Optimization method, device and equipment of air conditioning system and storage medium

Abstract

The invention provides an optimization method, device and equipment of an air conditioning system and a storage medium. The method comprises the following steps that annual operation data and actually-measured operation data of the air conditioning system are collected, annual meteorological typical day data are generated through the annual operation data, and meteorological typical year data are generated based on meteorological data of recent years; establishing a simulation building model in TRNSYS, inputting meteorological condition data, building information, personnel, illumination, equipment parameters and actual multi-mode operation conditions of the air conditioning system, and generating simulation optimization operation data of the air conditioning system; and deviation analysis is conducted on the actually-measured operation data and the simulated and optimized operation data, and when the deviation exceeds a preset threshold value, the air conditioning system automatically adjusts equipment operation parameters, so that the actual operation data of the air conditioning system is recovered to the simulated and optimized operation data. According to the method, the actual air conditioning system operation data and the analogue simulation data are combined, the operation strategy is dynamically adjusted, and the energy utilization efficiency is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of HVAC operation and design optimization, and in particular to an optimization method, apparatus, equipment and storage medium for an air conditioning system. Background Technology

[0002] Existing HVAC system designs are typically based on static maximum daily load assumptions, lacking the ability to adapt to various operating conditions and dynamic load characteristics in actual operation. Traditional meteorological parameter systems rely on outdated data (such as meteorological years from 1971 to 2000), failing to reflect current climate characteristics, resulting in significant deviations between design daily parameters and actual operation. Furthermore, the system often operates at a "powered but not efficient" pace, with equipment running outside of its high-efficiency range, leading to high energy consumption and low operating efficiency.

[0003] Secondly, since the changes in parameter settings of different equipment and units affect hourly energy consumption, fixed HVAC system control strategies are no longer suitable. It is necessary to explore and optimize operating strategies that take into account meteorological data, building information, personnel, lighting, equipment parameters, etc. Therefore, a suitable and flexible HVAC system control strategy is needed. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an optimization method and apparatus for air conditioning systems, aiming to solve the problem of energy waste in the operation of existing air conditioning systems.

[0005] To achieve the above objectives, the present invention provides an optimization method for an air conditioning system, comprising the following steps:

[0006] Collect the annual operating data and measured operating data of the air conditioning system, use the annual operating data to generate typical daily meteorological data for the whole year, and generate typical annual meteorological data based on meteorological data from recent years.

[0007] A simulated building model is established in TRNSYS. Meteorological condition data, building information, personnel, lighting, equipment parameters, and the actual multimodal operating conditions of the air conditioning system are input. Simulated optimized operating data of the air conditioning system is generated. The meteorological condition data includes the typical meteorological year data and the typical meteorological day data for the whole year.

[0008] The measured operating data and the simulated optimized operating data are analyzed for deviation. When the deviation exceeds a preset threshold, the air conditioning system automatically adjusts the equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

[0009] The process of establishing a simulated building model in TRNSYS, inputting meteorological condition data, building information, personnel, lighting, equipment parameters, and the actual multimodal operating conditions of the air conditioning system, and generating simulated optimized operating data for the air conditioning system, further includes: optimizing the simulated building model using response surface methodology, which includes: systematically changing the input variables of the simulated building model using experimental design methods to collect experimental data; setting equipment operating limits for the air conditioning system; based on the TRNSYS simulation output data, minimizing system energy consumption as the optimization objective, describing the relationship between system energy consumption and input variables using a quadratic polynomial model; and solving the simulated building model using an optimization algorithm.

[0010] The setting of equipment operating limits for the air conditioning system further includes: setting the water temperature range on the heat source side and the load side; designing the water flow range on the heat source side and the load side according to the pipeline design specifications; and setting the rated value of the operating power of the heat source equipment.

