Building water chilling unit load distribution method based on joint simulation

Abstract

The invention discloses a building water chilling unit load distribution method based on joint simulation, and the method comprises the following steps: S100, building an air conditioning system model in simulation software, and reserving a load distribution proportion as an input port; s200, establishing a relation model with the energy efficiency ratio and the total energy consumption by taking the load distribution proportion of each unit and the cooling demand of the building as input variables and the energy consumption of each component of the air conditioning system as output; s300, the actual building hourly cooling demand prediction table serves as an independent variable of the building cooling demand, and the optimal load distribution proportion under the input of each building cooling demand is calculated through an optimization algorithm; and S400, making a hourly optimal load distribution proportion table in a research time period, calculating annual cumulative power consumption, and performing comparison evaluation with a conventional expert control strategy. The method is suitable for solving the optimal load proportion of the building water chilling unit and energy-saving optimization scenes of other air conditioning systems, and is high in accuracy, flexibility, reliability and practicability.

Description

Technical Field

[0001] The present invention relates to the technical field of air-conditioning unit load distribution, and in particular to a building chiller load distribution method based on joint simulation. Background Art

[0002] The rapid development of human society and the economy has led to massive energy consumption, and the greenhouse gases generated by this energy use are a significant contributor to global warming. In the building sector, energy consumption accounts for approximately 40% of global energy use, with varying proportions across regions. Forty percent of total building energy consumption is used to meet the electricity needs of chillers in heating, ventilation, and air conditioning (HVAC) systems. Therefore, exploring HVAC energy-saving strategies will make a significant contribution to mitigating global warming.

[0003] Chillers in air conditioning systems have significantly different COPs (Combined Energy Performance Ratios) at different load rates, meaning the cooling capacity achieved per unit of electricity consumed varies. When a chiller operates at a load rate corresponding to a lower COP, energy is wasted. Air conditioning systems often have more than one chiller, but rather multiple chillers operating in parallel. The cooling load borne by each chiller is controlled by controlling the chilled water flow rate. The chiller's total COP can be considered a multivariate function, with the cooling load borne by each chiller as the independent variable and the performance parameters of each unit as the parameters of the multivariate function. To maximize energy savings, it is necessary to find an appropriate load distribution ratio that maximizes the value of this multivariate function. Therefore, the process of minimizing energy consumption in an air conditioning system with multiple chillers operating in parallel is called the optimal chiller load (OCL).

[0004] To address the OCL problem, researchers have focused on finding the extreme value of the multivariate function representing chiller energy consumption after constructing a mathematical model. To this end, researchers have proposed various algorithms to continuously improve convergence speed, convergence stability, and ultimately energy-saving optimization results. They have also developed more tailored solutions for various application scenarios. While mathematical modeling can achieve maximum accuracy for a real air-conditioning system, this also requires considering more parameters and variables. Overly complex mathematical models inevitably affect the representation of the multivariate function representing the chiller's energy consumption. Consequently, researchers generally choose to ignore relatively unimportant factors, resulting in a multivariate function that ultimately deviates from the actual operating conditions of the chillers in the air-conditioning system. Furthermore, the energy consumption of pumps, electric valves, and other components of the air-conditioning system used to control the partial load of parallel chillers is difficult to flexibly incorporate into the calculation. Simulation is an effective solution for accurately modeling the entire HVAC system (an integrated system that integrates heating, ventilation, and air conditioning functions) and monitoring the energy consumption of each component of the system. Some scholars have verified the optimization scheme by simulation in TRNSYS software (transient system simulation program) after proposing it. However, simulation is only used as a verification method and does not reflect the potential of simulation in solving OCL problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a building chiller load distribution method based on joint simulation. By introducing Dymola / Simulink joint simulation to replace the energy consumption multivariate function in solving the load distribution problem of multiple chillers, the problems of inaccurate chiller energy consumption function due to the complexity of the air-conditioning system model and the inability to flexibly incorporate the energy consumption of other components such as water pumps into the calculation during the chiller load optimization process are solved. The method of the present invention can accurately optimize the load distribution, improve the convergence efficiency, effectively avoid falling into the local optimal solution, and significantly improve the energy saving effect.

