoor temperature, G a is the building thermal conductivity, Q m (t) is the heat dissipated by personnel and equipment, Q s (t) is the solar radiation heat, k is the vaporization heat conductivity coefficient, t represents the time, W in is the indoor humidity, W set is the set temperature.

[0102] S4. Establish the lower layer model for the refined control of the central air conditioning, including the load control model of the terminal fan subsystem, the load control model of the chilled water pump subsystem, the load control model of the chiller subsystem, and the load control model of the cooling water pump subsystem; specifically,

[0103] The lower layer model established for the refined control of the central air conditioning is:

[0104] min P = min(P s , P q , P l , P d )

[0105] where P is the lower - layer objective function, P[[ID=1}} s is the load of the terminal fan subsystem, P q is the load of the chilled water pump subsystem, P l is the load of the chiller subsystem, P d is the load of the cooling water pump subsystem;

[0106] The load regulation model of the terminal fan subsystem is:

[0107]

[0108] where G n , G s are the rated air supply volume and the actual air supply volume respectively, f n , f s are the rated operating frequency and the actual operating frequency of the terminal fan respectively, P n is the rated operating power of the terminal fan;

[0109] The load regulation model of the chilled water pump subsystem is:

[0110]

[0111] where W q.i is the actual water inflow of the i - th chilled water pump, H q.i is the actual head of the i - th chilled water pump, ρ is the water density, g is the acceleration of gravity, n q.i is the working - point efficiency of the i - th chilled water pump;

[0112] The load regulation model of the chiller subsystem is:[[ID=}}}

[0113] minP l = a0 + a1(t cwj - t w1 ) + a2(t cwj - t w1 ) 2 + a3(t cwj - t w1 )Q th + a4Q th + a5Q th 2

[0114] where t w1 is the chilled water inlet temperature, t cwjis the inlet water temperature of the cooling water, and a0, a1, a2, a3, a4, a5 are the energy consumption model coefficients respectively, and Q th is the total refrigeration capacity;

[0115] The load regulation model of the cooling water pump subsystem is:

[0116]

[0117] where, W d.i is the actual water inflow of the i-th cooling water pump, H d.i is the actual head of the i-th cooling water pump, ρ is the water density, g is the acceleration of gravity, and n d.i is the working point efficiency of the i-th cooling water pump;

[0118] The constraint conditions are:

[0119]

[0120] where, G max , G min are the upper and lower limits of the air supply volume respectively, f i.s is the actual operating frequency of equipment i, f i.max , f i.min are the upper and lower limits of the operating frequency of equipment i respectively, t w2 is the outlet water temperature of the chilled water, are the upper and lower limits of the chilled water inlet and outlet temperatures respectively, are the upper and lower limits of the cooling water inlet temperature respectively, are the upper and lower limits of the head of the i-th chilled water pump respectively, are the upper and lower limits of the head of the i-th cooling water pump respectively, are the upper and lower limits of the chilled water flow rate of the i-th chilled water pump respectively, are the upper and lower limits of the cooling water flow rate of the i-th cooling water pump respectively.

[0121] S5. Establish a two-layer optimization model composed of the upper-layer model and the lower-layer model of the refined control of the central air conditioner, and use the ant colony algorithm to optimize the two-layer optimization model, such as Figure 2 and Figure 3 , to obtain the refined control scheme of the central air conditioner, such as Figure 4 .

[0122] S51. Initialize the population size of the ant colony and initialize the relevant parameters, including the heuristic information factor and the pheromone evaporation coefficient;

[0123] S52. Initialize the pheromone matrix. In the upper-layer model of the refined regulation of central air conditioning, a set of temperature and humidity combinations within the comfort range is regarded as a path, and in the lower-layer model of the refined regulation of central air conditioning, a set of equipment load values that meet the equipment operation constraints is regarded as a path. Initialize the initial value of pheromone for each path of the temperature and humidity combination group and each path of the equipment load value group respectively.

[0124] S53. Randomly initialize the position of each ant and record it in its respective taboo list information.

[0125] S54. Iteratively calculate for each ant, and calculate the next visited node according to the state transition probability.

[0126] S55. Update their respective taboo list information.

