Operation control method, training method, device and equipment of operation control model

Abstract

The application provides an operation control method, a training method and device of an operation control model, and equipment. The control method comprises: obtaining operation data of each node device in an HVAC system, and obtaining a disturbance parameter; determining at least two correction operation modes according to the disturbance parameter, a basic operation mode and a preset HVAC system operation mechanism function, and determining a correction system operation efficiency cop value corresponding to each correction operation mode; determining a control mode of the HVAC system; if the control mode is a first control mode, determining that an operation mode with the highest correction cop value is an optimal operation mode; if the control mode is a second control mode, obtaining a target cop value by using a pre-trained operation control model; determining that an operation mode with the smallest difference between the correction cop value and the target cop value is the optimal operation mode; and controlling the HVAC system to operate according to the optimal operation mode. The application takes into account high operation efficiency, and reduces consumption of energy resources.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of operation control of heating, ventilation and air conditioning, and particularly relate to an operation control method, a training method and device of an operation control model, and equipment. BACKGROUND

[0002] Heating, ventilation and air conditioning is an air conditioner with heating, ventilation and air conditioning functions. Since the main functions of heating, ventilation and air conditioning include heating, ventilation and air conditioning, which are abbreviated as HVAC (Heating, Ventilating and Air Conditioning), the three functions are collectively referred to as heating, ventilation and air conditioning.

[0003] During the operation of the HVAC system, the HVAC system will be disturbed by external state quantities outside the system (such as outdoor temperature, relative humidity, etc.). For example, changes in outdoor temperature and / or outdoor humidity will cause changes in the cooling effect of the cooling tower and the refrigerating capacity in the HVAC system, thereby interfering with the normal operation of the HVAC system and forming a disturbance to the HVAC system.

[0004] The existing operation control method often ignores the influence on the system operation efficiency cop (Coefficient of Performance, also known as performance coefficient) when offsetting the disturbance, resulting in the problem that although the disturbance is offset, the operation efficiency is severely reduced. SUMMARY

[0005] Embodiments of the present application provide an operation control method, a training method and device of an operation control model, and equipment to solve the problem that the existing operation control method often ignores the influence on the system operation efficiency cop when offsetting the disturbance, resulting in the problem that although the disturbance is offset, the operation efficiency is severely reduced. To solve the above technical problems, the present application is implemented as follows:

[0006] In a first aspect, embodiments of the present application provide an operation control method of a heating, ventilation and air conditioning HVAC system, comprising:

[0007] obtaining operation data of each node device in the HVAC system, and obtaining a disturbance parameter; wherein the disturbance parameter is an external state quantity outside the HVAC system that can form a disturbance to the HVAC system;

[0008] determining a basic operation mode of the HVAC system according to the operation data;

[0009] determine at least two corrected operation modes of the HVAC system according to the disturbance variable, the basic operation mode and a preset HVAC system operation mechanism function, and determine a corrected system operation efficiency cop value of the HVAC system corresponding to each of the corrected operation modes, wherein the corrected operation mode is an operation mode capable of offsetting the disturbance formed on the HVAC system by the disturbance variable;

[0010] determine a control mode of the HVAC system;

[0011] if the control mode is a first control mode, determine an operation mode with the highest corrected cop value among all the corrected operation modes as an optimal operation mode;

[0012] if the control mode is a second control mode, obtain a target cop value of the HVAC system according to the operation data by using a pre-trained operation control model, and determine an operation mode with the smallest difference between the corrected cop value and the target cop value among all the corrected operation modes as the optimal operation mode;

[0013] control the HVAC system to operate in the optimal operation mode.

[0014] Optionally,

[0015] the disturbance variable comprises at least one of the following: an outdoor temperature change amount, an outdoor humidity change amount.

[0016] Optionally,

[0017] the node device comprises: a cooling side device, a freezing side device and a refrigeration host.

[0018] In a second aspect, an embodiment of the present application provides a training method of an operation control model, comprising:

[0019] obtain a historical operation data set of each node device in a heating ventilation air conditioning (HVAC) system;

[0020] extract historical operation data associated with a system operation efficiency cop value of the HVAC system from the historical operation data set to obtain a first operation data set according to a preset HVAC system operation mechanism function, and determine an influence coefficient of each first operation data in the first operation data set on the cop value according to the HVAC system operation mechanism function;

[0021] generate a weight value according to the influence coefficient, and label the first operation data according to the weight value, and obtain a second operation data set by collecting all the labeled first operation data;

[0022] train a control model by using the second set of operation data, to obtain the operation control model.

[0023] Optionally,

[0024] use the HVAC system operation mechanism function as a target function for training the control model.

[0025] Optionally,

[0026] The operation control model is an equivalent model.

[0027] In a third aspect, an embodiment of the present application provides an operation control device of a heating ventilation air conditioning (HVAC) system, which comprises:

[0028] A first obtaining module is configured to obtain operation data of each node device in the HVAC system and obtain a disturbance parameter; the disturbance parameter is an external state variable that can form a disturbance to the HVAC system and is outside the HVAC system.

[0029] A first executing module is configured to determine a basic operation mode of the HVAC system according to the operation data.

[0030] The first executing module is further configured to determine at least two corrected operation modes of the HVAC system and determine corrected system coefficient of performance (cop) values of the HVAC system corresponding to the corrected operation modes according to the disturbance parameter, the basic operation mode and a preset HVAC system operation mechanism function; the corrected operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system.

[0031] The first executing module is further configured to determine a control mode of the HVAC system.

[0032] The first executing module is further configured to, if the control mode is a first control mode, determine an operation mode with the highest corrected cop value in all the corrected operation modes as an optimal operation mode.

[0033] The first executing module is further configured to, if the control mode is a second control mode, obtain a target cop value of the HVAC system according to the operation data by using a pre-trained operation control model; and determine an operation mode with the smallest difference between a corrected cop value and the target cop value in all the corrected operation modes as the optimal operation mode.

[0034] The first executing module is further configured to control the HVAC system to operate in the optimal operation mode.

[0035] In a fourth aspect, an embodiment of the present application provides a training device of an operation control model, which comprises:

[0036] The second obtaining module is configured to obtain a historical operation data set of each node device in the HVAC system.

[0037] The second execution module is configured to extract, according to a preset HVAC system operation mechanism function, historical operation data associated with a system operation efficiency cop value of the HVAC system from the historical operation data set to obtain a first operation data set, and determine an influence coefficient of each first operation data in the first operation data set on the cop value according to the HVAC system operation mechanism function.

[0038] The second execution module is further configured to generate a weight value according to the influence coefficient, label the first operation data according to the weight value, and obtain a second operation data set by collecting all the labeled first operation data.

[0039] The training module is configured to train a control model by using the second operation data set to obtain the operation control model.

[0040] In a fifth aspect, an electronic device is provided, which includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, the steps in the HVAC system operation control method according to any one of the first aspect and the steps in the training method of the operation control model according to any one of the second aspect are implemented.

[0041] In a sixth aspect, a readable storage medium is provided, which stores a program or instructions. When the program or instructions are executed by a processor, the steps in the HVAC system operation control method according to any one of the first aspect and the steps in the training method of the operation control model according to any one of the second aspect are implemented.