[0006] To achieve the above object, the present invention adopts a technical solution: providing a building chiller load distribution method based on joint simulation, comprising the following steps:

[0007] S100: Establishing an air conditioning system model in the simulation software; at the same time, reserving a load distribution ratio as an input port of the air conditioning system model in the simulation software;

[0008] S200: In the air conditioning system model of the simulation software, the load distribution ratio of each unit and the building cooling demand are used as input variables, and the energy consumption of each component of the air conditioning system is used as output. A relationship model with the energy efficiency ratio and total energy consumption is established;

[0009] S300: Using a relational model and an actual hourly cooling demand forecast table of a building as an independent variable of the building cooling demand, an optimization algorithm is used to calculate the optimal load distribution ratio under the cooling demand input of each building;

[0010] S400: Prepare an hourly optimal load distribution ratio table within the study period, calculate the annual cumulative power consumption, and compare and evaluate it with the conventional expert control strategy.

[0011] Preferably, in the present technical solution, in the step S100, the air-conditioning system model is established based on the structure of the actual air-conditioning system, and the air-conditioning system components in the voxel library are used in the simulation software to model the air-conditioning system that needs to be optimized, and the various performance parameters of the system in the modeling are specified based on the unit performance data obtained based on the actual operating effect of the unit.

[0012] Preferably, in the present technical solution, in step S100, the curve fitting coefficient expression of the energy efficiency ratio C0P of the chiller in the air-conditioning system model changing with the load rate is:

[0013] η PL =a0+a1·y PL +a2·y PL 2 +a3·y PL 3 + · · #

[0014] Among them, η PL : is the ratio of the actual energy efficiency ratio COP to the rated energy efficiency ratio COP; a n : is the fitting coefficient; y PL : is the load rate.

[0015] Preferably, in this technical solution, the unit performance parameters include steady-state hydraulic characteristic parameters of the pipe network, which are expressed by the following formula:

[0016]

[0017] Where v1 is the average velocity at the initial section of the pipe; v2 is the average velocity at the final section of the pipe; △P is the pressure loss between the two points at the beginning and end of the pipe; g is the acceleration due to gravity; Z1 and Z2 are the heights of point 1 and point 2 respectively; and γ is the specific gravity of the fluid.

[0018] The pressure loss ΔP in the pipe section is the sum of the total resistance loss along the pipe section and the total local head loss between the two sections at the beginning and end of the pipe section. The calculation formula is:

[0019]

[0020] Where ΔP y : Total pressure loss along the pipe section; ΔP jb: total loss of local resistance in the pipe section; R: resistance loss rate along the pipe section, Pa / m; l: length of the pipe section, m; (: local resistance coefficient; d: diameter of the pipe; ρ: density of the fluid; λ: friction coefficient.

[0021] Preferably, in this technical solution, the unit performance parameters include the performance parameters of the water pump, the performance of the water pump is determined by the characteristics of the pipe network and the water pump itself, and the function of changing the supply flow synchronously with the user load is achieved by adjusting the speed of the water pump;

[0022] Pump head H (m), pump efficiency η and hourly flow rate Q l The mathematical model is as follows:

[0023]

[0024] Where, n: actual speed of the pump, 1 / min; n o : Rated speed of the pump, 1 / min; Q l : Pump delivery flow, m 3 / h; η: water pump efficiency; k i : water pump characteristic fitting coefficient;

[0025] The calculation formula for the water pump input power W is as follows:

[0026]

[0027] Where, H: pump head;

[0028] When the medium flows through the pipe network, heat loss will occur. The formula for calculating heat loss Q is:

[0029]

[0030] Where: Q: heat loss of the transported fluid; D o : Pipe outer diameter; D i : Pipeline inner diameter L: Pipeline length, m; ρ: Fluid density; c p : Specific heat capacity of fluid, J / (kg·℃); ΔT: Fluid temperature difference, ℃; k: Thermal conductivity of insulation material.