[0127] S56. Determine whether the taboo list is full. If not, return to step S54; if so, continue to the next step S57.

[0128] S57. Calculate the current upper-layer objective function value and lower-layer objective function value, and update the global pheromone according to the pheromone.

[0129] S58. Determine whether the current situation has fallen into a local optimum. If so, continue to execute step S59; if not, go to step S510.

[0130] S59. Use the variable neighborhood idea to update the local pheromone and continue step S510.

[0131] S510. Determine whether the termination condition is met, that is, the maximum number of ant colony iterations is reached. If not, jump to step S53; if so, the program ends and the refined regulation plan of the central air conditioning is obtained.

[0132] In step S5, the obtained refined regulation plan of the central air conditioning includes adjusting the air volume of the terminal fan, adjusting the flow rate of the operating chilled water pump, shutting down some chilled water pumps and adjusting the flow rate of the operating chilled water pump, adjusting the chilled water temperature of the chiller, putting some chillers on standby and adjusting the chilled water temperature of the operating chiller, adjusting the flow rate of the operating cooling water pump, and shutting down some cooling water pumps and adjusting the flow rate of the operating cooling water pump.

[0133] This refined regulation method of central air conditioning based on double-layer optimization can effectively reduce the operating energy consumption and improve the economy on the premise of ensuring user comfort through the double-layer optimization model. It can achieve the refined regulation of central air conditioning, reduce the power consumption during peak hours, and is conducive to alleviating the contradiction of tight power supply and ensuring the safe operation of the power grid.

[0134] The refined control method of central air conditioning based on double - layer optimization first establishes a building humidity - heat scale model according to the data of user satisfaction with the indoor environment comfort in public buildings. Secondly, it establishes an upper - layer model for the refined control of central air conditioning that meets comfort requirements and has the minimum operating cost, and uses the Shapley value method to allocate the proportion of comfort and economy in the upper - layer model. Then, it establishes a lower - layer model for the refined control of central air conditioning, including the control models of the central air - conditioning subsystems of the terminal fan, chilled - water pump, chiller, and chilled - water pump. Finally, the ant colony algorithm is used to optimize the double - layer optimization model to obtain a refined control scheme for central air conditioning. This method can fully explore the potential of central air - conditioning control, reduce the operating energy consumption, achieve refined control of each subsystem within the central air - conditioning system while ensuring user comfort, and provide theoretical and practical value for the actual engineering application of the central air - conditioning system to participate in demand - side response.

[0135] The refined control method of central air conditioning based on double - layer optimization adopts a double - layer control strategy. Since the load controllability of central air conditioning is large, it can achieve refined control of central air conditioning, effectively reduce the operating energy consumption of central air conditioning, and regulating the central air - conditioning load is of great significance for changing the user load curve and realizing peak shaving and valley filling of the power grid. It can be promoted to on - site engineering applications and has high engineering practice value.

[0136] Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A refined control method for central air conditioners based on double-layer optimization, characterized in that: It includes the following steps: S1. Collect data on users' satisfaction with the indoor environment comfort, including the optimal indoor comfort temperature T c and the optimal comfort humidity W c , establish a building temperature scale model and a building humidity scale model and formulate a comfort range; In step S1, a building temperature scale model and a building humidity scale model are established. Specifically, S11. Establish the building temperature scale model as: λ T_PMV = β1(T in - T c ), where λ T_PMV is the building temperature scale, T in is the indoor temperature, and β1 is the temperature coefficient; S12. Establish the building humidity scale model as: λ W_PMV = β2|m * T in |(W in - W c ), where λ W_PMV is the building humidity scale, W in is the indoor humidity, m is the temperature influence factor on humidity, representing the influence degree of different temperatures on humidity comfort in different building environments, and β2 is the humidity coefficient; S2. Establish an upper-layer model for refined control of the central air conditioner, including a central air conditioner control model with the best comfort and a central air conditioner control model with the minimum operating cost; specifically, Establish an upper-layer model for refined control of the central air conditioner: minC = Y1C c + Y2C m Among them, C is the upper-level objective function, Y1 and Y2 are the economic distribution coefficient and the comfort distribution coefficient respectively, C c is the economic objective function, C m is the comfort objective function; The central air conditioner control model with the best comfort is: Among them, T set is the set temperature, W set is the set temperature, γ1 and γ2 are the comfort temperature coefficient and the comfort humidity coefficient respectively, e is the exponential function, β1 is the temperature coefficient, m is the temperature influence factor on humidity, and β2 is the humidity coefficient; The central air conditioner control target model with the minimum operating cost is: Among them, γ3 and γ4 are the economic temperature coefficient and the economic humidity coefficient respectively; The constraint conditions are: Among them, T max and T min are the upper and lower limits of the set temperature respectively; W max and W min are the upper and lower limits of the set humidity respectively. S3. Use the Shapley value method, that is, the Shapley value method, to allocate the proportion of comfort and economy in the upper-layer model of refined control of the central air conditioner, and calculate the total central air conditioner cooling capacity; S4. Establish a lower-layer model for refined control of the central air conditioner, including a load control model for the terminal fan subsystem, a load control model for the chilled water pump subsystem, a load control model for the chiller subsystem, and a load control model for the cooling water pump subsystem; S5. Establish a two-layer optimization model composed of the upper-layer model and the lower-layer model for refined control of the central air conditioner, and use the ant colony algorithm to optimize the two-layer optimization model to obtain a refined control scheme for the central air conditioner.