[0042] In the embodiment of the present application, operation data of each node device in the HVAC system is acquired, and a disturbance parameter is acquired; wherein the disturbance parameter is an external state variable outside the HVAC system and capable of forming a disturbance to the HVAC system; a basic operation mode of the HVAC system is determined according to the operation data; at least two correction operation modes of the HVAC system are determined according to the disturbance parameter, the basic operation mode and a preset HVAC system operation mechanism function, and a correction system operation efficiency cop value of the HVAC system corresponding to each correction operation mode is determined; wherein the correction operation mode is an operation mode capable of offsetting the disturbance formed by the disturbance parameter to the HVAC system; a control mode of the HVAC system is determined; if the control mode is a first control mode, an operation mode with the highest correction cop value in all the correction operation modes is determined as an optimal operation mode; if the control mode is a second control mode, a target cop value of the HVAC system is obtained according to the operation data by using a pre-trained operation control model; an operation mode with the smallest difference between the correction cop value and the target cop value in all the correction operation modes is determined as the optimal operation mode, and the HVAC system is controlled to operate according to the optimal operation mode, the optimal operation mode is determined by comparing cop values (the operation mode with the highest correction cop value or the operation mode with the smallest difference between the correction cop value and the target cop value), high operation efficiency of the HVAC system is considered in the case of offsetting the disturbance, high operation efficiency of the HVAC system is realized, and consumption of energy resources is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a better understanding of the preferred embodiments, and are not to be considered limitations of the present application. Furthermore, like reference numerals refer to like elements throughout the drawings. In the drawings:

[0044] Figure 1 A flowchart of an operation control method of an HVAC system according to an embodiment of the present application;

[0045] Figure 2 A p-h diagram of a vapor compression heat pump cycle process;

[0046] Figure 3 A schematic diagram of a cooling tower working principle;

[0047] Figure 4 A flowchart of an operation control method of an HVAC system according to an embodiment of the present application;

[0048] Figure 5A control system architecture schematic diagram for applying the operation control method of the embodiment of the present application is shown in FIG. 1.

[0049] Figure 6 A flowchart of the training method of the operation control model of the embodiment of the present application is shown in FIG. 6.

[0050] Figure 7 A flowchart of the training method of the operation control model of the embodiment of the present application is shown in FIG. 6.

[0051] Figure 8 A principle block diagram of the operation control device of the HVAC system of the embodiment of the present application is shown in FIG. 7.

[0052] Figure 9 A principle block diagram of the training device of the operation control model of the embodiment of the present application is shown in FIG. 8.

[0053] Figure 10 A principle block diagram of the electronic device of the embodiment of the present application is shown in FIG. 9. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0055] The embodiment of the present application provides an operation control method of an HVAC system, as shown in FIG. 1. Figure 1 Figure 1 A flowchart of the operation control method of the HVAC system of the embodiment of the present application is shown in FIG. 2. The operation control method comprises the following steps.

[0056] Step 11: obtaining operation data of each node device in the HVAC system, and obtaining a disturbance parameter; wherein the disturbance parameter is an external state quantity that can form a disturbance to the HVAC system and is outside the HVAC system;

[0057] Step 12: determining a basic operation mode of the HVAC system according to the operation data;

[0058] Step 13: determining at least two correction operation modes of the HVAC system and determining correction system operation efficiency cop values of the HVAC system corresponding to each correction operation mode according to the disturbance parameter, the basic operation mode and a preset HVAC system operation mechanism function; wherein the correction operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system;

[0059] Step 14: determining a control mode of the HVAC system; ​

[0060] Step 15: if the control mode is the first control mode, determining that the operation mode with the highest modified cop value in all modified operation modes is the optimal operation mode;

[0061] Step 16: if the control mode is the second control mode, obtaining the target cop value of the HVAC system according to the operation data by using the pre-trained operation control model; determining that the operation mode with the smallest difference between the modified cop value and the target cop value in all modified operation modes is the optimal operation mode;

[0062] Step 17: controlling the HVAC system to operate according to the optimal operation mode.

[0063] In some embodiments of the present application, optionally, the node device includes at least one of the following devices: a cooling tower, a cooling water pump, a chilled water pump, a cooling load, and a refrigeration host. Correspondingly, the operation data can be the number of start-ups, the operation frequency, and the power consumption of each node device. In the case where the node device includes the refrigeration host, the operation data can further include the cooling capacity, the load rate, the condensing temperature, and the evaporating temperature of the refrigeration host.

[0064] In the embodiments of the present application, the disturbance parameter can be, for example, the outdoor temperature variation and / or the outdoor humidity variation. The outdoor temperature variation and / or the outdoor humidity variation can cause the cooling effect of the cooling tower and the refrigeration capacity of the HVAC system to change, thereby interfering with the normal operation of the HVAC system (i.e., forming a disturbance to the HVAC system). It can be understood that the disturbance degree of the disturbance parameter to the HVAC system is related to the numerical value of the disturbance parameter itself.

[0065] In the specific execution process of step 11, the disturbance parameter of the preset monitoring area can be obtained by a sensor. The preset detection area is an area outside the HVAC system, which can be specifically set by the user according to the operation control requirement, for example: the HVAC system controls the indoor temperature and humidity variation, then the outdoor is outside the HVAC system, and the user can determine a certain area of the outdoor as the sampling area (i.e., the preset detection area) to obtain the disturbance parameter of the area.

[0066] It can be understood that the disturbance parameters obtained from different detection areas can be different, which can directly affect the execution efficiency of the subsequent steps. Therefore, the user can determine an optimal detection area through experimental comparison, or the user can set at least two detection areas, and determine the disturbance parameter of step 11 as the average of the disturbance parameters of the at least two detection areas. The above setting is beneficial to avoid the interference of external accidental factors on the operation control, and ensures the high accuracy of the operation control.

[0067] In the embodiments of the present application, the preset HVAC system operation mechanism function includes: 1, a refrigeration host model, 2, a cooling tower model, 3, a circulating water pump model.

[0068] 1. Refrigeration host model

[0069] The host is composed of a compressor, a condenser, an evaporator and an expansion valve. The refrigeration working medium is compressed into high-temperature and high-pressure gas by the compressor, enters the condenser to be cooled into high-pressure liquid, and after the expansion valve, the pressure is reduced, enters the evaporator to be evaporated by heat absorption, and then enters the compressor to complete a refrigeration cycle.

[0070] Referring to Figure 2 , as shown, Figure 2 is a p-h diagram (pressure-enthalpy diagram) of the vapor compression heat pump cycle process.

[0071] 1) Evaporator model

[0072] The circulating working medium evaporates and absorbs heat in the evaporator. The working medium is vaporized from the saturated state 4' to the superheated state 1'. In the process of 4'→1, the circulating working medium is in a saturated state, and 1→4' is in a superheated state.

[0073] The calculation formula of the circulating working medium flow rate is: In the formula: Q evap is the heat absorption amount of the circulating working medium in the evaporator, which is determined according to the terminal load.

[0074] The calculation formula of the evaporating temperature T evap is: T evap =T sat (P 4′ ), in which: T sat is the saturation temperature of the working medium at a given pressure.

[0075] 2) Compressor model

[0076] In the heat pump cycle process, the compressor consumes power and compresses the circulating working medium. The temperature and pressure of the working medium increase.

[0077] The theoretical compression process is adiabatic compression (i.e. isentropic compression). The pressure, temperature and enthalpy of the circulating working medium in the compression process increase from (P 1’ , T 1’ , h 1’ ) to (P2, T2, h 2,s ), and the entropy remains unchanged, i.e. s2=s1'.

[0078] The calculation formula of the compression work w consumed by the compressor for compressing one unit of circulating working medium is: w=h 2,s -h 1′ .

[0079] In the actual compression process, due to the existence of irreversible factors such as heat transfer outside the compressor, the compression process deviates from the adiabatic compression. The pressure, temperature and enthalpy of the circulating working medium increase from (P1’ T 1’ h 1’ ) to (P2, T2, h 2’ ), the entropy value increases from s 1’ to s 2’ . The calculation formula of the compression work w' consumed by compressing a unit mass of the cycle working medium is: w' = h 2′ - h 1′ .

[0080] The actual compression process deviates from the adiabatic compression process to some extent, which is usually measured by the compressor adiabatic efficiency. The calculation formula of the adiabatic compression efficiency η s is: η s = w / w' = (h 2，s - h 1′ ) / (h 2′ - h 1′ ).

[0081] 3) Condenser model

[0082] The condenser of the refrigeration host machine is used to discharge the terminal heat and the work consumed by the compression process to the air, and is connected with the cooling tower. The heat discharge amount is related to the outlet water temperature of the cooling tower and the cooling water flow rate. The change process of the cycle working medium in the condenser is approximately an isobaric process.