[0031] Preferably, in the present technical solution, in step S200, the load ratio of each chiller is reserved as input and the energy consumption of each component is output by modeling in step S102. The model is encapsulated as a functional module (FMU module) and imported into the integration platform (Simulink). A relationship model with energy efficiency ratio and total energy consumption is established, which is expressed by the following formula:

[0032] P chi =P chi1 +P chi2 +P chi3

[0033]

[0034] Among them, P chi1 、P chi2 、P chi3 is the power consumption of the first, second and third chillers, output from the simulation results, P chi is the total energy consumption of multiple chillers; P demand is the user side load power, W CoP is the total energy efficiency ratio of multiple chillers.

[0035] Preferably, in this technical solution, in step S300, the optimization algorithm is an iterative algorithm for finding an extreme value;

[0036] The Dymola simulation system is encapsulated and interconnected with the iterative algorithm for finding extreme values through the Simulink interface to achieve automatic iterative calling.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) The present invention uses Dymola software to model the air-conditioning system, ensuring the accuracy and comprehensiveness of the system energy consumption calculation, and supports the update of components such as chillers, water pumps, and valves, facilitating the flexible calculation of the energy consumption of each component.

[0039] (2) In the case of optimizing the energy consumption of building chillers, the present invention can accurately optimize the load distribution and achieve significant energy-saving effects, and has good applicability to large public buildings. By using Dymola / Simulink for joint simulation and control, it can be easily expanded to more heating, ventilation and air conditioning (HVAC) systems or other complex energy optimization problems.

[0040] (3) The iterative algorithm for finding the extreme value adopted by the present invention can efficiently approach the global optimal solution within a limited number of iterations, thereby improving the convergence efficiency and effectively avoiding falling into the local optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is the overall structural diagram of the actual chiller of the present invention;

[0042] Figure 2 This is the overall structural diagram of the air conditioning system modeled in the simulation software Dymola according to the present invention;

[0043] Figure 3 The structure diagram of the present invention in Dymola / Simulink joint simulation;

[0044] Figure 4is a flow chart of the allocation method of the present invention;

[0045] Figure 5 Schematic diagram comparing the energy efficiency ratios of the chiller of the present invention using an iterative algorithm control for finding extreme values and an expert strategy under different load distribution strategies. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; the same or similar concepts or processes may not be repeated in some embodiments.

[0048] The present invention provides a building chiller load distribution method based on joint simulation, based on Figure 1 、 Figure 2 and Figure 3 The structural diagram of Figure 4 The flow chart of the building chiller load distribution method based on joint simulation is shown, which includes the following steps:

[0049] S100: Establishing an air conditioning system model in the simulation software; at the same time, reserving a load distribution ratio as an input port of the air conditioning system model in the simulation software;

[0050] S200: In the air conditioning system model of the simulation software, the load distribution ratio of each unit and the building cooling demand are used as input variables, and the energy consumption of each component of the air conditioning system is used as output. A relationship model with the energy efficiency ratio and total energy consumption is established;

[0051] S300: Using a relational model and an actual hourly cooling demand forecast table of a building as an independent variable of the building cooling demand, an optimization algorithm is used to calculate the optimal load distribution ratio under the cooling demand input of each building;

[0052] S400: Prepare an hourly optimal load distribution ratio table within the study period, calculate the annual cumulative power consumption, and compare and evaluate it with the conventional expert control strategy.

[0053] In this embodiment, the preferred technical solutions are as follows:

[0054] S100: Establish an air conditioning system model in the simulation software;

[0055] In this embodiment, the preferred simulation software is the Dynamic Modeling Laboratory multidisciplinary system simulation platform (Dymola), a powerful multi-domain physical modeling and simulation software widely used in fields such as automotive, aerospace, robotics, and industrial automation. It uses a component-based modeling approach, with models constructed from reusable components. It can describe the interactions between different physical domains (such as mechanical, thermal, and electrical), supports the object-oriented modeling language Modelica, and offers a high degree of integration, multi-domain simulation capabilities, and openness.

[0056] The application of Dymola in the field of air conditioning systems has the following advantages:

[0057] (1) Multi-physics coupling: Air conditioning systems involve the interaction of multiple physical fields, such as thermodynamics, fluid mechanics, heat transfer, and electrical control. Dymola can couple modeling and simulation of these multi-physics fields on a single platform, eliminating the need for multiple specialized software for joint simulation. This avoids data transmission and interface issues between different software, and improves modeling and simulation efficiency.

