Central air-conditioning machine room system energy efficiency real-time simulation and optimization method

Abstract

A real-time energy efficiency simulation and optimization method for a central air-conditioning machine room system comprises the steps that S1, working condition data of the central air-conditioning machine room system are collected in real time, and historical non-abnormal data are selected as background data for fitting a mathematical relationship model between variables; s2, establishing an energy consumption model at least comprising a water chilling unit, an air-conditioning water pump, a cooling water pump and a cooling tower and a correlation model of each working condition, and performing fitting by using the background data to obtain a fitting coefficient of each model; s3, establishing an energy consumption incidence relation of coupling of the models, and constructing an energy efficiency simulation system; calculating a system simulation result under a current environment condition and a control strategy based on the energy efficiency simulation system; and S4, changing the control strategy for multiple times by using an optimization algorithm, obtaining a simulation result, performing judgment, and finally outputting an optimal control strategy. According to the invention, low-error prediction of the energy efficiency of the machine room is realized through a simulation system with high equipment association degree, control strategy optimization is realized in combination with an optimization algorithm, and the whole-year real-time efficient operation of the machine room is finally realized.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption simulation and optimization control of large-scale central air-conditioning room systems in public buildings, in particular to a real-time simulation and optimization method for energy efficiency of central air-conditioning room systems. Background Art

[0002] Central air conditioning rooms account for 40% of energy consumption in public buildings. Most control systems in these rooms lack systematic optimization capabilities, and each device operates under independent closed-loop control. Real-time, full-operation-condition simulation optimization is an effective approach to achieving systematic optimization of computer rooms. Accurate energy consumption models for each device under all operating conditions are fundamental to achieving system simulation and reducing overall errors. Reasonable correlations between these models are also prerequisites for feasible system optimization. These shortcomings are also addressed by existing technologies.

[0003] CN116697530A discloses a high-efficiency adaptive energy-saving control system and method for a computer room based on working condition prediction. Although it provides a method of analyzing and predicting the working conditions of the terminal, various equipment in the computer room, air conditioners and cooling pipes, and decomposing this demand to each device to optimize system control, so as to reduce energy consumption and improve system efficiency, it lacks a model for quantitatively solving its needs and a specific linkage idea between the needs of each device, and cannot implement precise optimization and global optimization and achieve efficient and low-consumption operation throughout the year.

[0004] CN114165854A discloses an intelligent optimization control method based on a dynamic simulation platform for a central air-conditioning system. It provides quantitative solution models for various types of equipment, builds an energy consumption simulation system by connecting the models in series, and optimizes an efficient control strategy using a coded genetic algorithm. However, each model has key factor defects or indirect correlations, resulting in a large overall prediction error. Furthermore, the impact of outdoor weather on system energy efficiency is not considered and integrated into the simulation system, making it impossible to achieve accurate and stable optimization control and ultimately efficient operation throughout the year.

[0005] CN119538570A discloses a data-driven method and device for simulating the energy consumption of an air-conditioning system. The method determines whether the system energy consumption is abnormal by establishing an energy consumption prediction model and collecting real-time data for prediction. The method can realize a prediction scheme for system energy consumption based on the control state of the equipment. However, this scheme cannot predict the mutual influence of each prediction model under changes in the control state, and therefore lacks the feasibility of simulation optimization. Summary of the Invention

[0006] The purpose of the present invention is to solve at least one of the above technical problems and provide a real-time simulation and optimization method for the energy efficiency of a central air-conditioning room system. Through a simulation system with high equipment correlation, low-error prediction of the energy efficiency of the room is achieved, and the control strategy is optimized by combining the optimization algorithm, ultimately achieving real-time and efficient operation of the room throughout the year.

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

[0008] A method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system includes the following steps:

[0009] S1: Real-time collection of various operating data of the central air-conditioning room system, real-time analysis and verification of the data, storage of verified non-abnormal data; and selection of historical non-abnormal data as background data for fitting the mathematical relationship model between variables;

[0010] S2: establishing an energy consumption model including at least a chiller, an air conditioning water pump, a cooling water pump, and a cooling tower, as well as a correlation model for each working condition, and fitting them using the background data to obtain a fitting coefficient for each model;

[0011] S3: Integrate the energy consumption models in S2, establish energy consumption correlations among the coupled models, and construct an energy efficiency simulation system. Based on the energy efficiency simulation system, load real-time indoor and outdoor environmental parameters as fixed input conditions, set control strategy-related variables as adjustable inputs, and calculate system simulation results under the current environmental conditions and control strategy.

[0012] S4: Use the optimization algorithm to change the control strategy multiple times and obtain simulation results and make evaluations, and finally output the optimal control strategy.

[0013] Furthermore, in S2, the method for establishing the energy consumption model of the chiller includes: establishing an energy consumption function of the chiller using the total heat brought by the air-conditioning water into the evaporator and the pressure difference between the condenser and the evaporator as two main input variables, obtaining an energy consumption model expression of the chiller with a fitting coefficient, and fitting the model using the background data to obtain the fitting coefficient;

[0014] or,

[0015] The method for establishing an energy consumption model for an air conditioning water pump includes: establishing a total energy consumption function for the air conditioning water pump using the number of air conditioning water pumps that are simultaneously turned on and the frequency of a single air conditioning water pump as two main input variables; and establishing an air conditioning water flow function using the total energy consumption of the air conditioning water pump, the number of air conditioning water pumps that are simultaneously turned on, and the pressure difference between the outlet and inlet of the air conditioning water pump as three main input variables, obtaining an expression for the total energy consumption model of the air conditioning water pump and an expression for the air conditioning water flow model with fitting coefficients, and performing fitting using the background data to obtain the fitting coefficients;

[0016] or,

[0017] The method for establishing an energy consumption model of a cooling water pump includes: establishing a total energy consumption function of the cooling water pump using the number of cooling water pumps simultaneously turned on and the frequency of a single cooling water pump as two main input variables; and establishing a cooling water flow function using the total energy consumption of the cooling water pump, the number of cooling water pumps simultaneously turned on, and the pressure difference between the outlet and inlet of the cooling water pump as three main input variables, obtaining a total energy consumption model expression of the cooling water pump and a cooling water flow model expression with fitting coefficients, and performing fitting using the background data to obtain the fitting coefficients;

[0018] or,

[0019] The method for establishing an energy consumption model of a cooling tower includes: taking the frequency of a single fan of the cooling tower and the number of cooling tower openings as two main input variables to establish a total fan air volume function and a total energy consumption function of the cooling tower; taking the total heat brought into the evaporator by air-conditioning water as the main input variable to establish a cooling load function of the cooling tower; and taking the outdoor wet-bulb temperature, cooling load, total fan air volume, the difference between the outdoor dry-bulb temperature and the wet-bulb temperature, and the difference between the outdoor dry-bulb temperature and the average temperature of the cooling tower inlet and outlet water as the main input variables to establish a cooling tower outlet water temperature function; obtaining a cooling tower total energy consumption model expression with a fitting coefficient, and fitting using the background data to obtain the fitting coefficient; and also obtaining a cooling tower fan total air volume model expression and a cooling tower outlet water temperature model expression.

[0020] Furthermore, in S2, the working condition association model includes a total energy consumption model of the computer room system. The total energy consumption W0 of the computer room is expressed as follows:

[0021] W0=W WaterChiller +W KTpump +W LQpump +W LQtower

[0022] Where W WaterChiller is the total energy consumption of the chiller, W KTpump is the total energy consumption of the air conditioning water pump, W LQpump is the total energy consumption of the cooling water pump, W LQtower is the total energy consumption of the cooling tower.