[0083] The cycle working medium changes in the condenser. The cycle working medium in the superheated state after compression is condensed and releases heat in the condenser, and the state of the working medium is cooled from the superheated state 2' to the supercooled state 3'. For the purpose of simplifying the model, the condensation process is regarded as an isobaric process, i.e. P 2’ = P 3’ . The difference between the temperature of the state 3' point and the saturation temperature corresponding to the exhaust pressure is referred to as the supercooling degree.

[0084] The calculation formula of the supercooling degree T super_cool is: T super_cool = T3-T 3′ .

[0085] The calculation formula of the saturated condensation temperature T 3′ is: T 3′ = T sat (P 3′ ).

[0086] The calculation formula of the heat discharge amount of the condenser is: In the formula, Q cond is the heat discharged through the condenser, in kJ; h 2′ is the enthalpy value of the working medium at the inlet of the condenser, in kJ / kg; and h 3′ is the enthalpy value of the working medium at the outlet of the condenser, in kJ / kg.

[0087] 4) Expansion valve model

[0088] The working fluid flows through the expansion valve for a short time, and the heat exchange amount with the outside can be ignored during the process. The throttling process is regarded as an isenthalpic process. The enthalpy of the working fluid before and after the expansion valve remains unchanged, i.e., h3'=h4'.

[0089] 2. Cooling tower model

[0090] In the cooling tower, the hot water flow is in direct contact with the air for heat and mass transfer, and the heat exchange process includes both sensible heat and latent heat. The sensible heat part (secondary) is driven by the temperature difference between the air and the water, and the water temperature decreases and the air flow temperature increases during the sensible heat exchange process. This process is related to the water flow temperature and the dry bulb temperature of the air. The latent heat part (main) refers to the heat exchanged between the water flow and the air during the water evaporation process

[0091] Referring to Figure 3 As shown, the ambient air is drawn upward through the falling water, and most cooling towers have filling materials to increase the contact area between the water and the air surface. A cooling tower is usually composed of several tower cells, which are connected in parallel to share the water collecting tank.

[0092] At present, most researchers use the component-based cooling tower model proposed by Braum after 1989, and the mathematical expression is as follows:

[0093]

[0094] In the formula: ε a is the heat exchange coefficient of the cooling tower;

[0095] is the air mass flow in the cooling tower, which is related to the fan speed N, and the unit is kg / s;

[0096] h a，i is the enthalpy of the inlet air in the cooling tower, which is related to the outdoor air temperature T a and the outdoor humidity Unit: kJ / kg;

[0097] h a，w，i is the enthalpy of the water surface saturated air in the cooling tower, which is related to the outdoor air temperature T a and the outdoor humidity Unit: kJ / kg.

[0098] 2-1. Determination method of ε a

[0099] When the Lewis number is 1, for the counterflow cooling tower:

[0100]

[0101] ​When Lewis number is 1, for cross-flow cooling tower:

[0102]

[0103] Where: Ntu is the number of heat transfer units;

[0104] m * is the ratio of heat capacity of air and cooling water in the cooling tower;

[0105] C s is the specific heat capacity of saturated air at constant pressure, unit kJ / (kg*K);

[0106] is the water flow rate at the inlet of the cooling tower, unit kg / s;

[0107] h D is the mass transfer coefficient, unit m / s;

[0108] A v is the surface area of water droplets per unit volume, unit m 2 ;

[0109] V cell is the exchange volume of the cooling tower group, unit m 3 .

[0110] 2-2, C s determination method

[0111] Specific heat capacity C s is determined by the water inlet and outlet conditions and enthalpy value, and the specific formula is:

[0112]

[0113] Where: h s,w,i is the water surface saturated air enthalpy value at the inlet of the cooling tower, unit kJ / kg;

[0114] h s,w,o is the water surface saturated air enthalpy value at the outlet of the cooling tower, unit kJ / kg;

[0115] T w,i is the water temperature at the inlet of the cooling tower, unit K;

[0116] T w，o is the water temperature at the outlet of the cooling tower, unit K.

[0117] 2-3, T w，o determination method

[0118]

[0119] Where: is the mass flow rate of the water entering the cooling tower, kg / s;

[0120] is the mass flow rate of the water leaving the cooling tower, which is about 96% to 99% (inclusive) of the mass flow rate of the water entering the cooling tower;

[0121] c p，w is the specific heat capacity of water at constant pressure, unit: kJ / (kg*K);

[0122] T ref is the reference temperature of water, equal to 273.15 K.

[0123] 2-4、 determination method

[0124]

[0125] in the formula: ω a，o is the humidity content of the air leaving the cooling tower, kg / kg of dry air;

[0126] ω a，i is the humidity content of the air entering the cooling tower, kg / kg of dry air;

[0127] ω a，o = ω s，w，e + (ω a，i - ω s，w，e ) exp (-Ntu)

[0128] in the formula: ω s，w，e is obtained from the heat transfer equation, and the actual enthalpy value is obtained by using the data of the psychrometric chart.

[0129] 3. Circulating water pump model

[0130] Consumption is the consumption of electric energy, and the equipment that does work on the circulating water, after passing through the water pump, the pressure of the circulating water is increased. The purpose of modeling the circulating water pump model is to predict the trend of changes in the circulating water head, flow rate and consumption power when the rotating speed of the circulating water pump changes. The modeling process of the circulating water pump is as follows:

[0131] P pump = L x H x g x ε ρ x η / 3600

[0132] in the formula: L is the circulating water flow rate, unit: m 3 / h;

[0133] H is the head, unit: m;

[0134] δ ρ is the relative density ratio of the circulating working medium to water, and when the circulating working medium is water, the value is 1;

[0135] η is the pump efficiency, usually 80%~90% (including 80% and 90%).

[0136] The variation expression of flow, head and power with pump rotating speed is as follows:

[0137] The relationship expression of flow and rotating speed is as follows:

[0138] The relationship expression of head and rotating speed is as follows:

[0139] The relationship expression of shaft power and rotating speed is as follows:

[0140] In the formula, L is the circulating water flow, unit m 3 / h;

[0141] H is the head, unit m;

[0142] N is the rotating speed, unit rpm;

[0143] P is the power, unit kW.

[0144] In the embodiment of the application, the above-mentioned HVAC system operation mechanism function can ensure that the corrected operation mode conforms to the HVAC system operation mechanism, and a high-accuracy corrected cop value is obtained. Based on the high-accuracy corrected cop value, the embodiment of the application can ensure that the optimal operation mode determined by comparing the cop values has high accuracy (steps 15 and 16), and high-accuracy HVAC system operation control is realized.

[0145] In some embodiments of the application, the disturbance parameter is at least one.

[0146] In some embodiments of the application, in step 13, at least two corrected operation modes of the HVAC system are determined according to the disturbance parameter, the basic operation mode and the preset HVAC system operation mechanism function. That is, the disturbance parameter is substituted into the HVAC system operation mechanism model on the basis of the basic operation mode to determine the disturbance influence of the disturbance parameter on the basic operation mode. The original set parameters in the basic operation mode are further adjusted to obtain an operation mode capable of offsetting the disturbance formed by the disturbance parameter on the HVAC system. It can be understood that the adjustment scheme of adjusting the original set parameters in the basic operation mode is not unique, and thus at least two adjustment schemes (equivalent to at least two corrected operation modes in the embodiment of the application) can be obtained.

[0147] It should be noted that the user determines the disturbance has been offset according to the actual needs. The following is described in conjunction with an example. The example is that a certain disturbance parameter only disturbs one type of operation data index in the HVAC system, and all adjustment schemes are adjustment schemes for restoring the disturbed operation data index. A certain disturbance parameter disturbs at least two operation indexes in the HVAC system, and the user can specify whether to offset the disturbance to all operation indexes or only to part of the operation indexes, and then all adjustment schemes (i.e., all modified operation modes) are adjustment schemes for restoring the disturbance operation index specified by the user. In actual application, the operation data indexes disturbed are often coupled or contradictory to each other, and the user can determine to make adjustment schemes for each disturbed operation data index, and / or the user can determine to make an overall adjustment scheme for all disturbed operation data indexes.