[0058] (2) Component-based and modular modeling: Dymola, based on the Modelica language, adopts an object-oriented modeling approach and supports component-based and modular modeling. Users can model each component of the air conditioning system (such as the compressor, condenser, evaporator, expansion valve, etc.) as an independent component, and then build the entire system model by connecting these components. This modeling approach not only improves the flexibility and reusability of the modeling, but also facilitates the expansion and modification of the system.

[0059] (3) Powerful symbolic manipulation: Dymola has powerful symbolic manipulation capabilities that can automatically process complex mathematical equations and physical models and efficiently solve differential algebraic equations (DAEs). This allows users to focus on the physical meaning of the model and the design of the system without having to pay too much attention to the details of the mathematical solution.

[0060] (4) Efficient solvers: Dymola has built-in multiple efficient solvers that can quickly and accurately solve complex system models. For systems such as air conditioning systems that contain a large number of nonlinear equations and dynamic characteristics, Dymola's solvers can provide high-precision simulation results, helping users better understand and optimize system performance.

[0061] Based on the superior characteristics of Dymola, in step S100, this embodiment uses the structure and parameters of the air-conditioning system and the actual chiller (such as Figure 1) In the simulation software, establish Figure 2 The air conditioning system model shown in FIG; and the load distribution ratio is reserved as an input port. More specifically, according to the structure of the actual air conditioning system (such as Figure 1 As shown), the air conditioning system that needs to be optimized is modeled using the air conditioning system components in the Building library (voxel library) in the Dymola software (as shown Figure 2 ), and based on the unit performance data obtained from actual operation, the performance parameters of the system in the modeling are specified to ensure that the simulation results are representative of the actual unit. Specifically, in step S100, specific unit performance parameters and load rate energy efficiency ratio curves are set based on the actual chiller, and a complete air conditioning system model is constructed to achieve accurate modeling and simulation of the optimization objectives.

[0062] In this embodiment, in step S100, the actual chiller system consists of three units connected in parallel. The actual operating performance is calculated by collecting operating data of the chiller at different load rates. Because the operating data does not have an obvious linear relationship, a nonlinear polynomial model is used for fitting. The curve fitting coefficient of the chiller's COP as it changes with the load rate in the air conditioning system model is designed as shown in formula (1):

[0063] η PL =a0+a1·y PL +a2·y PL 2 +a3·y PL 3 +··# (1)

[0064] Where η PL : is the ratio of the actual energy efficiency ratio COP to the rated energy efficiency ratio COP; a n : is the fitting coefficient; n: a positive integer; y PL : is the load rate.

[0065] In simulation software, air conditioning system modeling involves establishing a characteristic model of the unit based on the unit's operating parameters, including the steady-state hydraulic characteristics of the pipe network and the performance parameters of the water pumps.

[0066] 1. Steady-state hydraulic characteristic parameters of the pipe network

[0067] The Bernoulli equation can describe the steady-state hydraulic characteristics of the air conditioning system network:

[0068]

[0069] Where, v1: average flow velocity at the initial section of the pipe; v2: average flow velocity at the end section of the pipe; △P: pressure loss of the fluid between the two points at the beginning and end of the pipe: acceleration due to gravity; Z1, Z2: heights of point 1 and point 2; γ: specific gravity of the fluid.

[0070] The pressure loss ΔP in the pipe section is the sum of the total resistance loss along the pipe section and the total local head loss between the two sections at the beginning and end of the pipe section. The calculation formula is:

[0071]

[0072] Where ΔP y : Total pressure loss along the pipe section; ΔP jb : total loss of local resistance in the pipe section; R: resistance loss rate along the pipe section, Pa / m; l: length of the pipe section, m; (: local resistance coefficient; d: diameter of the pipe; μ: density of the fluid; λ: friction coefficient.