[0023] Furthermore, the energy consumption function of the chiller is: W WaterChiler =f(Q KT ,P Δ )

[0024] Where W WaterChiller is the total energy consumption of the chiller; Q KT is the total heat brought into the evaporator by the air-conditioning water, P Δ It is the pressure difference between the refrigerant suction and discharge from the compressor;

[0025] Since the total heat brought into the evaporator by the air-conditioning water is equal to the heat exchanged by the air-conditioning water, and the pressure difference between the refrigerant sucked in and discharged by the compressor is equal to the difference between the condenser pressure and the evaporator pressure, then:

[0026] Q KT =C Water ·(T KTin -T KTout )·F KTpump

[0027] P Δ =P LNQ -P ZFQ

[0028] In the above formula, C Water is the specific heat capacity of water; T KTin T is the air conditioning water inlet temperature before entering the evaporator; KTout F is the outlet temperature of the air-conditioning water after leaving the evaporator; KTpump is the air conditioning water flow; P LNQ is the absolute pressure of the high-pressure refrigerant in the condenser; P ZFQ is the absolute pressure of the low-pressure refrigerant in the evaporator; where:

[0029] P ZFQ =B1·T KTout

[0030]

[0031] Wherein, B1 and C1 are comprehensive heat transfer coefficients, obtained by fitting the background data; T LQout T is the outlet temperature of cooling water after leaving the condenser; LQin is the cooling water inlet temperature before entering the condenser;

[0032] The energy consumption model expression of the chiller is:

[0033] W WaterChiller =A1·Q KT +A2·Q KT 2 +A3·Q KT ·P Δ +A4·Q KT +A5·Q KT 2 +A6

[0034] In the above formula, A1 to A6 are fitting coefficients, which are obtained by fitting the chiller energy consumption model expression based on multiple non-abnormal background data recorded in S1 at the same time.

[0035] Furthermore, the total energy consumption function of the air conditioning water pump is: W KTpump =f(N KTpump, HZ Set )

[0036] Where W KTpump is the total energy consumption of the air conditioning water pump; N KTpump The number of air conditioning water pumps that are turned on at the same time; HZ Set The frequency of a single air-conditioning water pump is the same for each air-conditioning water pump that is turned on.

[0037] Set independent model expressions according to the number of air conditioning water pumps turned on:

[0038]

[0039] Where: n is the number of air conditioning water pumps turned on; D 1_1 ~D n_2 is the fitting coefficient, obtained by fitting multiple non-abnormal background data recorded in S1 at the same time in the above independent model expressions;

[0040] The air conditioning water flow function is: F KTpump =f(W KTpump , N KTpump , H KToutin )

[0041] Where H KToutin The pressure difference between the outlet and inlet of the air conditioning water pump;

[0042] Set independent model expressions according to the number of air conditioning water pumps turned on:

[0043]

[0044] Where: n is the number of pumps turned on; E 1_1 ~E n_5 is the fitting coefficient, obtained by fitting multiple non-anomalous background data recorded in S1 at the same time in the above independent model expression; where:

[0045]

[0046] Where: F 1_1 ~F 24_2 is the fitting coefficient, obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal;

[0047] or,

[0048] The total energy consumption function of the cooling water pump is: W LQpump =f(N LQpump HZ SHt )

[0049] Where W LQpump is the total energy consumption of the cooling water pump; NLQpump The number of cooling water pumps that are turned on at the same time; HZ SHt The frequency of a single cooling water pump, and the frequency of each cooling water pump is the same;

[0050] Set independent model expressions according to the number of cooling water pumps turned on:

[0051]

[0052] Where: n is the number of pumps turned on; G 1_1 ~G n_2 is the fitting coefficient, obtained by fitting multiple non-abnormal background data recorded in S1 at the same time in the above independent model expressions;

[0053] The cooling water flow function is: F LQpump =f(W LQpump , N LQpump , H LQoutin )

[0054] Where H LQoutin is the pressure difference between the cooling water pump outlet and inlet;

[0055] Set independent model expressions according to the number of cooling water pumps turned on:

[0056]

[0057] Where: n is the number of cooling water pumps turned on; H 1_1 ~H n_5 The fitting coefficients are obtained by fitting multiple non-anomalous background data recorded in S1 at the same time in the above independent model expressions; where:

[0058]

[0059] Where: I 1_1 ~I n_2 The fitting coefficient is obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal.

[0060] Furthermore, the total air volume function of the cooling tower fan and the total energy consumption function of the cooling tower are respectively:

[0061] FL LQtower =f(HZ LQtower , N LLQtower )

[0062] W LQtower =f(HZ LQtower , N LQtower )

[0063] Where FLLQtower is the total air volume of the cooling tower, W LQtower is the total energy consumption of the cooling tower, HZ LQtower is the frequency of a single fan, N LQtower The number of cooling towers opened; the specific expression is:

[0064] FL LQtower =c·N LQtower ·HZ LQtower

[0065]

[0066] Where c is a constant, J1~J2 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal;

[0067] Cooling tower heat dissipation Q LQtower The heat brought into the chiller by the air-conditioning water, the cooling load model and expression are as follows:

[0068] Q LQtower =f(Q KT )

[0069] Q LQtower =(K1·Q KT +K2)

[0070] Where: Q KT is the total heat brought into the evaporator by the air-conditioning water; K1~K2 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal;

[0071] The cooling tower outlet water temperature function is:

[0072] T LQin =f(T Air_w , Q LQtower ,FL LQtower , T Air_d -T LQavg , T Air_d -T Air_w ,(T LQout -T LQwaterin )·Q LQtower )

[0073] Where: T LQin is the cooling tower outlet water temperature; T Air_w is the outdoor atmospheric wet-bulb temperature; T Air_d is the outdoor atmospheric dry-bulb temperature; T LQavg is the average heat exchange temperature of cooling water; T LQwaterin is the feed water temperature in the cooling water tank; where:

[0074]

[0075] Where, T LQout is the cooling tower inlet water temperature; L1~L4 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal.

[0076] Furthermore, S1 specifically includes the following steps:

[0077] Arrange various working condition sensors inside and outside the computer room, establish communication links between each working condition sensor and the host computer, and use the host computer to perform real-time analysis and verification of the received data. According to the pre-set classification restrictions of each received data and its combination, the data without abnormalities after verification is written into the corresponding structure table of the relational database carried by the host computer to achieve orderly storage of data; select historical non-abnormal data as background data for fitting the mathematical relationship model between variables.

[0078] Furthermore, S2 also includes a building air conditioning load model, and the building load function is: Q KT =f(T Air_d , RH Air , Lux Air )

[0079] Where Q KT This is the building load at this time, that is, the total heat brought into the evaporator by the air-conditioning water; T Air_d is the outdoor atmospheric dry bulb temperature; RH Air Lux is atmospheric humidity; Air is the current outdoor light intensity; the specific expression is as follows:

[0080] Q KT =M1·T Air_d +M2·T Air_d 2 +M3·T Air_d RH Air +M4·RH Air +M5·Lux Air

[0081] Where: M1~M5 are fitting coefficients, which are obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal;

[0082] The average heat exchange temperature function of air conditioning water supply and return water is:

[0083] Where, T KTavg is the average heat exchange temperature of air conditioning water supply and return water, T inner RH is the dry bulb temperature of the air in the representative area of ​​the building; inneris the air humidity in the representative indoor area, T Air_d is the outdoor atmospheric dry bulb temperature; RH Air is the atmospheric humidity; the specific expression is as follows: T KTavg =N1·T inner +N2·RH inner +N3·T Air_d +N4·RH Air +N5

[0084] Where: N1~N5 are fitting coefficients, which are obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal.

[0085] Furthermore, in S3, the method for establishing the energy efficiency simulation system includes the following steps:

[0086] S3-1: Set inputs, including environmental quantification and control strategy groups;

[0087] The environmental quantification is as follows:

[0088] In the above quantitative analysis, T Air_w is the outdoor atmospheric wet-bulb temperature, T Air_d is the outdoor atmospheric dry bulb temperature, RH Air Lux is atmospheric humidity; Air is the current outdoor light intensity; T inner is the indoor dry bulb temperature, RH inner The required humidity for the room; Time day is the current week number, Time hour is the current hour; where T Air_w and RH Air From the sensor, T inner and RH inner From artificial experience setting or setting by month; T LQwaterin The soft water temperature inside the cooling water make-up tank;

[0089] The control strategy groups are as follows:

[0090] Among them, N KTpump The number of air conditioning water pumps turned on, HZ Set The frequency of a single air-conditioning water pump, and the frequency of each air-conditioning water pump is the same; N LQpump The number of cooling water pumps turned on, HZ SHt The frequency of a single cooling water pump, and the frequency of each cooling water pump is the same; N LQtower The number of cooling towers opened, HZ LQtower The fan frequency of a single cooling tower is the same for each cooling tower that is turned on.