[0148] In step 14 of the embodiment of the present application, the control mode of the HVAC system is determined. In some embodiments of the present application, the control mode can be preset by the user, and step 14 is to verify which control mode is preset by the user. In some embodiments of the present application, the control mode can be determined according to other indexes (for example, the disturbance parameter), and step 14 is to determine which control mode the HVAC system is according to the change degree of the external state quantity. In actual application, whether the value of the disturbance parameter exceeds the preset disturbance parameter threshold is verified, and which control mode the HVAC system is is determined according to the verification result.

[0149] It should be noted that in the first control mode, the operation mode with the highest operation efficiency is determined as the optimal operation mode, that is, high system operation efficiency is prioritized to ensure low energy consumption operation of the system, energy saving and emission reduction, and high accuracy of system operation control is considered. In the second operation mode, the operation mode with the smallest difference from the target cop value is determined as the optimal operation mode, that is, high accuracy of system operation control is prioritized, and high system operation efficiency is considered. Through step 14, the priority requirement of the embodiment of the present application is determined, which can greatly improve the flexibility of operation control. For example, the user can set the high electricity price period in a day to execute the first control mode, and the execution process of step 14 is: determining whether the current time is in the high electricity price period; if it is in the high electricity price period, the control mode is determined as the first control mode, the low energy consumption operation of the system is prioritized to save energy and reduce emissions, and the user's electricity bill is reduced. For example, if the control mode is determined according to other indicators (for example: disturbance parameters), the execution process of step 14 is: verifying whether the disturbance parameter value exceeds the preset disturbance parameter threshold value, if it exceeds, it indicates that the disturbance caused by the disturbance parameter will seriously affect the system operation, and the control mode is determined as the second control mode, the high accuracy of system operation control is prioritized, and the purpose of quickly offsetting the disturbance is achieved through high-accuracy operation control. It can be understood that there is a situation that the current time is in the high electricity price period and the disturbance parameter value exceeds the preset disturbance parameter threshold value. In this case, the execution process of step 14 can be: sending information to the interactive terminal associated with the user, the information requesting the user to decide which control mode to use; the control mode decided by the user is determined as the control mode of the HVAC system.

[0150] The embodiments of the present application are explained and described below in combination with specific examples:

[0151] Referring to Figure 4 shown, the operation control in this example includes four parts: initial running state, disturbance, global optimization and issued instruction, wherein:

[0152] The initial running state part is the basic operation mode in the embodiment of the present application. In this example, the node device includes three types of devices, namely cooling side device, freezing side device and refrigeration host. The operation data of each node device is obtained, including:

[0153] The operation data cooling_0 of the cooling side device: the number of cooling towers turned on, the running frequency of the cooling tower and the power consumption of the cooling tower; the number of cooling water pumps turned on, the running frequency of the cooling water pump and the power consumption of the cooling water pump.

[0154] The operation data chill_0 of the freezing side device: end cooling load, air conditioning set temperature value, number of refrigerated water pumps turned on, running frequency of the refrigerated water pump and power consumption of the refrigerated water pump.

[0155] The running data of the refrigeration host chiller_0: cooling capacity, number of start-ups, load rate, condensing temperature, evaporating temperature and power consumption.

[0156] The disturbance part, i.e. obtaining the disturbance variable. In the present example, the disturbance variable includes the outdoor temperature change and the outdoor humidity change. The outdoor temperature change and / or the outdoor humidity change will cause the cooling effect of the cooling tower in the HVAC system and the refrigeration capacity to change, thereby interfering with the normal operation of the HVAC system (i.e. forming a disturbance to the HVAC system).

[0157] The global optimization part, i.e. according to the disturbance variable, the basic running mode and the preset HVAC system running mechanism function, at least two correction running modes of the HVAC system are determined, and the correction system running efficiency cop value of the HVAC system corresponding to each correction running mode is determined.

[0158] In the present example, the correction running mode has n, n is a positive integer and n≥3. According to the order from bottom to top of the global optimization part in the present example, the correction running mode 1, the correction running mode 2 and the correction running mode n are determined. Figure 4 The global optimization part in the present example is described in order from bottom to top as follows:

[0159] The correction running mode 1 includes: cooling_1, chill_1, chiller_1, i.e. setting the parameters of the node device according to cooling_1, chill_1, chiller_1 can offset the disturbance; controlling the HVAC system to run according to the correction running mode 1, the system running efficiency of the HVAC system is COP_1.

[0160] The correction running mode 2 includes cooling_2, chill_2, chiller_2, i.e. setting the parameters of the node device according to cooling_2, chill_2, chiller_2 can offset the disturbance; controlling the HVAC system to run according to the correction running mode 2, the system running efficiency of the HVAC system is COP_2.

[0161] The running mode between the correction running mode 2 and the correction running mode n cannot be exhausted, so the expression is omitted here, and the omission should not be considered unclear. Figure 4 In the present example, the correction running mode 2 and the correction running mode n use ellipsis to indicate the correction running modes omitted between them.

[0162] The correction running mode n includes cooling_n, chill_n, chiller_n, i.e. setting the parameters of the node device according to cooling_n, chill_n, chiller_n can offset the disturbance; controlling the HVAC system to run according to the correction running mode n, the system running efficiency of the HVAC system is COP_n.

[0163] In the example, the control mode of the HVAC system is the first control mode, and the operation mode with the highest cop value (i.e., COP_i) in COP_1, COP_2, …, COP_n is determined as the optimal operation mode (i.e., the modified operation mode i) in the example, and the specific expression is: COP_i = max(COP_1, COP_2, …, COP_n).

[0164] The instruction is issued, that is, the HVAC system is controlled to operate according to the optimal operation mode. In the example, the cooling_i, chill_i, and chiller_i corresponding to the optimal operation mode (the modified operation mode i) are used to set the parameters of the node device.

[0165] In some embodiments of the present application, the training step of the pre-trained operation control model includes:

[0166] A historical operation data set of each node device in the heating ventilation air conditioning (HVAC) system is obtained.

[0167] According to a preset HVAC system operation mechanism function, historical operation data associated with the system operation efficiency cop value of the HVAC system is extracted from the historical operation data set to obtain a first operation data set, and an influence coefficient of each first operation data in the first operation data set on the cop value is determined according to the HVAC system operation mechanism function.

[0168] A weight value is generated according to the influence coefficient, the first operation data is labeled according to the weight value, and a second operation data set is obtained by collecting all the labeled first operation data.

[0169] The second operation data set is used to train the control model to obtain the operation control model.

[0170] In some embodiments of the present application, the first operation data set is obtained by extracting the historical operation data associated with the system operation efficiency cop value of the HVAC system from the historical operation data set according to the preset HVAC system operation mechanism function, that is, determining which data in the historical operation data set are associated with the system operation efficiency cop value according to the HVAC system operation mechanism function, and then extracting the associated historical operation data from the historical data set to obtain the first operation data set. The first operation data, that is, the historical operation data associated with the system operation efficiency cop value of the HVAC system, includes the outdoor dry-bulb temperature, the outdoor relative humidity, the outdoor light intensity, the refrigeration host load rate, the chilled water supply temperature and flow, and the cooling water supply temperature and flow.

[0171] According to the HVAC system operation mechanism function, the influence coefficient of each first operation data on the cop value is determined, the weight value is generated according to the influence coefficient, and the first operation data is marked according to the weight value, and the second operation data set is obtained by collecting all the marked first operation data. It can be understood that the second operation data represents the operation data (i.e. the first operation data) associated with the operation efficiency cop value, and represents the influence degree of each associated operation data on the cop value (the weight value corresponding to each first operation data). In particular, the first operation data includes disturbance parameters (for example: the first operation data includes: outdoor dry-bulb temperature, outdoor relative humidity and outdoor light intensity, etc.). Therefore, the second operation data set is used to train the control model, and the obtained operation control model can obtain the influence degree of the disturbance parameter on the HVAC system, and obtain the cop value (i.e. the target cop value) of the HVAC system under the disturbance of the disturbance parameter.