[0073] 2. Performance parameters of water pump

[0074] The performance of the water pump is determined by the characteristics of the pipe network and the water pump itself, and the pump speed is adjusted to achieve the function of changing the supply flow in sync with the user load. l The mathematical model is as follows:

[0075]

[0076] Where, n: actual speed of the pump n0: rated speed of the pump; Q l : Pump delivery flow η: Pump efficiency; k1, k2, k3, k4, k5: Pump characteristic fitting coefficients.

[0077] The calculation formula for the water pump input power W is as follows:

[0078]

[0079] Where, H: pump head, m.

[0080] When the medium flows through the pipe network, heat loss will occur. The heat loss Q can be calculated using the following formula:

[0081]

[0082] Where: Q: heat loss of the conveying fluid, J; D o 、D i : outer and inner diameters of the pipe, m; L, pipe length, m; ρ: fluid density, kg / m 3 ;c p : specific heat capacity of fluid, J / (kg·℃); ΔT: fluid temperature difference, ℃; k: thermal conductivity of insulation material, W / (m·K).

[0083] Based on the above characteristics, the present invention builds models for various components of the air-conditioning system in the Dymola simulation system, making it relatively simple to build the air-conditioning system model. In addition, due to the powerful simulation capabilities of the Dymola simulation system, the simulation calculation output results are highly consistent with the data of the actual air-conditioning system, which ensures the accuracy and comprehensiveness of the energy consumption calculation.

[0084] S200: In the air-conditioning system of the simulation software, the Dymola model is first exported as an FMU module and imported into Simulink. Then, the load distribution ratio of each unit and the building cooling demand are used as input variables, and the energy consumption of each component of the air-conditioning system is used as output to establish a relationship model with the energy efficiency ratio and total energy consumption.

[0085] In step S200, the load ratio of each chiller (which can also be regarded as load) reserved in the modeling of step S100 is used as input, and the energy consumption of each component is used as output. The model is encapsulated as a functional module (FMU module) and imported into the integration platform (Simulink, which is also a visual simulation tool in MATLAB). The calculation of energy efficiency ratio, total energy consumption, etc. is constructed as shown in formula (8):

[0086] P chi =P chi1 +P chi2 +P chi3

[0087]

[0088] Among them, P chi1 、P chi2 、P chi3 is the power consumption of the first, second and third chillers, output from the simulation results, P chi is the total energy consumption of multiple chillers; P demand is the user side load power, W CoP is the total energy efficiency ratio of multiple chillers. demand , the load distribution ratio of each unit x1, x2, x3 is the independent variable, the total energy consumption of multiple chillers is the black box of the function, and the function value is obtained by simulation under different independent variable inputs. The function P chi The expression is:

[0089] P chi =f(P demand , x1, x2, x3) (9)

[0090] S300: By using a relational model and an actual hourly cooling demand forecast table of a building as an independent variable of the building cooling demand, an optimization algorithm is used to calculate the optimal load distribution ratio under each building cooling demand input.

[0091] Specifically, the input of the Dymola simulation model is the load distribution of each chiller, and the output is the energy consumption per unit time under the distribution strategy. The model has multiple extreme values. To facilitate algorithm call, this simulation model is encapsulated and interconnected with the iterative algorithm for finding extreme values through the integrated platform (Simulink) interface to achieve automatic iterative call. More specifically, by setting a reasonable single simulation time, the simulation time should be less than the minimum time unit of the user-side load input (generally 1 hour), and the system reaches a steady state, the simulation results are used as output and integrated into a multivariate function with the chiller load distribution ratio as the independent variable and the total energy consumption of the air-conditioning system as the function (see formula (3)). Since the multivariate function has multiple extreme values, in step S300, it is preferred to use the iterative algorithm for finding extreme values of the present invention as the optimization algorithm to find the minimum value of the multivariate function, that is, to find the chiller load distribution ratio strategy that minimizes total energy consumption.