[0091] S3-2: Correlation calculation: Input the environmental quantitative and control strategy groups, perform correlation point calculations, establish energy consumption relationships between models, and finally obtain the total energy consumption of the chiller W. WaterChiller , total energy consumption of air conditioning water pump W KTpump , total energy consumption of cooling water pump W LQpump And the total energy consumption of cooling tower W LQtower Finally, the total energy consumption W0 and system efficiency S0 of the computer room system are obtained as follows:

[0092] W0=W WaterChiller +W KTpump +W LQpump +W LQtower

[0093]

[0094] Where Q KT The total heat brought by the air-conditioning water into the evaporator.

[0095] Furthermore, in S4, the real-time optimization method includes the following steps:

[0096] S4-1: Set the optimization goal: take the energy efficiency of the computer room system as the goal, set the optimization variable to the computer room system efficiency S0, and the optimization direction to the positive direction;

[0097] S4-2: Set constraints: During the simulation process, the optimization nodes are constantly monitored. If the working conditions of each node do not meet the constraints, the optimization is considered invalid. Constraint group 1 is the number of available units and variable frequency of each device, and constraint group 2 is the working condition range for safe operation of the equipment.

[0098] Constraint group 1 is as follows:

[0099]

[0100] Where: N KTpump 、N KTpump_Max 、HZ Set Respectively represent the number of air conditioning water pumps that are turned on, the number of air conditioning water pumps that are available, and the frequency of the unified setting of air conditioning water pump input; N LQpump 、N LQpump_Max 、HZ SHt Respectively represent the number of cooling water pumps that are turned on, the number of cooling water pumps that are available, and the frequency of the unified setting of cooling water pump input; N LQtower 、N LQtower_Max 、HZ LQtower They represent the number of cooling towers that are turned on, the number of cooling towers that are available, and the frequency of the unified setting of cooling tower inputs respectively;

[0101] Constraint group 2 is as follows:

[0102]

[0103] Where, T KTout P is the outlet temperature of the air-conditioning water after leaving the evaporator; LNQ is the absolute pressure of the high-pressure refrigerant in the condenser; P ZFQ is the absolute pressure of the low-pressure refrigerant in the evaporator; T Air_w is the outdoor atmospheric wet-bulb temperature; T LQin is the cooling tower outlet water temperature; F KTpump is the air conditioning water flow; F LQpump is the cooling water flow rate;

[0104] S4-3: Use the optimization algorithm to perform iterative optimization, taking the variables of constraint condition 1 as the optimization items, and output the optimal control strategy.

[0105] Beneficial effects of the present invention:

[0106] (1) Through real-time data collection and analysis, non-abnormal data is obtained as background data, which can be brought into the energy consumption model of each device for fitting to obtain fitting coefficients. On the one hand, the real-time data can reflect the current operating status and environmental conditions of the equipment. The fitting coefficients can dynamically correct the model deviation and avoid long-term prediction inaccuracies caused by equipment aging or working condition deviation; on the other hand, it can ensure the purity of the training set of the input model and reduce the pollution of noise on the fitting results;

[0107] (2) By establishing energy consumption models for chillers, air conditioning water pumps, cooling water pumps, and cooling towers, as well as associated models for various operating conditions, precise and global optimization can be achieved, ensuring efficient and low-consumption operation throughout the year. Furthermore, the impact of outdoor weather on system energy efficiency is integrated into the simulation system, enabling precise and stable optimization control and ultimately efficient operation throughout the year.

[0108] (3) In the energy consumption model of the chiller, by taking the pressure difference between the condenser and the evaporator as a more direct factor affecting the energy consumption of the compressor, the model is more accurate than the existing method of taking the air conditioning and cooling water temperature as indirect factors, and is more conducive to serving as a basis for judging the performance changes of the unit equipment;

[0109] (4) By adding the resistance factor along the way to the models of air-conditioning water pumps and cooling water pumps, the long-term accuracy is greatly improved compared to directly solving the water flow rate by the water pump frequency, speed or power. In addition, by removing the terminal influencing factors in the flow model, the abnormal performance of the water pump can be accurately judged. It can not only be used for fault diagnosis but also for judging performance degradation. It can also serve as a basis for cooling tower cleaning and water pump equipment maintenance.

[0110] (5) The energy efficiency simulation system established by the present invention not only has higher accuracy compared with historical data, but also, through the correlation of energy consumption between devices possessed by this simulation model architecture, it can enable subsequent actual simulation optimization and implementation control to still have the advantage of low error between simulation optimization results and implementation results under the new control strategy group;

[0111] (6) The present invention improves the system operation efficiency through the entire process, and ultimately achieves a reduction in energy consumption of public buildings throughout the year; it solves the problem that the models of various devices in the existing simulation system are inaccurate, which leads to large errors in the overall energy efficiency prediction; and solves the problem that the simulation architecture lacks internal correlation and each device independently optimizes, resulting in the optimization strategy losing sight of one thing while focusing on another, and further failing to achieve efficient operation of the system. DETAILED DESCRIPTION

[0112] The present invention will be further described in detail below with reference to specific embodiments.

[0113] In this embodiment, the central air-conditioning room system process equipment mainly includes: a chiller, a cooling tower, an air-conditioning water circulation pump (hereinafter referred to as the air-conditioning water pump), a cooling water circulation pump (hereinafter referred to as the cooling water pump), and a cooling water replenishment device; the chiller includes a compressor, an evaporator, and a condenser; the evaporator has low-pressure refrigerant inside and medium-temperature air-conditioning water outside; the condenser has high-pressure refrigerant inside and medium-temperature cooling water outside;

[0114] The process flow of the central air-conditioning room system described in this embodiment is divided into three thermal cycles:

[0115] 1. Air conditioning water circulation: The fan coil unit with automatic air volume control in the building transfers the heat of the indoor air to the air conditioning water and increases the temperature of the air conditioning water. Driven by the air conditioning water pump, the heated air conditioning water is sent back to the outside of the evaporator of the chiller in the computer room through the air conditioning return pipe, transferring the heat to the low-pressure refrigerant to cool the air conditioning water. After leaving the evaporator, it returns to the fan coil unit through the air conditioning water supply pipe to realize the air conditioning water circulation.

[0116] 2. Refrigerant circulation: The low-pressure refrigerant obtains heat from the air-conditioning water in the evaporator, and after being pressurized by the compressor, it becomes a high-temperature high-pressure refrigerant and is sent to the condenser. After transferring heat to the medium-temperature cooling water outside the condenser, it passes through the expansion valve and becomes a low-temperature low-pressure refrigerant again, completing the refrigerant circulation.

[0117] 3. Cooling water circulation: The cooling water that obtains heat in the evaporator enters the cooling tower under the push of the cooling water pump. The heat of the cooling water is taken away by surface evaporation in the cooling tower, and it becomes low-temperature cooling water. Under the action of the cooling water pump, it enters the unit again to complete the cooling water circulation. During this period, a cooling water make-up pipe will be set on the pipeline to replenish the cooling water.

[0118] The overall solution described in this embodiment is based on the above process equipment and process, and specifically includes the following steps:

[0119] S1: Real-time data collection and analysis records:

[0120] After the sensors are deployed, a stable and reliable communication link is established between the sensors and the host computer through the corresponding communication module. The host computer uses specially developed data acquisition software to perform real-time analysis and verification of the received data, and writes the verified data without abnormalities into the corresponding structure table of the relational database installed on the host computer. The application features of collection, analysis and recording are as follows:

[0121] S1.1: Data verification features:

[0122] Preset classification limits for each received data type and its combination. For example, the amplitude limit for the air conditioning water supply temperature sensor data is set to 3-30. If it exceeds this range, the data is considered abnormal. Other sensor data types are also configured with corresponding abnormality judgments.

[0123] S1.2: Database table structure characteristics of the host computer:

[0124] Set the table structure to use timestamp as the primary key and index key, set the index type to UNIQUE, and use various sensor data fields. Insert data row by row into the corresponding time series data table in the relational database on the host computer at the same interval through a timer to achieve orderly storage of data, so that subsequent data query, data cleaning and other operations based on time series can be performed.

[0125] S1.3: Data application-oriented features:

[0126] Select historical non-abnormal data as the background data for fitting the mathematical relationship model between variables. Select real-time data as the input source of the simulation system. If there is any missing or abnormal data at this time, the real-time optimization will be suspended.