[0011] The quadratic polynomial model of the system's energy consumption is as follows:

[0012]

[0013] Where E represents the system energy consumption in kW; i, j = 1, 2, 3, 4; x1 to x4 represent the input variables, which are respectively represented as the temperature difference between the cold and hot source sides, the water flow rate between the cold and hot source sides, the water temperature difference between the load side, and the water flow rate between the load side; β represents the regression coefficient, which is determined by least squares fitting.

[0014] The process of generating typical meteorological year data based on meteorological data from recent years also includes the following steps: standardizing the meteorological data from recent years, extracting key features using principal component analysis, and generating the typical meteorological year data through orthogonal transformation.

[0015] The standardization process of the meteorological data from recent years, the extraction of key features using principal component analysis, and the generation of typical meteorological year data through orthogonal transformation further include:

[0016] The meteorological data from recent years, including dry-bulb temperature, wet-bulb temperature, solar radiation, and wind speed, are standardized according to the following formula.

[0017]

[0018] Where X represents the meteorological data from the past few years, X norm The data represents standardized meteorological data from recent years, where μ is the mean and σ is the standard deviation.

[0019] Calculate the covariance matrix of the standardized meteorological data from recent years;

[0020] Extract the eigenvalues ​​of the covariance matrix and the corresponding eigenvectors of the eigenvalues, and select the top m principal components with a cumulative contribution rate ≥ 85% from the eigenvalues, where m is a positive integer;

[0021] The meteorological data from recent years are projected onto a space composed of m principal components through orthogonal transformation to generate a reconstructed dataset, wherein the meteorological data of a typical year is the reconstructed dataset.

[0022] The process of generating typical meteorological day data for the whole year using the annual operational data further includes the following steps: performing time series clustering on the annual operational data using the K-means algorithm to generate the typical meteorological day data for the whole year, which includes the following sub-steps:

[0023] Extract hourly variation trend coefficients of meteorological elements based on actual meteorological observation data of the past 30 years or more, and update meteorological data in real time;

[0024] The hourly variation trend coefficients of different meteorological elements are combined into an hourly variation trend coefficient matrix. Cluster analysis is performed on the hourly variation trend coefficient matrix, and the centroids of the clusters are extracted as hourly variation coefficients.

[0025] By combining the hourly coefficients with the daily average and daily difference of the main parameters of the design day, the hourly values ​​of the design day are generated in reverse.

[0026] The hourly values ​​of the design day are used to generate meteorological parameter variation curves corresponding to various standard design day types.

[0027] The various standard design days include seven types: summer air conditioning, dehumidification, and fresh air; and winter air conditioning, heating, humidification, and fresh air. The main parameters of the design days include dry-bulb temperature, moisture content, and enthalpy.

[0028] The optimization method for the air conditioning system also includes: applying the optimization results to regionalized case studies for different climate regions and building types to verify their applicability; and using air source heat pumps for cooling in summer and ground source heat pumps for cooling or heating at other times, based on the load differences between winter heating and summer cooling.

[0029] On the other hand, the present invention also provides an air conditioning system optimization device, which employs the above-described air conditioning system optimization method, wherein the air conditioning system optimization device includes at least:

[0030] The data acquisition and processing module is used to collect the annual operation data and measured operation data of the air conditioning system, generate typical daily meteorological data for the whole year using the annual operation data, and generate typical annual meteorological data based on meteorological data from recent years.

[0031] The model building module is used to build a simulated building model in TRNSYS. It inputs meteorological condition data, building information, personnel lighting, equipment parameters and the actual operating conditions of the air conditioning system, and generates simulated optimized operating data of the air conditioning system. The meteorological condition data includes the typical meteorological year data and the typical meteorological day data of the whole year.

[0032] The adjustment module is used to perform deviation analysis between the measured operating data and the simulated optimized operating data. When the deviation exceeds a preset threshold, the air conditioning system automatically adjusts the equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

[0033] On the other hand, the present invention also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the above-described method for optimizing the air conditioning system.

[0034] On the other hand, the present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the above-described optimization method for the air conditioning system.