[0172] It should be noted that the target cop value is only a calculated value of the model. Moreover, the operation control cannot be realized only according to the target cop value.

[0173] In order to realize the operation control, the embodiment of the present application determines the optimal operation mode (i.e. the modified operation mode i) in which the difference between the modified cop value and the target cop value is the smallest among all the modified operation modes, when the control mode is the second control mode, and obtains the target cop value of the HVAC system according to the operation data by using the pre-trained operation control model. (That is, the modified operation mode closest to the result obtained by the operation control model is regarded as the optimal operation mode). Further, by controlling the HVAC system according to the cooling_i, chill_i and chiller_i corresponding to the optimal operation mode (modified operation mode i), the parameters of each node device are set, and the HVAC system is controlled to operate according to the optimal operation mode.

[0174] In the embodiment of the present application, when the target cop value obtained by the operation control model can meet the control demand, the huge cost of computing power and time is avoided, and the operation control benefit can be effectively improved.

[0175] In some embodiments of the present application, the disturbance parameter includes at least one of the following: outdoor temperature variation, outdoor humidity variation. In practical application, the outdoor temperature variation is the outdoor dry-bulb temperature variation, and the outdoor humidity variation is the outdoor relative humidity variation.

[0176] The control system architecture of the operation control method is explained and described below in combination with specific examples:

[0177] For example, as shown in Figure 5 Figure 5 ​A schematic diagram of a control system architecture for applying the operation control method of the embodiments of the present application. In this example, the operation data of each node device is obtained by the station control platforms 1-6 through the SCADA data link. The station control platforms 1-6 send the obtained operation data to the centralized control platform (i.e. the six-station centralized control platform in the figure) through the six-station centralized data link. The centralized control platform executes the operation control method of steps 11-15 and step 17. In step 17, according to cooling_i, chill_i, chiller_i, the station control platforms 1-6 issue instructions to the node devices that need to be set, so as to realize the setting of various parameters of the node devices according to the corresponding cooling_i, chill_i, chiller_i of the optimal operation mode (corrected operation mode i) of the HVAC system.

[0178] In this example, the centralized control platform determines the control mode of the HVAC system by executing step 14. If the control mode is the second control mode, the centralized control platform calls the pre-trained operation control model deployed on the cloud platform through the DTU or gateway of the demand response new quantity link, and obtains the target cop value of the HVAC system according to the operation data by using the pre-trained operation control model. The operation mode in which the corrected cop value and the target cop value have the smallest difference among all the corrected operation modes is determined as the optimal operation mode. In this example, the node devices include: a refrigeration host, a cooling tower, a cooling circulating pump, an intelligent electric meter, an energy meter, a terminal temperature, and a (terminal) humidity.

[0179] In the embodiment of the present application, the operation data of each node device in the HVAC system is acquired, and a disturbance parameter is acquired; the disturbance parameter is an external state variable that can form a disturbance to the HVAC system and is outside the HVAC system; a basic operation mode of the HVAC system is determined according to the operation data; at least two correction operation modes of the HVAC system are determined according to the disturbance parameter, the basic operation mode and a preset HVAC system operation mechanism function, and a correction system operation efficiency cop value of the HVAC system corresponding to each correction operation mode is determined; the correction operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system; a control mode of the HVAC system is determined; if the control mode is a first control mode, the operation mode with the highest correction cop value in all the correction operation modes is determined as an optimal operation mode; if the control mode is a second control mode, a target cop value of the HVAC system is obtained according to the operation data by using a pre-trained operation control model; the operation mode with the smallest difference between the correction cop value and the target cop value in all the correction operation modes is determined as the optimal operation mode, and the HVAC system is controlled to operate according to the optimal operation mode, the optimal operation mode is determined by comparing the cop values (the operation mode with the highest correction cop value or the operation mode with the smallest difference between the correction cop value and the target cop value), the high operation efficiency of the HVAC system is considered in the case of offsetting the disturbance, the high operation efficiency of the HVAC system is realized, and the consumption of energy resources is reduced.

[0180] In some embodiments of the present application, the node device includes a cooling side device, a freezing side device and a refrigeration host.

[0181] For example, Figure 4 As shown, the initial operation state part is the basic operation mode in the embodiment of the present application. In this example, the node device includes three types of devices, i.e. a cooling side device, a freezing side device and a refrigeration host. The operation data of each node device includes:

[0182] The operation data cooling_0 of the cooling side device includes the number of cooling towers turned on, the operation frequency of the cooling tower and the power consumption of the cooling tower, the number of cooling water pumps turned on, the operation frequency of the cooling water pump and the power consumption of the cooling water pump.

[0183] The operation data chill_0 of the freezing side device includes the terminal cold load, the air conditioning set temperature value, the number of freezing water pumps turned on, the operation frequency of the freezing water pump and the power consumption of the freezing water pump.

[0184] The running data of the chiller host chiller_0: cooling capacity, number of start-ups, load rate, condensing temperature, evaporating temperature and power consumption.

[0185] The embodiment of the present application provides a training method of an operation control model, referring to Figure 6 , Figure 6 The embodiment of the present application provides a training method of an operation control model, referring to

[0186] Step 21: obtaining a historical running data set of each node device in the HVAC system;

[0187] Step 22: according to the preset HVAC system running mechanism function, extracting historical running data associated with the system running efficiency cop value of the HVAC system from the historical running data set to obtain a first running data set; and according to the HVAC system running mechanism function, determining the influence coefficient of each first running data in the first running data set on the cop value;

[0188] Step 23: generating a weight value according to the influence coefficient, and labeling the first running data according to the weight value, and collecting all the labeled first running data to obtain a second running data set;

[0189] Step 24: training the control model by using the second running data set to obtain the operation control model.

[0190] In the embodiment of the present application, the preset HVAC system running mechanism function includes: 1, a chiller host model, 2, a cooling tower model, 3, a circulating water pump model.

[0191] 1, chiller host model

[0192] The host is composed of a compressor, a condenser, an evaporator and an expansion valve. The refrigerant is compressed into a high-temperature and high-pressure gas by the compressor, enters the condenser to be cooled into a high-pressure liquid, and then the pressure is reduced by the expansion valve, enters the evaporator to be evaporated by absorbing heat, and then enters the compressor to complete a refrigeration cycle.

[0193] Referring to Figure 2 , Figure 2 It is a p-h diagram of the vapor compression heat pump cycle process.

[0194] 1) evaporator model

[0195] The circulating working medium evaporates and absorbs heat in the evaporator. The working medium is vaporized from the saturated state 4' to the superheated state 1'. In the process of 4'→1, the circulating working medium is in a saturated state, and 1→4' is in a superheated state.

[0196] The calculation formula of the working medium circulation flow is: wherein Q evap is the heat absorbed by the cycle working medium in the evaporator, determined according to the terminal load.

[0197] Evaporation temperature T evap The calculation formula of T evap = T sat 1 = T 4′ , wherein T sat is the saturation temperature of the working medium at a given pressure.

[0198] 2) Compressor model

[0199] In the heat pump cycle process, the compressor consumes power while compressing the suction cycle working medium, and the temperature and pressure of the working medium increase.

[0200] The theoretical compression process is adiabatic compression (i.e. isentropic compression), and the pressure, temperature and enthalpy of the cycle working medium during the compression process increase from (P 1’ , T 1’ , h 1’ ) to (P2, T2, h 2，s ), and the entropy remains unchanged, i.e. s2 = s1'.

[0201] The calculation formula of the compression work w consumed by the compressor per unit mass of cycle working medium is: w = h 2，s -h 1′ .