[0127] S2: Establish relevant models for each device system

[0128] The background data obtained by S1 is applied to the energy consumption model of each device. The energy consumption models of the chiller, air conditioning water pump, cooling water pump, cooling tower, and various related models are fitted separately, and finally combined into a real-time energy efficiency model of the central air-conditioning room system.

[0129] S2.1: Chiller-related models

[0130] The energy consumption from the chiller is the main source of energy consumption for the entire system. The main energy consumption is used to drive the power consumption of the refrigerant compressor. The compressor sucks low-pressure refrigerant from the evaporator, pressurizes it into high-pressure refrigerant and sends it to the condenser. The suction amount is affected by the total heat brought into the evaporator by the air-conditioning water returning from the terminal. All compressors share the same evaporator and condenser, so the refrigerant pressure difference is the same. Therefore, the energy consumption of the chiller is related to the average pressure difference between the high and low pressure refrigerants and the heat from the air-conditioning water. The energy consumption of the chiller can be obtained as W WaterChiller function.

[0131] W WaterChiller =f(Q KT , P Δ ) (1)

[0132] In the above formula: W WaterChiller is the total power of all compressors in the chiller, that is, the total energy consumption of the chiller; Q KT is the total heat brought into the evaporator by the air-conditioning water, P Δ It is the pressure difference between the refrigerant suction and discharge from the compressor.

[0133] The total heat brought by the air-conditioning water into the evaporator is equal to the heat exchange of the air-conditioning water. The pressure difference between the refrigerant sucked in and discharged by the compressor is equal to the difference between the condenser pressure and the evaporator pressure. Then:

[0134] Q KT =C Water ·(T KTin -T KTout )F KTpump (2)

[0135] P Δ =R LNQ -P ZFQ (3)

[0136] In the above formula, C Water is the specific heat capacity of water; T KTin T is the temperature of the air conditioning pipe section outside the unit before the air conditioning water enters the evaporator, referred to as the air conditioning water inlet temperature; KTout The temperature of the air-conditioning pipe section outside the unit after the air-conditioning water leaves the evaporator, referred to as the air-conditioning water outlet temperature; F KTpump is the total circulation volume of air-conditioning water, hereinafter referred to as air-conditioning water flow; P LNQ is the absolute pressure of the high-pressure refrigerant in the condenser; P ZFQ It is the absolute pressure of the low-pressure refrigerant in the evaporator.

[0137] The final chiller energy consumption model expression is as follows:

[0138] W WaterChille =A1·Q KT +A2·Q KT2 +A3·Q KT ·P Δ +A4·Q KT +A5·Q KT 2 +A6 (4)

[0139] In the above formula, A1~A6 are fitting coefficients. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the six fitting coefficients of A1~A6 can be obtained by fitting in formula (4). Then, the total heat Q brought into the evaporator by the air-conditioning water of the required air conditioner in the simulation environment is substituted into formula (4). KT , the pressure difference P between the compressor suction and discharge refrigerant Δ , the chiller's energy consumption prediction results can be obtained. The above formula uses the pressure difference between the condenser and evaporator as a more direct factor affecting compressor energy consumption. Compared with the existing method that uses air conditioning and cooling water temperature as indirect factors, the model is more accurate and more useful as a basis for judging changes in unit equipment performance.

[0140] The evaporator pressure is controlled by the compressor load, and the compressor load control is affected by the air conditioning water outlet temperature. The absolute pressure model of the low-pressure refrigerant in the evaporator can be expressed as follows:

[0141] P ZFQ =f(T KTout ) (5)

[0142] The dynamic balance between the low-pressure refrigerant in the evaporator and the outlet temperature of the air-conditioning water is mainly affected by the heat exchange area and the heat transfer coefficient. The specific expression of the absolute pressure model of the low-pressure refrigerant in the evaporator is:

[0143] P ZFQ =B1·T KTout (6)

[0144] The above formula B1 is the comprehensive heat transfer coefficient. According to the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficient B1 can be obtained by fitting in formula (6).

[0145] The absolute pressure of the condenser's high-pressure refrigerant under dynamic equilibrium is related to the heat exchange area, heat transfer coefficient, and average heat exchange temperature of the cooling water heat exchange tube outside the condenser. The model for the absolute pressure of the high-pressure refrigerant in the condenser is:

[0146] P LNQ =f(T LQout ) (7)

[0147] The absolute pressure of the high-pressure refrigerant in the condenser is linearly related to the average heat exchange temperature, and its model expression is as follows:

[0148]

[0149] In the above formula: T LQout T is the temperature of the cooling pipe section outside the unit after the cooling water leaves the condenser, that is, the outlet temperature of the cooling water after leaving the condenser; LQin is the temperature of the cooling pipe section outside the unit before the cooling water enters the condenser, that is, the cooling water inlet temperature before entering the condenser; C1 is the comprehensive heat transfer coefficient. According to the multiple background data recorded in S1 at the same time that are not abnormal, the fitting coefficient C1 can be obtained by fitting in formula (8).

[0150] S2.2: Air conditioning water pump related models

[0151] The total energy consumption of the air conditioning water pump is affected by the number of units that are simultaneously turned on and the frequency regulation of the independent inverters at the upper end of the circuit. The regulation frequency of each inverter is set to be consistent. The model is:

[0152] W KTpump =f(N KTpump , HZ Set ) (9)

[0153] In the above formula, W KTpump is the total energy consumption of the air conditioning water pump; N KTpump The number of air conditioning water pumps that are turned on at the same time; HZ Set It is the frequency of a single air-conditioning water pump, and the frequency of each air-conditioning water pump that is turned on is the same.

[0154] Set independent model expressions according to the number of pumps turned on:

[0155]

[0156] In the above formula: n is the number of air conditioning water pumps turned on; D 1_1 ~D n_2 is the fitting coefficient. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficient D can be obtained by fitting in formula (10). 1_1 ~D n_2 .

[0157] The air conditioning water pump pushes the air conditioning water to overcome the resistance along the air conditioning pipe, so that the air conditioning water can circulate between the building end and the chiller in the central air conditioning room, bringing the heat of the building indoor air to the chiller; the air conditioning water pump can adjust the frequency and change the motor load to adjust the air conditioning water flow. Therefore, the air conditioning water flow F KTpump Related to the resistance along the pipeline and the load of the air conditioning water pump:

[0158] F KTpump =f(W KTpump ,N KTpump , H KToutin ) (11)

[0159] In the above formula, W KTpump is the total energy consumption of the air conditioning water pump; N KTpump H is the number of air conditioning water pumps that are turned on at the same time; KToutin It is the pressure difference between the outlet and inlet of the air conditioning water pump.

[0160] Set independent model expressions according to the number of pumps turned on:

[0161]

[0162] In the above formula: n is the number of pumps turned on; E 1_1 ~E n_5 is the fitting coefficient. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficient E can be obtained by fitting in formula (12). 1_1 ~E n_5 .

[0163] H KToutin As the pressure difference between the outlet and inlet of the air conditioning water pump, the resistance overcome by the air conditioning water circulation process is usually in W KTpump Keep stable when W KTpump The change of the internal business formats of public buildings changes with the regularity of 24 hours, and the influence of the synchronous ratio of W KTpump With H KToutin The relationship between H KToutin The function is as follows:

[0164]

[0165] According to theoretical analysis and actual data, H KToutin With W KTpump It is a linear relationship, and the specific expression is as follows:

[0166]

[0167] In the above formula: F 1_1 ~F 24_2 is the fitting coefficient. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficient F can be obtained by fitting in formula (14). 1_1 ~F 24_2 .

[0168] Compared with directly solving the air conditioning water flow rate through the air conditioning water pump frequency, speed or power, this embodiment adds the along-the-line resistance factor, which has higher long-term accuracy. In addition, the terminal influencing factors are removed from the flow model, which can accurately judge abnormal equipment performance. It can not only be used to judge equipment failures but also to judge performance degradation, serving as a basis for equipment maintenance.

[0169] S2.3: Cooling water pump related models

[0170] The total energy consumption of the cooling water pumps is affected by the number of units that are simultaneously turned on and the frequency regulation of the independent inverters at the upper end of their circuits. The regulation frequency of each inverter is set to be consistent. The model is:

[0171] W LQpump =f(N LQpump , HZ SHt ) (15)

[0172] In the above formula, W LQpump is the total energy consumption of the cooling water pump; N LQpump The number of cooling water pumps that are turned on at the same time; HZ SHt It is the frequency of a single cooling water pump, and the frequency of each cooling water pump that is turned on is the same.