2. The refined control method for central air-conditioning based on double-layer optimization according to claim 1, characterized in that: In step S1, a comfort interval is defined as λ T_PMV ∈[-2, 2] and λ W_PMV ∈[-2, 2], λ T_PMV =-2 and λ W_PMV =-2 represents the lowest indoor environmental comfort level, λ T_PMV =-0 and λ W_PMV =-0 represents the optimal indoor environmental comfort level, λ T_PMV =2 and λ W_PMV =2 represents the highest indoor environmental comfort level, λ T_PMV and λ W_PMV are both dimensionless constant values.

3. The refined control method for central air conditioners based on double-layer optimization according to claim 1, characterized in that: In step S3, use the Shapley value method to allocate the proportion of comfort and economy in the upper-layer model of refined control of the central air conditioner. Specifically, The objective function of the upper-layer model for refined control of the central air conditioner includes two sub-objective members: the central air conditioner control target model with the best comfort and the central air conditioner control target model with the minimum operating cost. Different sub-objectives form different coalitions s. When the sub-objective member i participates in the objective function of the upper-layer model, there are (|s|-1)! permutation methods. Among them, |s| is the number of sub-objectives included in the coalition s, and the remaining (n-|s|) members have (n-|s|)! permutation orders. The distribution coefficient of the sub-objective to the objective function of the upper-layer model for refined control of the central air conditioner is: Among them, N is the number of sets, n is the number of members, y(s) is the characteristic function of the coalition s, representing the maximum benefit achieved by the coalition s through the mutual cooperation of each objective, and y(s / i) is the benefit obtained after removing the main body i from the subset s.

4. The refined control method for central air-conditioning based on double-layer optimization according to claim 1, characterized in that: In step S3, calculate the total central air conditioner cooling capacity. Specifically, Q th q(t) = Q S q(t) + Q L q(t) = (T out (t) - T set (t))G a + Q m q(t) + Q s q(t) + k(W in - W set ) Among them, Q th (t) is the total cooling capacity, Q S (t) is the sensible heat, Q L (t) is the latent heat, T out is the outdoor temperature, G a is the building thermal conductivity, Q m (t) is the heat dissipated by people and equipment, Q s (t) is the solar radiation heat, k is the vaporization heat conductivity coefficient, t represents the time, W in is the indoor humidity, W set is the set temperature.