[0202] In the actual compression process, due to the existence of irreversible factors such as heat transfer to the outside of the compressor, the compression process deviates from the adiabatic compression process. The pressure, temperature and enthalpy of the cycle working medium increase from (P 1’ , T 1’ , h 1’ ) to (P2, T2, h 2’ ), and the entropy increases from s1' to s2'. The calculation formula of the compression work w' consumed by the compressor per unit mass of cycle working medium is: w' = h 2′ -h 1′ .

[0203] The adiabatic efficiency of the compressor is usually used to measure the degree of deviation of the actual compression process from the adiabatic compression process. The calculation formula of the adiabatic compression efficiency η s is: η s = w / w' = (h 2,s -h 1′ ) / (h 2′ -h 1′ ).

[0204] 3) Condenser model

[0205] The condenser of the refrigeration host machine is used to discharge the terminal heat and the power consumed in the compression process to the air, and is communicated with the cooling tower. The heat discharge amount is related to the outlet water temperature of the cooling tower and the cooling water flow. The change process of the circulating working medium in the condenser is approximately an isobaric process.

[0206] The circulating working medium changes its state in the condenser. The circulating working medium in the superheated state after compression is condensed and discharged in the condenser, and the state of the working medium is cooled from the superheated state 2' to the supercooled state 3'. For the simplified model, the condensation process is regarded as an isobaric process, i.e. P 2’ = P 3’ The difference between the temperature of the state 3' point and the saturation temperature corresponding to the exhaust pressure is called the supercooling degree.

[0207] The calculation formula of the supercooling degree T super_cool is T super_cool = T3-T 3′ .

[0208] The calculation formula of the saturated condensation temperature T 3′ is T 3′ = T sat (P 3′ ).

[0209] The calculation formula of the heat discharge amount of the condenser is: In the formula, Q cond is the heat discharged through the condenser, in kJ; h 2′ is the enthalpy value of the working medium at the inlet of the condenser, in kJ / kg; and h 3′ is the enthalpy value of the working medium at the outlet of the condenser, in kJ / kg.

[0210] 4) Expansion valve model

[0211] The time of the working medium flowing through the expansion valve is short, and the heat exchange amount with the outside during the process can be ignored. The enthalpy values of the working medium before and after the expansion valve remain unchanged, i.e. h3' = h4'.

[0212] 2, Cooling tower model

[0213] In the cooling tower, the hot water flow and the air directly contact to perform heat transfer and mass transfer. The heat exchange process includes two parts of sensible heat and latent heat. The sensible heat part (secondary) takes the temperature difference between the air and the water as the heat exchange driving force. In the sensible heat exchange process, the water temperature decreases, and the air flow temperature increases. The process is related to the water flow temperature and the dry bulb temperature of the air. The latent heat part (main) refers to the heat exchanged between the water flow and the air in the water evaporation process

[0214] See Figure 3As shown, ambient air is drawn upward through the falling water, and most cooling towers have fill material to increase the surface area of the water exposed to the air. A cooling tower is usually composed of several cells that share a sump in parallel.

[0215] Most researchers use the component-based cooling tower model proposed by Braum in 1989, which is mathematically expressed as:

[0216]

[0217] wherein ε a is the heat exchange coefficient of the cooling tower;

[0218] is the mass flow rate of air in the cooling tower, which is related to the fan speed N, and has a unit of kg / s;

[0219] h a,i is the enthalpy of the inlet air in the cooling tower, which is related to the outdoor air temperature T a and the outdoor humidity and has a unit of kJ / kg;

[0220] h a，w，i is the enthalpy of the saturated air on the surface of the inlet water in the cooling tower, which is related to the outdoor air temperature T a and the outdoor humidity and has a unit of kJ / kg.

[0221] 2-1, Determination method of ε a

[0222] When the Lewis number is 1, for a counterflow cooling tower:

[0223]

[0224] When the Lewis number is 1, for a crossflow cooling tower:

[0225]

[0226] wherein Ntu is the number of heat transfer units;

[0227] m * is the heat capacity ratio of the air and the cooling water in the cooling tower;

[0228] C s is the constant-pressure specific heat capacity of the saturated air, which has a unit of kJ / (kg*K);

[0229] is the inlet water flow rate of the cooling tower, which has a unit of kg / s;

[0230] h D ​Kg / s

[0231] A v Kg / s 2 ;

[0232] Kg / s cell Kg / s 3 .

[0233] Kg / s s Kg / s

[0234] Kg / s s Kg / s

[0235]

[0236] Kg / s s，w，i Kg / s

[0237] Kg / s s，w，o Kg / s

[0238] Kg / s w，i Kg / s

[0239] Kg / s w，o Kg / s

[0240] Kg / s w，o Kg / s

[0241]

[0242] Kg / s Kg / s

[0243] Kg / s Kg / s

[0244] Kg / s p，w Kg / s Kg / s

[0245] Kg / s ref Kg / s Kg / s

[0246] Kg / s Kg / s

[0247]

[0248] where ω a，o is the cooling tower outlet air humidity ratio, kg / kg dry air;

[0249] ω a,i is the cooling tower inlet air humidity ratio, kg / kg dry air;

[0250] ω a，o = ω s，w，e + (ω a，i - ω s,w，e ) exp(-Ntu)

[0251] where ω s,w,e is obtained from heat transfer equation, and actual enthalpy is obtained from psychrometric chart.

[0252] 3. Circulating water pump model

[0253] Consumption is the consumption of electric energy, and the equipment that does work on circulating water. After the circulating water passes through the pump, the pressure of the circulating water is increased. The purpose of modeling the circulating water pump model is to predict the change trend of the circulating water head, flow rate, and consumption power when the rotating speed of the circulating water pump changes. The modeling process of the circulating water pump is as follows:

[0254] P pump = L x H x g x ε ρ x η / 3600

[0255] where L is the circulating water flow rate, unit m 3 / h;

[0256] H is the head, unit m;

[0257] ε ρ is the relative density ratio of the circulating working medium to water, and is taken as 1 when the circulating working medium is water;

[0258] η is the pump efficiency, and is usually taken as 80% to 90% (including 80% and 90%).

[0259] The change law expressions of the flow rate, head, and power with the pump rotating speed are as follows:

[0260] The relationship expression of the flow rate and the rotating speed is:

[0261] The relationship expression of the head and the rotating speed is:

[0262] The relationship expression of the shaft power and the rotating speed is:

[0263] where L is the circulating water flow rate, unit m 3 / h;

[0264] H is the head, unit m;

[0265] N is the rotating speed, unit rpm;

[0266] P is the power, unit kW.

[0267] According to the HVAC system operation mechanism function, the influence coefficient of each first operation data on the cop value is determined, the weight value is generated according to the influence coefficient, and the first operation data is marked according to the weight value, and the second operation data set is obtained by collecting all the marked first operation data. It can be understood that the second operation data represents the operation data (i.e. the first operation data) associated with the operation efficiency cop value, and represents the influence degree of each associated operation data on the cop value (the weight value corresponding to each first operation data). In particular, the first operation data includes disturbance parameters (for example: the first operation data includes: outdoor dry bulb temperature, outdoor relative humidity and outdoor light intensity, etc.). Therefore, the second operation set is used to train the control model, and the obtained operation control model can obtain the influence degree of the disturbance parameter on the HVAC system, and obtain the cop value (i.e. the target cop value) of the HVAC system under the disturbance of the disturbance parameter.

[0268] It should be noted that the target cop value is only a calculated value of the model. Moreover, the operation control of the HVAC system cannot be realized only according to the target cop value.