[0173] Set independent model expressions according to the number of pumps turned on:

[0174]

[0175] In the above formula: n is the number of pumps turned on; G 1_1 ~G n_2 is the fitting coefficient. According to the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficient G can be obtained by fitting in formula (16). 1_1 ~G n_2 .

[0176] The cooling water pump pushes the cooling water to overcome the resistance along the cooling pipe, so that the cooling water can circulate between the building end and the chiller in the central cooling room, bringing the heat of the building indoor air to the chiller; the cooling water pump can adjust the cooling water flow by adjusting the frequency and changing the motor load. Therefore, the cooling water flow F LQpump Related to the resistance along the pipeline and the load of the cooling water pump:

[0177] F LQpump =f(W LQpump , N LQpump , H LQoutin ) (17)

[0178] In the above formula, W LQpump is the total energy consumption of the cooling water pump; N LQpump The number of cooling water pumps that are turned on at the same time; H LQoutin It is the pressure difference between the outlet and inlet of the cooling water pump.

[0179] Set independent model expressions according to the number of pumps turned on:

[0180]

[0181] In the above formula: n is the number of cooling water pumps turned on; H 1_1 ~H n_5 The fitting coefficient H can be obtained by fitting the multiple background data recorded in S1 at the same time and then solving them in formula (18). 1_1 ~H n_5 .

[0182] H LQoutin As the pressure difference between the outlet and inlet of the cooling water pump, the resistance overcome by the cooling water circulation process is usually in W LQpump Keep stable when W LQpump Change and change, affecting W LQpump With H LQoutin The main factor of the relationship is the number of cooling towers open, with the number of cooling towers open N LQtower The increase of H LQoutin The function is as follows:

[0183] H LQoutin =f(W LQpump , N LQtower ) (19)

[0184] According to theoretical analysis and actual data, H LQoutin With W LQpump It is a linear relationship, and the specific expression is as follows:

[0185]

[0186] In the above formula: I 1_1 ~I n_2 The fitting coefficient is obtained by fitting the multiple background data recorded in S1 at the same time and without abnormality in formula (20). 1_1 ~I n_2 .

[0187] Compared with directly solving the cooling water flow rate through the cooling water pump frequency, speed or power, this embodiment adds the resistance factor along the way, which has higher long-term accuracy and removes the terminal influencing factors in the flow model. It can accurately judge the abnormal performance of the cooling water pump, which can not only be used for fault diagnosis but also for judging performance degradation, and at the same time serve as the basis for cooling tower cleaning and cooling water pump equipment maintenance.

[0188] S2.4: Cooling Tower Related Models

[0189] The main energy consumption of the cooling tower comes from the useful power consumption of the fan motor. The total air volume of the fan and the total energy consumption of the cooling tower are the sum of the air volume and power consumption at the corresponding frequency and speed of each cooling tower. The total air volume function of the cooling tower fan and the total energy consumption function of the cooling tower are:

[0190] FL LQtower =f(HZ LQtower , N LQtower ) (twenty one)

[0191] W LQtower =f(HZ LQtower , N LQtower ) (twenty two)

[0192] In the above formula, FL LQtower is the total air volume of the cooling tower, W LQtower is the total energy consumption of the cooling tower, HZ LQtower is the frequency of a single fan, N LQtower The number of cooling towers opened; considering that the fan frequency of the cooling tower remains the same when it is opened, FL LQtower and W LQtower The specific expression is as follows:

[0193] FL LQtower =c·N LQtower ·HZ LQtower (twenty three)

[0194]

[0195] In the above formula, c is a constant, J1~J2 are fitting coefficients, and the fitting coefficients J1~J2 can be obtained by fitting in formula (24) based on multiple background data recorded in S1 that are not abnormal at the same time.

[0196] Cooling tower heat dissipation Q LQtower The heat brought into the chiller by the air-conditioning water, the cooling load model and expression are as follows:

[0197] Q LQtower =f(Q KT ) (25)

[0198] Q LQtower =(K1·Q KT +K2) (26)

[0199] In the above formula: Q KT is the total heat brought into the evaporator by the air-conditioning water, which can be obtained from formula (2); K1~K2 are fitting coefficients. According to the multiple background data recorded in S1 at the same time that are not abnormal, the fitting coefficients K1~K2 can be obtained by fitting in formula (26).

[0200] The cooling tower's outlet water temperature is a key indicator affecting the energy consumption of the computer room system and a crucial link between the cooling water system and the chiller. Cooling towers reduce cooling water temperature through three primary methods: evaporative heat dissipation, contact heat exchange with air, and replenishment of water from the cooling water makeup tank. The first two types of cooling water-air heat exchange can be roughly viewed as two processes: heat and mass exchange between the cooling water and saturated air at wet-bulb temperature, and heat and mass exchange between the saturated air and outdoor air at dry-bulb temperature.

[0201] According to the above analysis, the cooling tower outlet water temperature is affected by the outdoor wet-bulb temperature, cooling load, fan air volume, the difference between the outdoor dry-bulb temperature and the wet-bulb temperature, and the difference between the outdoor dry-bulb temperature and the average temperature of the cooling tower inlet and outlet water. The relevant functions are as follows:

[0202] T LQin =f(T Air_w , Q LQtwer , FL LQtower ,T Air_d -T LQawg , T Air_d -T Air_w ,(T LQout -T LQwaterin )Q LQtower )(27)

[0203] In the above formula: T LQin Q is the cooling water temperature after leaving the cooling tower (referred to as cooling tower outlet water temperature); LQtower FL is the total heat dissipation of cooling water in the cooling tower; LQtower is the total air volume of the cooling tower; T Air_w is the outdoor atmospheric wet-bulb temperature, which is the wet-bulb temperature value of the air temperature and humidity collector near the cooling inlet; T Air_d is the outdoor atmospheric dry-bulb temperature, which is the dry-bulb temperature value of the air temperature and humidity collector near the cooling inlet; T LQavg is the average heat exchange temperature of cooling water; T LQwaterin Is the water supply temperature in the cooling water tank. The average water temperature in and out of the cooling tower is T LQavg and cooling tower inlet water temperature T LQout The expression is as follows:

[0204]

[0205] In the above formula, T LQout It is the cooling water temperature before entering the cooling tower (referred to as cooling tower inlet water temperature).

[0206] In T LQin Among the related variables, LQtower 、T Air_d -T LQavg 、T Air_d-T Air_w 、(T LQout -T LQwaterin )·Q LQtower They are directly proportional to FL LQtower Inversely proportional relationship, T Air_w As the benchmark value, combined with formula (23), the specific function model expression is as follows:

[0207]

[0208] In the above formula, L1~L4 are fitting coefficients. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficients L1~L4 can be obtained by fitting in formula (30).

[0209] When applying the model to calculate T LQin When , due to the self-nested structure in formula (29), it is necessary to iterate the calculation. The iterative expression is as follows:

[0210]

[0211] In the above formula: is the initial value, n is the number of iterations, is the technical result after n iterations, the third formula is the iteration termination condition, ε LQinIteration Set to 0.1.

[0212] S2.5: Total energy consumption model of computer room system

[0213] The energy consumption equipment in the computer room comes from the above-mentioned chillers, air conditioning water pumps, cooling water pumps, and cooling towers. The total energy consumption W0 of the computer room is expressed as follows:

[0214] W0=W WaterChiller +W KTpump +W LQpump +W LQtower (32)

[0215] S2.6: Building Air Conditioning Load Related Models

[0216] The load of public buildings is mainly affected by the weather and varies regularly with the week and 24-hour cycle. Therefore, the function is as follows:

[0217] Q KT =f(T Air_d , RH Air , Lux Air ) (33)

[0218] In the above formula, Q KT This is the building load at this time, that is, the total heat brought into the evaporator by the air-conditioning water; T Air_d is the outdoor atmospheric dry bulb temperature; RHAir Lux is atmospheric humidity; Air is the current outdoor illumination. The specific expression is as follows:

[0219] Q KT =M1·T Air_d +M2·T Air_d 2 +M3·T Air_d RH Air +M4·RH Air +M5·Lux Air (34)

[0220] In the above formula, M1~M5 are fitting coefficients. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficients M1~M5 can be obtained by fitting in formula (34).