[0035] As can be seen from the above solutions, the advantages of the present invention are:

[0036] This invention achieves a significant improvement in energy utilization efficiency by dynamically adjusting the operating strategy by combining actual operating data and simulation data.

[0037] This invention generates typical meteorological data for the whole year by integrating meteorological parameters of the new standard design days (including seven types of design days: summer air conditioning, dehumidification, fresh air, and winter heating and humidification), combined with dynamic hourly variation coefficients and cluster analysis, which significantly improves the accuracy of load simulation. Attached Figure Description

[0038] Figure 1A The flowchart (I) shows the optimization method of the air conditioning system of the present invention.

[0039] Figure 1B The flowchart (II) shows the optimization method of the air conditioning system of the present invention.

[0040] Figure 2 for Figure 1A Flowchart of step S10;

[0041] Figure 3 for Figure 2 Flowchart of step S100;

[0042] Figure 4 for Figure 2 Flowchart of step S101;

[0043] Figure 5 for Figure 1A Flowchart of step S20 (Part 1);

[0044] Figure 6 for Figure 1A Flowchart of step S20 (II);

[0045] Figure 7 for Figure 6 Flowchart of step S230 (Part 1);

[0046] Figure 8 for Figure 6 Flowchart of step S230 (II);

[0047] Figure 9 The flowchart (III) shows the optimization method of the air conditioning system of the present invention.

[0048] Figure 10 A comparison chart of cooling and heating capacities for air-source heat pumps and ground-source heat pumps;

[0049] Figure 11 A comparison chart of COP for air source heat pumps and heat pump units;

[0050] Figure 12 A comparison chart of total energy consumption for air source heat pump and ground source heat pump systems;

[0051] Figure 13A This is a normal distribution graph of the residuals;

[0052] Figure 13B A comparison chart of residuals and equation predictions;

[0053] Figure 13C A graph showing the correspondence between predicted values ​​and actual experimental values;

[0054] Figure 13D A 3D scatter plot;

[0055] Figure 14 A graph comparing actual measured values ​​and equation-predicted values;

[0056] Figure 15 A graph showing the system's energy consumption versus inlet and outlet water temperature and flow rate.

[0057] Figure 16 This is a schematic diagram of the structure of the optimization device for the air conditioning system of the present invention;

[0058] Figure 17 for Figure 16 A schematic diagram of the data acquisition and processing module in the middle;

[0059] Figure 18 for Figure 16 A schematic diagram of the structure of the model building module;

[0060] Figure 19 This is another structural schematic diagram of the optimization device for the air conditioning system of the present invention;

[0061] Figure 20 This is a schematic diagram of the structure of the electronic device of the present invention;

[0062] In the attached figures, the following labels are used:

[0063] 1-Optimization methods for air conditioning systems;

[0064] 2- Optimization device for air conditioning system;

[0065] 20 - Data Acquisition and Processing Module;

[0066] 200 - First Processing Unit;

[0067] 201 - Second Processing Unit;

[0068] 21-Model building module;

[0069] 210 - Building Block;

[0070] 211-Analog Data Setting Unit;

[0071] 212 - Simulation Unit;

[0072] 213 - Optimization Unit;

[0073] 22-Adjustment module;

[0074] 23-Regional Analysis Module;

[0075] 24-Season Switching Module;

[0076] 3-Electronic devices;

[0077] 30-Processor;

[0078] 31-Memory;

[0079] 310 - Computer programs;

[0080] S10~S50, S100, S101, S1000~S1003, S1010~S1013, S200~S230, S2300~S2302, S2310, S2311 - steps. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that references to "an embodiment," "embodiment," "example embodiment," etc., in the specification refer to the described embodiment including specific features, structures, or characteristics, but not necessarily including these specific features, structures, or characteristics. Furthermore, such expressions do not refer to the same embodiment. Moreover, when describing specific features, structures, or characteristics in conjunction with embodiments, whether or not explicitly described, it is indicated that incorporating such features, structures, or characteristics into other embodiments is within the knowledge scope of those skilled in the art.