[0269] The operation control model trained by the embodiment of the present application is applied to the operation control method of the heating ventilation air conditioning (HVAC) system, and the specific steps of the operation control method of the HVAC system can include:

[0270] Obtaining the operation data of each node device in the HVAC system, and obtaining the disturbance parameter; wherein the disturbance parameter is an external state quantity that can form a disturbance to the HVAC system and is outside the HVAC system;

[0271] Determining the basic operation mode of the HVAC system according to the operation data;

[0272] According to the disturbance parameter, the basic operation mode and the preset HVAC system operation mechanism function, at least two correction operation modes of the HVAC system are determined, and the correction system operation efficiency cop value corresponding to each correction operation mode of the HVAC system is determined; wherein the correction operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system;

[0273] Determining the control mode of the HVAC system;

[0274] If the control mode is the first control mode, the operation mode with the highest correction cop value in all correction operation modes is determined as the optimal operation mode;

[0275] If the control mode is the second control mode, the target COP value of the HVAC system is obtained from the operating data using a pre-trained operating control model; the operating mode with the smallest difference between the modified COP value and the target COP value among all modified operating modes is determined as the optimal operating mode;

[0276] Control the HVAC system to operate in the optimal operating mode.

[0277] The following explanation of the model training steps is provided with specific examples:

[0278] See Figure 7 As shown, the model training in this example includes four parts: feature value identification, selection of baseline operating conditions, calculation of the influence coefficient of each feature value on operating efficiency, and efficiency prediction for new operating conditions.

[0279] The feature value identification section involves acquiring the historical operating data set of each node device in the HVAC system. Based on a preset HVAC system operating mechanism function, historical operating data associated with the HVAC system's system operating efficiency (COP) value is extracted from the historical operating data set to obtain the first operating data set. Specifically, in this example, the first operating data includes: outdoor dry-bulb temperature, outdoor relative humidity, light intensity, chiller load, chilled water supply temperature, cooling water supply temperature, and cooling water circulation rate.

[0280] The baseline operating condition is selected by choosing the historical data with the highest recurrence frequency as the baseline operating condition for feature weighting.

[0281] Calculate the influence coefficient of each feature value on operating efficiency, that is, determine the influence coefficient of each first operating data in the first operating data set on the COP value according to the HVAC system operating mechanism function. In practical applications, similar to step 13, determining the influence coefficient according to the HVAC system operating mechanism function requires setting a benchmark operating condition, that is, in the above-mentioned selection of benchmark operating conditions, the historical data with the highest recurrence frequency is selected as the benchmark operating condition for feature weighting.

[0282] Specifically, in this example, the influence coefficients of outdoor dry-bulb temperature, outdoor relative humidity, light intensity, chiller load rate, chilled water supply temperature, cooling water supply temperature, and cooling water circulation volume on operating efficiency (i.e., the influence coefficients of each first operating data point on the COP value) are calculated sequentially. The influence coefficients in this example are the average influence coefficients of n influence coefficients obtained by n calculations for each characteristic quantity.

[0283] Weight values ​​are generated based on the influence coefficients, and the first set of running data is labeled according to the weight values. All labeled first set of running data is then collected to obtain the second set of running data. The control model is trained using the second set of running data to obtain the running control model.

[0284] The new working condition efficiency prediction part, i.e., the target cop value of the HVAC system is obtained according to the operation data by using the pre-trained operation control model in the embodiment. In the example, the target cop value is the average value of the cop values obtained by n times of prediction.

[0285] The operation control step is explained in combination with a specific example as follows:

[0286] Referring to Figure 4 , the operation control in the example includes four parts: initial operation state, disturbance, global optimization, and issued instruction, wherein:

[0287] The initial operation state part is the basic operation mode in the embodiment. In the example, the node devices include three categories of devices, i.e., cooling side devices, chilling side devices, and refrigeration host. The operation data of each node device is obtained, including:

[0288] The operation data cooling_0 of the cooling side device: the number of cooling towers turned on, the operation frequency of the cooling tower, and the power consumption of the cooling tower; the number of cooling water pumps turned on, the operation frequency of the cooling water pump, and the power consumption of the cooling water pump.

[0289] The operation data chill_0 of the chilling side device: end cooling load, air conditioning set temperature value, number of chilled water pumps turned on, operation frequency of the chilled water pump, and power consumption of the chilled water pump.

[0290] The operation data chiller_0 of the refrigeration host: cooling capacity, number of units turned on, load rate, condensing temperature, evaporating temperature, and power consumption.

[0291] The disturbance part is to obtain the disturbance parameters. In the example, the disturbance parameters include the outdoor temperature change and the outdoor humidity change. The outdoor temperature change and / or the outdoor humidity change will cause the cooling effect of the cooling tower and the refrigeration capacity to change, thereby interfering with the normal operation of the HVAC system (i.e., forming a disturbance to the HVAC system).

[0292] The global optimization part is to determine at least two corrected operation modes of the HVAC system according to the disturbance parameters, the basic operation mode, and the preset HVAC system operation mechanism function, and to determine the corrected system operation efficiency cop value of the HVAC system corresponding to each corrected operation mode.

[0293] In the example, the corrected operation mode has n, n is a positive integer and n≥3. According to Figure 4 , the global optimization part is sequentially explained from bottom to top in the following:

[0294] The modified operation mode 1 includes cooling_1, chill_1, chiller_1, that is, the parameters of the node device are set according to cooling_1, chill_1 and chiller_1, and the disturbance can be offset; and the HVAC system is controlled to operate according to the modified operation mode 1, and the system operation efficiency of the HVAC system is COP_1.

[0295] The modified operation mode 2 includes cooling_2, chill_2, chiller_2, that is, the parameters of the node device are set according to cooling_2, chill_2 and chiller_2, and the disturbance can be offset; and the HVAC system is controlled to operate according to the modified operation mode 2, and the system operation efficiency of the HVAC system is COP_2.

[0296] The operation modes between the modified operation mode 2 and the modified operation mode n cannot be enumerated, and therefore the expressions are omitted, and the omission should not be considered as unclear. Figure 4 In the modified operation mode 2 and the modified operation mode n, the omitted expressions between the two are indicated by ellipsis.

[0297] The modified operation mode n includes cooling_n, chill_n, chiller_n, that is, the parameters of the node device are set according to cooling_n, chill_n and chiller_n, and the disturbance can be offset; and the HVAC system is controlled to operate according to the modified operation mode n, and the system operation efficiency of the HVAC system is COP_n.

[0298] In the case where the control mode is the second control mode, the target cop value of the HVAC system is obtained according to the operation data by using the operation control model; and the operation mode (modified operation mode i) in which the difference between the modified cop value and the target cop value is smallest is determined as the optimal operation mode (that is, the modified operation mode which is closest to the result obtained by the operation control model is determined as the optimal operation mode). Further, the parameters of the node device are set according to cooling_i, chill_i and chiller_i corresponding to the optimal operation mode (modified operation mode i) by controlling the HVAC system to operate according to the optimal operation mode.

[0299] In some embodiments of the present application, optionally,

[0300] The HVAC system operation mechanism function is used as the target function of the training control model.

[0301] In the embodiment of the present application, the HVAC system operation mechanism function is used as the target function of the training control model, which can effectively ensure that the operation control model obtained by training conforms to the operation mechanism of the HVAC system, and ensure that the target cop value obtained by the operation control model has high accuracy.

[0302] In some embodiments of the present application, optionally,

[0303] The operation control model is an equivalent model.

[0304] The operation control model is an equivalent model.

[0305] The operation control model is an equivalent model.

[0306] The embodiment of the present application provides an operation control device of a heating, ventilation and air conditioning (HVAC) system, as shown in Figure 8 The principle block diagram of the operation control device of the HVAC system in the embodiment of the present application is shown in Figure 8 The operation control device 70 comprises:

[0307] The first acquisition module 71 is configured to acquire operation data of each node device in the HVAC system, and acquire a disturbance parameter; wherein the disturbance parameter is an external state variable that can form a disturbance to the HVAC system and is outside the HVAC system.

[0308] The first execution module 72 is configured to determine a basic operation mode of the HVAC system according to the operation data.

[0309] The first execution module 72 is further configured to determine at least two correction operation modes of the HVAC system and determine correction system operation efficiency cop values of the HVAC system corresponding to each correction operation mode according to the disturbance parameter, the basic operation mode and a preset HVAC system operation mechanism function; wherein the correction operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system.