5. The refined control method for central air conditioning based on double-layer optimization according to any one of claims 1-4, characterized in that: In step S4, establish a lower-layer model for refined control of the central air conditioner, including a load control model for the terminal fan subsystem, a load control model for the chilled water pump subsystem, a load control model for the chiller subsystem, and a load control model for the cooling water pump subsystem. Specifically, Establish a lower-layer model for refined control of the central air conditioner: minP = min(P s , P q , P l , P d ) Among them, P is the lower-layer objective function, P s is the load of the terminal fan subsystem, P q is the load of the chilled water pump subsystem, P l is the load of the chiller subsystem, P d is the load of the cooling water pump subsystem; The load control model for the terminal fan subsystem is: Among them, G n and G s are the rated air supply volume and the actual air supply volume respectively, f n and f s are the rated operating frequency and the actual operating frequency of the terminal fan respectively, and P n is the rated operating power of the terminal fan; The load control model for the chilled water pump subsystem is: Among them, W q.i is the actual water inflow of the i-th chilled water pump, H q.i is the actual head of the i-th chilled water pump, ρ is the water density, g is the acceleration of gravity, n q.i is the working point efficiency of the i-th chilled water pump; The load control model for the chiller subsystem is: minP l = a0 + a1(t cwj - t w1 ) + a2(t cwj - t w1 ) 2 + a3(t cwj - t w1 )Q th + a4Q th + a5Q th 2 Among them, t w1 is the inlet temperature of chilled water, and t cwj is the inlet temperature of cooling water. a0, a1, a2, a3, a4, and a5 are the energy consumption model coefficients respectively, and Q th is the total refrigeration capacity; The load control model for the cooling water pump subsystem is: Among them, W d.i is the actual water intake of the i-th cooling water pump, H d.i is the actual head of the i-th cooling water pump, ρ is the water density, g is the acceleration due to gravity, n d.i is the working point efficiency of the i-th cooling water pump; The constraint conditions are: Among them, G max and G min are the upper and lower limits of the air supply volume respectively, f i.s is the actual operating frequency of equipment i, f i.max and f i.min are the upper and lower limits of the operating frequency of equipment i respectively, t w2 is the chilled water outlet temperature, are the upper and lower limits of the chilled water inlet and outlet temperatures respectively, are the upper and lower limits of the cooling water inlet temperature respectively, are the upper and lower limits of the head of the i-th chilled water pump respectively, are the upper and lower limits of the head of the i-th cooling water pump respectively, are the upper and lower limits of the flow rate of the i-th chilled water respectively, are the upper and lower limits of the flow rate of the i-th cooling water respectively.

6. The refined control method for central air conditioning based on double-layer optimization according to any one of claims 1-4, characterized in that: In step S5, use the ant colony algorithm to optimize the two-layer optimization model to obtain a refined control scheme for the central air conditioner. Specifically, S51. Initialize the population size of the ant colony and initialize relevant parameters, including the heuristic information factor and the pheromone evaporation coefficient; S52. Initialize the pheromone matrix. A set of temperature and humidity combinations that meet the comfort range in the upper layer model of the refined control of central air conditioning is regarded as a path, and a set of equipment load values that meet the equipment operation constraints in the lower layer model of the refined control of central air conditioning is regarded as a path. Initialize the initial value of pheromone for each path of temperature and humidity combination and each path of equipment load value respectively; S53. Randomly initialize the position of each ant and record it in their respective taboo list information; S54. Iteratively calculate for each ant, and calculate the next visited node according to the state transition probability; S55. Update their respective taboo list information; S56. Determine whether the taboo list is full. If not, return to step S54; if so, continue to the next step S57; S57. Calculate the current upper layer objective function value and lower layer objective function value, and update the global pheromone according to the pheromone; S58. Judge whether the current situation has fallen into local optimum. If so, continue to execute step S59; if not, go to step S510; S59. Update the local pheromone using the variable neighborhood idea and continue step S510; S510. Determine whether the termination condition is met, that is, the maximum number of ant colony iterations is reached. If not, jump to step S53; if so, the program ends and the refined control plan for central air conditioning is obtained.

7. The refined control method for central air conditioning based on double-layer optimization according to any one of claims 1-4, characterized in that: In step S5, the obtained refined control plan for central air conditioning includes adjusting the air volume of the terminal fan, adjusting the flow rate of the operating chilled water pump, shutting down some chilled water pumps and adjusting the flow rate of the operating chilled water pump, adjusting the chilled water temperature of the chiller, putting some chillers on standby and adjusting the chilled water temperature of the operating chiller, adjusting the flow rate of the operating cooling water pump, and shutting down some cooling water pumps and adjusting the flow rate of the operating cooling water pump.

Images