[0082] The specification and claims use certain terms to refer to specific modules, components, or parts. Those skilled in the art will understand that users or manufacturers may use different names or terms to refer to the same module, component, or part. This specification and claims do not distinguish modules, components, or parts based on differences in name, but rather on differences in function. The terms "comprising" and "including" used throughout the specification and claims are open-ended and should be interpreted as "including but not limited to." Furthermore, the term "connection" here includes any direct and indirect electrical connection means. Indirect electrical connection means include connections via other means.

[0083] This invention proposes an optimization method for air conditioning systems, aiming to solve problems such as energy waste and unstable comfort levels in existing HVAC systems. By combining actual operating data and simulation data, and comprehensively considering building dynamic load, meteorological conditions, and user behavior patterns, a smart management system covering the entire lifecycle of the system is established. This invention is applicable to the optimized design of air conditioning systems in small and medium-sized buildings, meeting intermittent usage needs and improving energy efficiency and system response performance.

[0084] Figure 1A and Figure 1B A flowchart of an air conditioning system optimization method 1 provided in an embodiment of the present invention is shown below. Figure 1A As shown, the optimization method 1 for the air conditioning system includes the following steps:

[0085] S10: Collect annual operating data and measured operating data of the air conditioning system, generate typical daily meteorological data for the whole year using the annual operating data, and generate typical annual meteorological data based on meteorological data from recent years.

[0086] S20: Establish a simulated building model in TRNSYS, input meteorological condition data, building information, personnel lighting equipment parameters and actual multimodal operating conditions of the air conditioning system, and generate simulated optimized operating data of the air conditioning system. The meteorological condition data includes the typical meteorological year data and the typical meteorological day data of the whole year.

[0087] S30: Perform deviation analysis between the measured operating data and the simulated optimized operating data. When the deviation exceeds the preset threshold, the air conditioning system automatically adjusts the equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

[0088] In step S10:

[0089] like Figure 2 As shown, generating typical meteorological year data based on meteorological data from recent years also includes the following steps:

[0090] S100: Standardize the meteorological data from recent years, extract key features using principal component analysis, and generate the typical meteorological year data through orthogonal transformation.

[0091] like Figure 3 As shown, step S100 further includes:

[0092] S1000: Meteorological data from recent years (e.g., the last 30 years) are standardized using the following formula. This data includes dry-bulb temperature, wet-bulb temperature, solar radiation, and wind speed.

[0093]

[0094] Where X represents meteorological data from recent years, X norm The data represents standardized meteorological data from recent years, where μ is the mean and σ is the standard deviation.

[0095] S1001: Calculate the covariance matrix of standardized meteorological data from recent years and analyze the correlation between various meteorological data.

[0096] S1002: Extract the eigenvalues ​​and corresponding eigenvectors of the covariance matrix, sort them by eigenvalue size, and select the top m principal components with a cumulative contribution rate ≥ 85% among the eigenvalues, where m is a positive integer;

[0097] S1003: By orthogonally transforming meteorological data from nearly a few years, a space consisting of m principal components is projected to generate a reconstructed dataset, which is the typical meteorological year data and is used for subsequent simulation modeling.

[0098] like Figure 2 As shown, generating typical meteorological day data for the whole year using the annual operational data also includes the following steps:

[0099] S101: The K-means algorithm is used to perform time series clustering on the annual operating data to generate typical daily meteorological data for the whole year.

[0100] The K-means algorithm is a commonly used clustering analysis method that divides a dataset into K clusters, each represented by the centroid of its internal data points. For example... Figure 4 As shown, step S101 further includes the following steps:

[0101] S1010: Data preparation, normalizing the hourly operating data of the air conditioning system (such as inlet and outlet water temperature, flow rate, system energy consumption, etc.).