[0310] The first execution module 72 is further configured to determine the control mode of the HVAC system;

[0311] The first execution module 72 is further configured to, if the control mode is the first control mode, determine the operating mode with the highest correction cop value among all the correction operating modes as the optimal operating mode;

[0312] The first execution module 72 is further configured to, if the control mode is the second control mode, use a pre-trained operation control model to obtain the target COP value of the HVAC system based on the operation data; and determine the operation mode with the smallest difference between the modified COP value and the target COP value among all the modified operation modes as the optimal operation mode;

[0313] The first execution module 72 is also used to control the HVAC system to operate according to the optimal operating mode.

[0314] In some embodiments of the present invention, optionally,

[0315] The disturbance parameters include at least one of the following: outdoor temperature change and outdoor humidity change.

[0316] In some embodiments of the present invention, the node device may optionally include: a cooling-side device, a freezing-side device, and a refrigeration unit.

[0317] The operation control device 70 provided in this embodiment can achieve... Figures 1 to 5 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0318] This invention provides a training device for a runtime control model, see [link to relevant documentation]. Figure 9 As shown, Figure 9 This is a schematic diagram of the training device for the operation control model according to an embodiment of the present invention. The training device 80 includes:

[0319] The second acquisition module 81 is used to acquire the historical operating data set of each node device in the HVAC system;

[0320] The second execution module 82 is used to extract historical operating data associated with the system operating efficiency (COP) value of the HVAC system from the historical operating data set according to a preset HVAC system operating mechanism function to obtain a first operating data set; and to determine the influence coefficient of each first operating data in the first operating data set on the COP value according to the HVAC system operating mechanism function.

[0321] The second execution module 82 is further configured to generate weight values ​​based on the influence coefficient, label the first running data according to the weight values, and collect all the labeled first running data to obtain a second running data set;

[0322] Training module 83 is used to train the control model using the second set of running data to obtain the running control model.

[0323] In some embodiments of the present invention, optionally, the HVAC system operating mechanism function is used as the objective function for training the control model.

[0324] In some embodiments of the present invention, the operation control model may optionally be an equivalent model.

[0325] The training device 80 provided in this application embodiment can achieve Figure 6 and Figure 7 The various processes implemented in the method embodiment achieve the same technical effect, and will not be described again here to avoid repetition.

[0326] This invention provides an electronic device 90, see [link to relevant documentation]. Figure 10 As shown, Figure 10 This is a schematic block diagram of an electronic device 90 according to an embodiment of the present invention, including a processor 91, a memory 92, and a program or instructions stored in the memory 92 and executable on the processor 91. When the program or instructions are executed by the processor, they implement the steps in any of the operation control methods of the HVAC system of the present invention, as well as the steps in the training method of any of the operation control models of the present invention.

[0327] This invention provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement various processes of an embodiment of the operation control method for an HVAC system as described above, as well as various processes of an embodiment of the training method for an operation control model as described above, and can achieve the same technical effect. To avoid repetition, these will not be described again here.

[0328] The readable storage medium mentioned above includes, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0329] The embodiments of the present application are described above with reference to the accompanying drawings, but the present application is not limited to the above-described specific embodiments, and the above-described specific embodiments are merely illustrative, but not restrictive, and a person of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection of the present application.

Claims

1. A method of operating control of a heating, ventilation, and air conditioning (HVAC) system, the method comprising: The method comprises the following steps: acquiring operation data of each node device in an HVAC system, and acquiring a disturbance parameter; wherein the disturbance parameter is an external state variable that can form a disturbance to the HVAC system and is outside the HVAC system; determining a basic operation mode of the HVAC system according to the operation data; determining at least two corrected operation modes of the HVAC system according to the disturbance parameter, the basic operation mode, and a preset HVAC system operation mechanism function, and determining a corrected system operation efficiency cop value of the HVAC system corresponding to each corrected operation mode; wherein the corrected operation mode is an operation mode that can offset the disturbance formed by the disturbance parameter to the HVAC system; determining a control mode of the HVAC system; if the control mode is a first control mode, determining that an operation mode with the highest corrected cop value in all the corrected operation modes is an optimal operation mode; if the control mode is a second control mode, obtaining a target cop value of the HVAC system according to the disturbance parameter by using a pre-trained operation control model; and determining that an operation mode with the smallest difference between the corrected cop value and the target cop value in all the corrected operation modes is the optimal operation mode; controlling the HVAC system to operate in the optimal operation mode; The training method of the operation control model comprises the following steps: acquiring a historical operation data set of each node device in an HVAC system; extracting historical operation data associated with a system operation efficiency cop value of the HVAC system from the historical operation data set to obtain a first operation data set according to a preset HVAC system operation mechanism function; and determining an influence coefficient of each first operation data on the cop value in the first operation data set according to the HVAC system operation mechanism function; generating a weight value according to the influence coefficient, and labeling the first operation data according to the weight value; and collecting all the labeled first operation data to obtain a second operation data set; training a control model by using the second operation data set to obtain the operation control model.

2. The operation control method according to claim 1, wherein: the disturbance parameter comprises at least one of the following: an outdoor temperature change amount, and an outdoor humidity change amount.

3. The operation control method according to claim 1, wherein: the node device comprises a cooling side device, a freezing side device, and a refrigeration host.

4. The operation control method according to claim 1, wherein: the HVAC system operation mechanism function is used as an objective function for training the control model.

5. The operation control method according to claim 1, wherein: the operation control model is an equivalent model.

6. An operation control device of a heating, ventilation, and air conditioning (HVAC) system, characterized by comprising: The method comprises the following steps: a first acquisition module is configured to acquire operation data of each node device in an HVAC system, and acquire a disturbance parameter; wherein the disturbance parameter is an external state variable that can form a disturbance to the HVAC system and is outside the HVAC system; The first execution module is configured to determine a basic operation mode of the HVAC system according to the operation data; The first execution module is further configured to determine at least two corrected operation modes of the HVAC system according to the disturbance variable, the basic operation mode, and a preset HVAC system operation mechanism function, and determine a corrected system operation efficiency cop value of the HVAC system corresponding to each of the corrected operation modes; the corrected operation mode is an operation mode that can offset the disturbance formed on the HVAC system by the disturbance variable; The first execution module is further configured to determine a control mode of the HVAC system; The first execution module is further configured to, if the control mode is a first control mode, determine an operation mode with the highest corrected cop value in all the corrected operation modes as an optimal operation mode; The first execution module is further configured to, if the control mode is a second control mode, obtain a target cop value of the HVAC system according to the disturbance variable by using a pre-trained operation control model; and determine an operation mode with the smallest difference between the corrected cop value and the target cop value in all the corrected operation modes as the optimal operation mode; The first execution module is further configured to control the HVAC system to operate in the optimal operation mode; The training device of the operation control model comprises: A second acquisition module is configured to acquire a historical operation data set of each node device in a heating ventilation air conditioning (HVAC) system; A second execution module is configured to extract historical operation data associated with a system operation efficiency cop value of the HVAC system from the historical operation data set to obtain a first operation data set according to a preset HVAC system operation mechanism function; and determine an influence coefficient of each first operation data in the first operation data set on the cop value according to the HVAC system operation mechanism function; The second execution module is further configured to generate a weight value according to the influence coefficient, and label the first operation data according to the weight value; and collect all the labeled first operation data to obtain a second operation data set; A training module is configured to train a control model by using the second operation data set to obtain the operation control model.

7. An electronic device, comprising: A readable storage medium stores a program or instructions, and the program or instructions are executed by a processor to implement the steps in the HVAC system operation control method according to any one of claims 1 to 5.

8. A readable storage medium characterized by: The readable storage medium stores a program or instructions, and the program or instructions are executed by a processor to implement the steps in the HVAC system operation control method according to any one of claims 1 to 5.

Images