[0221] To ensure indoor and outdoor thermal balance in public buildings and meet corresponding indoor temperature and humidity requirements, it is necessary to provide a corresponding average heat exchange temperature on the hot water exchange side of the building's air conditioning fan coil unit. Therefore, the average heat exchange temperature function of the air conditioning water supply and return water is as follows:

[0222]

[0223] In the above formula, T KTavg is the average heat exchange temperature of air conditioning water supply and return water, T inner RH is the dry bulb temperature of the air in the representative area of ​​the building; inner is the air humidity in the representative indoor area, T Air_d is the outdoor atmospheric dry bulb temperature; RH Air is the atmospheric humidity. The specific expression is as follows:

[0224] T KTavg =N1·T inner +N2·RH inner +N3·T Air_d +N4·RH Air +N5 (36)

[0225] In the above formula, N1~N5 are fitting coefficients. Based on the background data of multiple non-abnormal data recorded in S1 at the same time, the fitting coefficients N1~N5 can be obtained by fitting in formula (36).

[0226] S3: Establishment of simulation system

[0227] The simulation optimization of the central air conditioning room is based on the simulation system established by associating the above models. The establishment steps are as follows:

[0228] S3.1: Set input quantity

[0229] The simulation system inputs described in this embodiment are divided into environmental quantities and control strategy groups.

[0230] The environmental quantification is as follows:

[0231]

[0232] The above quantitative: T Air_w is the outdoor atmospheric wet-bulb temperature, T Air_d is the outdoor atmospheric dry bulb temperature, RH Air Lux is atmospheric humidity; Air is the current outdoor light intensity; T inner is the indoor dry bulb temperature, RH inner The required humidity for the room; Time day is the current week number, Time hour is the current hour; where T Air_w and RH Air From the sensor, T inner and RH inner From artificial experience setting or setting by month; T LQwaterin It is the soft water temperature inside the cooling water make-up tank.

[0233] The control strategy groups are as follows:

[0234]

[0235] Where: N KTpump The number of air conditioning water pumps turned on, HZ Set The frequency of a single air-conditioning water pump, and the frequency of each air-conditioning water pump is the same; N LQpump The number of cooling water pumps turned on, HZ SHt The frequency of a single cooling water pump, and the frequency of each cooling water pump is the same; N LQtower The number of cooling towers opened, HZ LQtower It is the fan frequency of a single cooling tower, and the fan frequency of each cooling tower that is turned on is the same.

[0236] S3.2: Association Calculation

[0237] To calculate the outlet temperature of the chiller, enter the quantitative formula (37) and formula (38) in S3.1, and obtain Q through the air conditioning load model established in S2.6. KT , and T is obtained through the air conditioning water average heat transfer temperature model KTavg , and the air conditioning water flow model F established in S2.2 KTpump , and finally get the air conditioning water outlet temperature T KTout The expression is:

[0238]

[0239] To calculate the average heat exchange temperature of the chiller cooling water, enter the quantity in S3.1 and the above Q KT The calculation results of Q are obtained through the cooling tower heat dissipation model established in S2.4. LQtower , obtain T through the cooling tower outlet water temperature established in S2.4 LQin , the cooling water flow rate F is obtained through the cooling water flow model established in S2.3 LQpump , the final average heat exchange temperature of the cooling water in the chiller is T LQavg The expression is:

[0240]

[0241] To calculate the pressure difference between the high-pressure refrigerant inside the condenser and the low-pressure refrigerant inside the evaporator of the chiller, enter the above T KTout 、T LQavg The absolute pressure P of the low-pressure refrigerant in the evaporator can be obtained by the absolute pressure model of the high-pressure refrigerant in the evaporator established in S2.1. ZFQ The absolute pressure of the high-pressure refrigerant inside the condenser can be obtained by the absolute pressure model of the high-pressure refrigerant inside the condenser established in S2.1. LNQ .

[0242] The present invention ultimately realizes the quantitative correlation of the air-conditioning water circulation, the cooling water circulation, and the refrigerant circulation through the above-mentioned correlation point calculation steps, and ultimately realizes the correlation between the energy consumption and strategy of each device in the computer room through the following system energy consumption calculation, and ultimately realizes the feasibility of systematic simulation optimization.

[0243] S3.3: System energy consumption simulation

[0244] To calculate the energy consumption of the chiller, enter the quantity in S3.1 and the above Q KT The calculation results of W can be obtained through the chiller energy consumption model established in S2.1. WaterChiller , W can be obtained through the air conditioning water pump energy consumption model established in S2.2 KTpump , W can be obtained through the cooling water pump energy consumption model established in S2.3 LQpump , W can be obtained through the cooling tower energy consumption model established in S2.4 LQtower Finally, the energy consumption W0 and system efficiency S0 of the central air-conditioning room system in a public building are as follows:

[0245] W0=W WaterChiller +W KTpump +W LQpump +W LQtower (41)

[0246]

[0247] The present invention follows these steps: S1 deploys various operating condition sensors inside and outside the computer room, establishes communication between each sensor and a host computer, and analyzes and records data in real time into a feature database; S2 establishes models related to each device and fits them using background data that has passed secondary verification; S3 uses simulation methods to establish energy consumption relationships between the models. Ultimately, this method achieves accurate simulation of computer room system energy consumption, and optimizes the simulation by modifying the simulation model's input strategy group.

[0248] To verify the accuracy of the simulation system of the present invention, a certain time is input to obtain the total energy consumption of the computer room system W at that moment from the relational database. 0_real_i The energy consumption W of the simulation system is calculated based on the actual data and the readings of each sensor and control strategy. 0_simulation_i Comparison, using the indicators commonly used in statistics to measure the goodness of fit for evaluation, the evaluation index R 2 The calculation formula is as follows:

[0249]

[0250] The method of the present invention is used to simulate at n time points to obtain R 2 Reaching above 0.95, the energy consumption simulation accuracy of the central air conditioning room system in public buildings achieved by this invention is higher than that of existing technologies. Not only does this achieve higher accuracy when compared with historical data, but the correlation between energy consumption across devices provided by this simulation model architecture also allows for subsequent actual simulation optimization and implementation control to maintain the advantage of low error between simulation optimization results and implementation results under the new control strategy set.

[0251] S4: Real-time optimization method:

[0252] Based on this simulation system, we obtain quantitative inputs related to the real-time indoor and outdoor environment. By varying the variables within these inputs that serve as the control strategy, we calculate the system simulation results for the current indoor and outdoor environment and control strategy. By using an optimization algorithm to repeatedly change the control strategy and obtain and evaluate the simulation results, we ultimately determine the most efficient control strategy at that time, ensuring efficient operation of the computer room at that moment and ultimately achieving high system efficiency and low energy consumption throughout the year.

[0253] S4.1: Setting Optimization Goals

[0254] The simulation system of this embodiment is adapted to a variety of optimization algorithms (such as differential evolution algorithm, genetic algorithm, bat algorithm, gray wolf optimization algorithm, whale optimization algorithm, etc.). During the optimization process, the goal is to achieve efficient and energy-saving operation of the computer room system, the optimization variable is set as the computer room system efficiency S0, and the optimization direction is the positive direction.

[0255] S4.2: Set constraints

[0256] During the simulation process, the optimization nodes are constantly monitored. If the working conditions of each node do not meet the constraints, it is judged as invalid optimization. Constraint group 1 is the number of available units and variable frequency of each device, and constraint group 2 is the working condition range for safe operation of the equipment:

[0257] Constraint group 1 is as follows:

[0258]

[0259] In the above formula: N KTpump 、N KTpump_Max 、HZ Set Respectively represent the number of air conditioning water pumps that are turned on, the number of air conditioning water pumps that are available, and the frequency of the unified setting of air conditioning water pump input; N LQpump 、N LQpump_Max 、HZ SHt Respectively represent the number of cooling water pumps that are turned on, the number of cooling water pumps that are available, and the frequency of the unified setting of cooling water pump input; N LQtower 、N LQtower_Max 、HZ LQtower They represent the number of cooling towers that are turned on, the number of cooling towers that are available, and the frequency of unified settings of cooling tower inputs.

[0260] Constraint group 2 is as follows:

[0261]

[0262] S4.3: Optimization Algorithm

[0263] This embodiment can use genetic algorithms, particle swarm optimization algorithms, gradient descent methods and other methods for iterative optimization, all of which use the variables of constraint condition 1 as optimization items. Since the above algorithms are existing technologies, they will not be described in detail here.