[0092] In the present invention, the load distribution of building chillers has multiple extreme value solutions. The iterative algorithm for finding extreme values is used to introduce a mutation operation with the idea of genetic algorithm, so that the particle swarm optimization algorithm can jump out of the local optimum in time, thereby expanding the searchable area and improving the search ability for the global optimal solution. To achieve a balance between rapid convergence and global search, the particle fitness distribution is utilized to dynamically adjust the algorithm parameters, including inertia weight and learning factor, so that high-fitness particles retain the existing search trend as much as possible, while low-fitness particles are given the opportunity to strengthen exploration. Specifically, the user-side load of each hour in the study time period (for example, 8760 hours in a year, 2160 hours in a quarter, 720 hours in a month, etc.) is used as a fixed input, and the optimal load distribution ratio under each hour's load input is obtained through step S300 (for example, in the first hour, the user-side load is 3000kW, and the distribution method that minimizes the total energy consumption of each unit is 60%, 20%, and 20%, and then the optimal distribution ratio for each subsequent hour is obtained in sequence). Based on the results of each hour, the optimal load distribution ratio table for the study period can be listed hourly, and the total power consumption for the study period can be obtained by adding up the power consumption for each hour.

[0093] Conventional expert control strategies refer to rules for load distribution that are manually set based on expert experience. For example: if the maximum loads of the three units are equal and the total load is low, only the first chiller is turned on; when the load exceeds 80% of the first unit's maximum load, the second chiller is turned on, and the load is evenly shared by the first and second units; when the total load exceeds 80% of the sum of the maximum loads of the first two units, the third chiller is turned on, and the load is evenly shared by all three units. The above is just one example of an expert strategy. In actual situations, when the units have different models, maximum loads, and COP-load rate performance curves, the expert strategy may be more complex and deviate significantly from the optimal load distribution ratio. Therefore, modeling and algorithm optimization are needed to determine a better load distribution ratio.

[0094] Therefore, the total power consumption of the study period can be calculated according to the conventional expert strategy and the matching strategy obtained by modeling-algorithm optimization, and the saved power consumption of the study period can be obtained.

[0095] Simulation test:

[0096] A simulation test was conducted under the cooling demand scenario of a large public building, such as Figure 5 The results show that IPSO control generally achieves higher energy efficiency than conventional and average control methods, and the resulting load distribution scheme reduces power consumption by approximately 3% compared to the expert strategy. This demonstrates that the proposed method offers greater accuracy and flexibility in determining the optimal load ratio for chillers, making it suitable for determining the optimal load ratio for building chillers and other energy-saving optimization scenarios for air conditioning systems, demonstrating excellent reliability and practicality.

[0097] In summary, the present invention replaces the multivariate function of energy consumption of the chiller with the Dymola / Simulink joint simulation of the air-conditioning system modeling in the problem of optimizing the load of the chiller, abandons the mathematical derivation of the multivariate function of energy consumption, and uses the independent variables such as the refrigeration load and the partial load rate of the unit instead, and the output of the simulation result is used as the function value. After the modeling is completed, a suitable optimization algorithm is adopted, that is, the process of calling the function for calculation is converted into calling the simulation and obtaining the output result, so as to calculate the optimal load ratio of the chiller that makes the air-conditioning system energy consumption the lowest, tap the energy-saving potential of the air-conditioning system, improve operating efficiency and reduce operating costs. The method of the present invention is based on the Dymola / Simulink joint simulation to make the model take into account the degree of accuracy and updatability, which is convenient for directly calling the Simulink model for optimization in the optimization algorithm.

[0098] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A building chiller load distribution method based on joint simulation, characterized in that: The steps include: S100: Establishing an air conditioning system model in the simulation software; at the same time, reserving a load distribution ratio as an input port of the air conditioning system model in the simulation software; S200: In the air conditioning system model of the simulation software, the load distribution ratio of each unit and the building cooling demand are used as input variables, and the energy consumption of each component of the air conditioning system is used as output. A relationship model with the energy efficiency ratio and total energy consumption is established; S300: Using a relational model and an actual hourly cooling demand forecast table of a building as an independent variable of the building cooling demand, an optimization algorithm is used to calculate the optimal load distribution ratio under the cooling demand input of each building; S400: Prepare an hourly optimal load distribution ratio table within the study period, calculate the annual cumulative power consumption, and compare and evaluate it with the conventional expert control strategy.