[0264] In summary, the present invention aims to reduce the energy consumption of public buildings, obtain real-time and background data through real-time data collection and analysis and recording; establish energy consumption models of equipment such as chillers, air-conditioning water pumps, cooling water pumps, cooling towers, as well as various working condition correlation models of the computer room system and building air-conditioning load correlation models, and use background data for fitting to obtain characteristic coefficients of each model; these models are correlated to construct an energy efficiency real-time simulation system to achieve correlation calculation and low-error simulation of system energy efficiency; finally, with the energy efficiency of the computer room system as the goal, optimization variables and constraints are set, and real-time optimization is achieved with the help of a variety of optimization algorithms (such as genetic algorithms, etc.). The present invention improves the system operation efficiency through the above steps, and ultimately achieves a reduction in energy consumption of public buildings throughout the year. The present invention solves the problem that the models of various devices in the existing simulation system are not accurate, which leads to large errors in the overall energy efficiency prediction, and solves the problem that the simulation architecture lacks internal correlation and each device independently optimizes, resulting in the optimization strategy losing sight of one thing while focusing on another, and further failing to achieve efficient operation of the system.

[0265] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A real-time simulation and optimization method for energy efficiency of a central air-conditioning room system, characterized in that: The following steps are involved: S1: Real-time collection of various operating data of the central air-conditioning room system, real-time analysis and verification of the data, and storage of verified non-abnormal data; And select historical non-abnormal data as background data for fitting the mathematical relationship model between variables; S2: establishing an energy consumption model including at least a chiller, an air conditioning water pump, a cooling water pump, and a cooling tower, as well as a correlation model for each working condition, and fitting them using the background data to obtain a fitting coefficient for each model; S3: Integrate the energy consumption models in S2, establish energy consumption correlations among the coupled models, and construct an energy efficiency simulation system. Based on the energy efficiency simulation system, load real-time indoor and outdoor environmental parameters as fixed input conditions, set control strategy-related variables as adjustable inputs, and calculate system simulation results under the current environmental conditions and control strategy. S4: Use the optimization algorithm to change the control strategy multiple times and obtain simulation results and make evaluations, and finally output the optimal control strategy.

2. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 1 is characterized in that: In S2, the method for establishing the energy consumption model of the chiller includes: establishing an energy consumption function of the chiller using the total heat brought by the air-conditioning water into the evaporator and the pressure difference between the condenser and the evaporator as two main input variables, obtaining an energy consumption model expression of the chiller with a fitting coefficient, and fitting the model using the background data to obtain the fitting coefficient; or, The method for establishing an energy consumption model for an air conditioning water pump includes: establishing a total energy consumption function for the air conditioning water pump using the number of air conditioning water pumps that are simultaneously turned on and the frequency of a single air conditioning water pump as two main input variables; and establishing an air conditioning water flow function using the total energy consumption of the air conditioning water pump, the number of air conditioning water pumps that are simultaneously turned on, and the pressure difference between the outlet and inlet of the air conditioning water pump as three main input variables, obtaining an expression for the total energy consumption model of the air conditioning water pump and an expression for the air conditioning water flow model with fitting coefficients, and performing fitting using the background data to obtain the fitting coefficients; or, The method for establishing an energy consumption model of a cooling water pump includes: establishing a total energy consumption function of the cooling water pump using the number of cooling water pumps simultaneously turned on and the frequency of a single cooling water pump as two main input variables; and establishing a cooling water flow function using the total energy consumption of the cooling water pump, the number of cooling water pumps simultaneously turned on, and the pressure difference between the outlet and inlet of the cooling water pump as three main input variables, obtaining a total energy consumption model expression of the cooling water pump and a cooling water flow model expression with fitting coefficients, and performing fitting using the background data to obtain the fitting coefficients; or, The method for establishing an energy consumption model of a cooling tower includes: taking the frequency of a single fan of the cooling tower and the number of cooling tower openings as two main input variables to establish a total fan air volume function and a total energy consumption function of the cooling tower; taking the total heat brought into the evaporator by air-conditioning water as the main input variable to establish a cooling load function of the cooling tower; and taking the outdoor wet-bulb temperature, cooling load, total fan air volume, the difference between the outdoor dry-bulb temperature and the wet-bulb temperature, and the difference between the outdoor dry-bulb temperature and the average temperature of the cooling tower inlet and outlet water as the main input variables to establish a cooling tower outlet water temperature function; obtaining a cooling tower total energy consumption model expression with a fitting coefficient, and fitting using the background data to obtain the fitting coefficient; and also obtaining a cooling tower fan total air volume model expression and a cooling tower outlet water temperature model expression.

3. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 1 is characterized in that: In S2, the working condition association model includes the total energy consumption model of the computer room system. The total energy consumption W0 of the computer room is expressed as follows: W0=W WaterChiller +W KTpump +W LQpump +W LQtower Where W WaterChiller is the total energy consumption of the chiller, W KTpump is the total energy consumption of the air conditioning water pump, W LQpump is the total energy consumption of the cooling water pump, W LQtower is the total energy consumption of the cooling tower.

4. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 2 is characterized in that: The energy consumption function of the chiller is: W WaterChiller =f(Q KT , P Δ ) Where W WaterChiller is the total energy consumption of the chiller; Q KT is the total heat brought into the evaporator by the air-conditioning water, P Δ It is the pressure difference between the refrigerant suction and discharge from the compressor; Since the total heat brought into the evaporator by the air-conditioning water is equal to the heat exchanged by the air-conditioning water, and the pressure difference between the refrigerant sucked in and discharged by the compressor is equal to the difference between the condenser pressure and the evaporator pressure, then: Q KT =C Water ·(T KTin -T KTout )·F KTpump P Δ =P LNQ -P ZFQ In the above formula, C Water is the specific heat capacity of water; T KTin T is the air conditioning water inlet temperature before entering the evaporator; KTout F is the outlet temperature of the air-conditioning water after leaving the evaporator; KTpump is the air conditioning water flow; P LNQ is the absolute pressure of the high-pressure refrigerant in the condenser; P ZFQ is the absolute pressure of the low-pressure refrigerant in the evaporator; where: P ZFQ =B1·T KTout Wherein, B1 and C1 are comprehensive heat transfer coefficients, obtained by fitting the background data; T LQout T is the outlet temperature of cooling water after leaving the condenser; LQin is the cooling water inlet temperature before entering the condenser; The energy consumption model expression of the chiller is: <h2 style=";text-align:left;direction:ltr">W<h2 style=";text-align:left;direction:ltr"> WaterChiller <h2 style=";text-align:left;direction:ltr"> =A1·Q<h2 style=";text-align:left;direction:ltr"> KT <h2 style=";text-align:left;direction:ltr"> +A2·Q<h2 style=";text-align:left;direction:ltr"> KT <h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> +A3·Q<h2 style=";text-align:left;direction:ltr"> KT <h2 style=";text-align:left;direction:ltr"> ·P<h2 style=";text-align:left;direction:ltr"> Δ <h2 style=";text-align:left;direction:ltr"> +A4·Q<h2 style=";text-align:left;direction:ltr"> KT <h2 style=";text-align:left;direction:ltr"> +A5·Q<h2 style=";text-align:left;direction:ltr"> KT <h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> +A6 In the above formula, A1 to A6 are fitting coefficients, which are obtained by fitting the chiller energy consumption model expression based on multiple non-abnormal background data recorded in S1 at the same time.

5. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 2, characterized in that: The total energy consumption function of the air conditioning water pump is: W KTpump =f(N KTpump , HZ Set ) Where W KTpump is the total energy consumption of the air conditioning water pump; N KTpump The number of air conditioning water pumps that are turned on at the same time; HZ Set The frequency of a single air-conditioning water pump is the same for each air-conditioning water pump that is turned on. Set independent model expressions according to the number of air conditioning water pumps turned on: Where: n is the number of air conditioning water pumps turned on; D 1_1 ~D n_2 is the fitting coefficient, obtained by fitting multiple non-abnormal background data recorded in S1 at the same time in the above independent model expressions; The air conditioning water flow function is: F KTpump =f(W KTpump , N KTpump , H KToutin ) Where H KToutin The pressure difference between the outlet and inlet of the air conditioning water pump; Set independent model expressions according to the number of air conditioning water pumps turned on: Where: n is the number of pumps turned on; E 1_1 ~E n_5 is the fitting coefficient, obtained by fitting multiple non-anomalous background data recorded in S1 at the same time in the above independent model expression; where: Where: F 1_1 ~F 24_2 is the fitting coefficient, obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal; or, The total energy consumption function of the cooling water pump is: W LQpump =f(N LQpump , HZ SHt ) Where W LQpump is the total energy consumption of the cooling water pump; W LQpump The number of cooling water pumps that are turned on at the same time; HZ SHt The frequency of a single cooling water pump, and the frequency of each cooling water pump is the same; Set independent model expressions according to the number of cooling water pumps turned on: Where: n is the number of pumps turned on; G 1_1 ~G n_2 is the fitting coefficient, obtained by fitting multiple non-abnormal background data recorded in S1 at the same time in the above independent model expressions; The cooling water flow function is: F LQpump =f(W LQpump , N LQpump , H LQoutin ) Where H LQoutin is the pressure difference between the cooling water pump outlet and inlet; Set independent model expressions according to the number of cooling water pumps turned on: Where: n is the number of cooling water pumps turned on; H 1_1 ~H n_5 The fitting coefficients are obtained by fitting multiple non-anomalous background data recorded in S1 at the same time in the above independent model expressions; where: Where: I 1_1 ~I n_2 The fitting coefficient is obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal.

6. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 2, characterized in that: The total air volume function of the cooling tower fan and the total energy consumption function of the cooling tower are respectively: FL LQtower =f(HZ LQtower ,N LQtower ) W LQtower =f(HZ LQtower ,N LQtower ) Where, EL LQtower is the total air volume of the cooling tower, W LQtower is the total energy consumption of the cooling tower, HZ LQtower is the frequency of a single fan, N LQtower The number of cooling towers opened; the specific expression is: FL LQtower =c·N LQtower ·HZ LQower Where c is a constant, J1~J2 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal; Cooling tower heat dissipation Q LQtower The heat brought into the chiller by the air-conditioning water, the cooling load model and expression are as follows: Q LQtower =f(Q KT ) Q LQtower =(K1·Q KT +K2) Where: Q KT is the total heat brought into the evaporator by the air-conditioning water; K1~K2 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal; The cooling tower outlet water temperature function is: T LQin =f(T Air_w ,Q LQtower ,FL LQtower ,T Air_d -T LQavg ,T Air_d -T Air_w (T LQout -T LQwaterin )·Q LQtower ) Where: T LQin is the cooling tower outlet water temperature; T Air_w is the outdoor atmospheric wet-bulb temperature; T Air_d is the outdoor atmospheric dry bulb temperature; T LQavg is the average temperature of cooling water heat exchange; T LQwaterin is the feed water temperature in the cooling water tank; where: Where, T LQout is the cooling tower inlet water temperature; L1~L4 are fitting coefficients, which are obtained by fitting the above formula based on multiple background data recorded in S1 at the same time that are not abnormal.

7. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 1, characterized in that: S1 specifically includes the following steps: Arrange various working condition sensors inside and outside the computer room, establish communication links between each working condition sensor and the host computer, and use the host computer to perform real-time analysis and verification of the received data. According to the pre-set classification restrictions of each received data and its combination, the data without abnormalities after verification is written into the corresponding structure table of the relational database carried by the host computer to achieve orderly storage of data; select historical non-abnormal data as background data for fitting the mathematical relationship model between variables.

8. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 4 is characterized in that: S2 also includes the building air conditioning load model, and the building load function is: Q KT =f(T Air_d , RH Air , Lux Air ) Where Q KT This is the building load at this time, that is, the total heat brought into the evaporator by the air-conditioning water; T Air_d is the outdoor atmospheric dry bulb temperature; RH Air Lux is atmospheric humidity; Air is the current outdoor light intensity; the specific expression is as follows: Q KT =M1·T Air_d +M2·T Air_d 2 +M3·T Air_d ·RH Air +M4·RH Air +M5·Lux Air Where: M1~M5 are fitting coefficients, which are obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal; The average heat exchange temperature function of air conditioning water supply and return water is: Where, T KTavg is the average heat exchange temperature of air conditioning water supply and return water, T inner RH is the dry bulb temperature of the air in the representative area of ​​the building; inner is the air humidity in the representative indoor area, T Air_d is the outdoor atmospheric dry bulb temperature; RH Air is the atmospheric humidity; the specific expression is as follows: T KTavg =N1·T inner +N2·RH inner +N3·T Air_d +N4·RH Air +N5 Where: N1~N5 are fitting coefficients, which are obtained by fitting the above expression based on multiple background data recorded in S1 at the same time that are not abnormal.

9. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to any one of claims 2 to 8, characterized in that: In S3, the method for establishing the energy efficiency simulation system includes the following steps: S3-1: Set inputs, including environmental quantification and control strategy groups; The environmental quantification is as follows: In the above quantitative analysis, T Air_w is the outdoor atmospheric wet-bulb temperature, T Air_d is the outdoor atmospheric dry bulb temperature, RH Air Lux is atmospheric humidity; Air is the current outdoor light intensity; T inner is the indoor dry bulb temperature, RH inner The required humidity for the room; Time day is the current week number, Time hour is the current hour; where T Air_w and RH Air From the sensor, T inner and RH inner From artificial experience setting or setting by month; T LQwaterin The soft water temperature inside the cooling water make-up tank; The control strategy groups are as follows: Among them, N KTpump The number of air conditioning water pumps turned on, HZ Set The frequency of a single air-conditioning water pump, and the frequency of each air-conditioning water pump is the same; N LQpump The number of cooling water pumps turned on, HZ SHt The frequency of a single cooling water pump, and the frequency of each cooling water pump is the same; N LQtower The number of cooling towers opened, HZ LQtower The fan frequency of a single cooling tower is the same for each cooling tower that is turned on. S3-2: Correlation calculation: Input the environmental quantitative and control strategy groups, perform correlation point calculations, establish energy consumption relationships between models, and finally obtain the total energy consumption of the chiller W. WaterChiller , total energy consumption of air conditioning water pump W KTpump , total energy consumption of cooling water pump W LQpump And the total energy consumption of cooling tower W LQtower Finally, the total energy consumption W0 and system efficiency S0 of the computer room system are obtained as follows: W0=W WaterChiller +W KTpump +W LQpump +W LQtower Where Q KT The total heat brought by the air-conditioning water into the evaporator.

10. The method for real-time simulation and optimization of energy efficiency of a central air-conditioning room system according to claim 9, characterized in that: In S4, the real-time optimization method includes the following steps: S4-1: Set the optimization goal: take the energy efficiency of the computer room system as the goal, set the optimization variable to the computer room system efficiency S0, and the optimization direction to the positive direction; S4-2: Set constraints: During the simulation process, the optimization nodes are constantly monitored. If the working conditions of each node do not meet the constraints, the optimization is considered invalid. Constraint group 1 is the number of available units and variable frequency of each device, and constraint group 2 is the working condition range for safe operation of the equipment. Constraint group 1 is as follows: Where: M KTpump 、N KTpump_Max 、HZ Set Respectively represent the number of air conditioning water pumps that are turned on, the number of air conditioning water pumps that are available, and the frequency of the unified setting of air conditioning water pump input; N LQpump 、N LQpump_Max 、HZ SHt Respectively represent the number of cooling water pumps that are turned on, the number of cooling water pumps that are available, and the frequency of the unified setting of cooling water pump input; N LQtower 、N LQtower_Max 、HZ LQtower They represent the number of cooling towers that are turned on, the number of cooling towers that are available, and the frequency of the unified setting of cooling tower inputs respectively; Constraint group 2 is as follows: Where, T KTout P is the outlet temperature of the air-conditioning water after leaving the evaporator; LNQ is the absolute pressure of the high-pressure refrigerant in the condenser; P ZFQ is the absolute pressure of the low-pressure refrigerant in the evaporator; T Air_w is the outdoor atmospheric wet-bulb temperature; T LQin is the cooling tower outlet water temperature; F KTpump is the air conditioning water flow; F LQpump is the cooling water flow rate; S4-3: Use the optimization algorithm to perform iterative optimization, taking the variables of constraint condition 1 as the optimization items, and output the optimal control strategy.

Images