2. The building chiller load distribution method based on joint simulation according to claim 1 is characterized in that: In step S100, the air-conditioning system model is established based on the structure of the actual air-conditioning system. The air-conditioning system to be optimized is modeled in the simulation software using the air-conditioning system components in the voxel library, and the various performance parameters of the system in the modeling are specified based on the unit performance data obtained from the actual operation effect of the unit.

3. The load distribution method for building chillers based on joint simulation according to claim 2 is characterized in that: In step S100, the curve fitting coefficient expression of the energy efficiency ratio COP of the chiller in the air-conditioning system model changing with the load rate is: <h2 style=";text-align:left;direction:ltr">η<h2 style=";text-align:left;direction:ltr"> PL <h2 style=";text-align:left;direction:ltr"> =a0+a1·y<h2 style=";text-align:left;direction:ltr"> PL <h2 style=";text-align:left;direction:ltr"> +a2·y<h2 style=";text-align:left;direction:ltr"> PL <h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> +a3·yp<h2 style=";text-align:left;direction:ltr"> L <h2 style=";text-align:left;direction:ltr"> 3 <h2 style=";text-align:left;direction:ltr"> +··# Among them, η PL : is the ratio of the actual energy efficiency ratio COP to the rated energy efficiency ratio COP; a n : is the fitting coefficient; y PL : is the load rate.

4. The building chiller load distribution method based on joint simulation according to claim 2 is characterized in that: The unit performance parameters include the steady-state hydraulic characteristic parameters of the pipe network, which are expressed by the following formula: Where v1 is the average velocity at the initial section of the pipe; v2 is the average velocity at the final section of the pipe; ΔP is the pressure loss between the beginning and end points of the pipe; g is the acceleration due to gravity; Z1 is the height of point 1 in the pipe; Z2 is the height of point 2; and γ is the specific gravity of the fluid. The pressure loss ΔP in the pipe section is the sum of the total resistance loss along the pipe section and the total local head loss between the two sections at the beginning and end of the pipe section. The calculation formula is: Where ΔP y : Total pressure loss along the pipe section; ΔP jb : total loss of local resistance in the pipe section; R: resistance loss rate along the pipe section; l: length of the pipe section; (: local resistance coefficient; d: diameter of the pipe; ρ: density of the fluid; λ: friction coefficient.

5. The building chiller load distribution method based on joint simulation according to claim 2 is characterized in that: The performance parameters of the unit include the performance parameters of the water pump. The performance of the water pump is determined by the characteristics of the pipe network and the water pump itself. The water pump speed is adjusted to achieve the function of changing the supply flow synchronously with the user load. Pump head H (m), pump efficiency η and hourly flow rate Q l The mathematical model is as follows: Where, n: actual speed of the water pump; n0: rated speed of the pump; Q l : Pump delivery flow rate; η: pump efficiency; k i : water pump characteristic fitting coefficient; The calculation formula for the water pump input power W is as follows: Where, H: pump head; When the medium flows through the pipe network, heat loss will occur. The formula for calculating heat loss Q is: Where: Q: heat loss of the transported fluid; D o : Pipe outer diameter; D i : inner diameter of the pipe; L, pipe length; ρ: fluid density; c p : specific heat capacity of fluid; ΔT: fluid temperature difference; k: thermal conductivity of insulation material.

6. The building chiller load distribution method based on joint simulation according to claim 1 is characterized in that: In step S200, the load ratio of each chiller reserved in the modeling in step S102 is used as input, and the energy consumption of each component is used as output. The model is encapsulated as a functional module (FMU module) and imported into the integration platform (Simulink). A relationship model with energy efficiency ratio and total energy consumption is established, which is expressed by the following formula: P chi =P chi1 +P chi2 +P chi3 Among them, P chi1 、P chi2 、P chi3 is the power consumption of the first, second and third chillers, output from the simulation results, P chi is the total energy consumption of multiple chillers; P demand is the user side load power, W COP is the total energy efficiency ratio of multiple chillers.

7. The building chiller load distribution method based on joint simulation according to claim 1 is characterized in that: In step S300, the optimization algorithm is an iterative algorithm for finding an extreme value; The Dymola simulation system is encapsulated and interconnected with the iterative algorithm for finding extreme values through the Simulink interface to achieve automatic iterative calling.

Images