[0102] S1011: Determine the number of clusters. The optimal number of clusters k=7 is determined by the Elbow Method, which corresponds to 7 typical operating conditions (including summer air conditioning, dehumidification, and fresh air, and winter air conditioning, heating, humidification, and fresh air).

[0103] S1012: Cluster analysis, using Euclidean distance as the metric, iteratively optimizes cluster centers.

[0104] S1013: Extract typical operating conditions and select the operating parameters (including dry-bulb temperature, moisture content, and enthalpy) corresponding to each type of center point as representative operating conditions for optimization strategy formulation.

[0105] Step S1013 also includes: using the hourly conversion coefficient in combination with the daily average and daily difference of the main parameters of the design day to generate the hourly value of the design day in reverse. The "daily difference" refers to the difference between the maximum and minimum values ​​of a certain meteorological element in a day. The "daily average" refers to the average value of a certain meteorological element measured four times in a day at 02:00, 08:00, 14:00 and 20:00. It can also be calculated as the average of the highest and lowest values ​​of a certain meteorological element on a certain day.

[0106] In step S20:

[0107] like Figure 5 As shown, step S20 further includes the following steps:

[0108] S200: Create a simulated building model in TRNSYS, and input information such as building envelope characteristics, interior layout, equipment operating parameters, and human activity patterns.

[0109] S210: Input the meteorological data of a typical year and the meteorological parameters corresponding to the seven standard design days to be generated into the TRNSYS model, and set the simulation conditions, including indoor temperature and humidity setpoints, personnel density, and lighting equipment usage modes.

[0110] S220: TRNSYS simulation and real-time adjustment of simulation parameters;

[0111] Step S220 specifically includes:

[0112] (1) Run the simulation model and output hourly building cooling load, heating load and energy consumption data.

[0113] (2) Adjust the simulation parameters and observe the energy consumption changes and operating characteristics of the air conditioning system under different operating conditions.

[0114] Simulation parameter adjustments can be made, for example, by adjusting the parameters on the cold and heat source sides to observe changes in system energy consumption, as detailed below:

[0115] Adjusting the temperature difference between the inlet and outlet water on the cold and heat source sides: gradually increasing it from 5℃ to 10℃, simulations showed that when the temperature difference was 7.5℃, the EER (energy efficiency ratio) of the air conditioning system increased by 12%.

[0116] Adjusting the water flow rate: varying it within the range of 100,000 to 200,000 kg / h, simulations showed that when the water flow rate was 150,000 kg / h, the pipeline pressure drop decreased by 18%, and the pump consumption decreased.

[0117] Therefore, the simulation optimization operation data includes: inlet and outlet water flow rate of 150,000 kg / h on the cold and heat source side, and inlet and outlet water temperature difference of 7.5℃ on the cold and heat source side.

[0118] The above parameter adjustments are based on TRNSYS simulation results, and their rationality is verified by combining the thermodynamic balance equation.

[0119] like Figures 6 to 7 As shown, step S20 further includes:

[0120] S230: Optimizes simulated building models using response surface design methods, including:

[0121] S2300: The experimental design method is used to systematically change the input variables of the simulated building model in order to collect experimental data;

[0122] S2301: Set the equipment operation limits of the air conditioning system. Based on the TRNSYS simulation output data, with the optimization objective of minimizing system energy consumption, the relationship between system energy consumption and input variables is described by a quadratic polynomial model.

[0123] S2302: Solve the simulated building model using optimization algorithms.

[0124] The objective function for system energy consumption is as follows:

[0125] MinimizeF = E;

[0126] Where F represents the objective function and E represents the system energy consumption.

[0127] Step S2301 further includes: setting the water temperature range on the heat source side and the load side; designing the water flow range on the heat source side and the load side according to the pipeline design specifications; and setting the rated value of the operating power of the heat source equipment.

[0128] For example, the water temperature range on the cold and heat source side is set to: 5℃≦T in / out ≤10℃, where T in / out This indicates the water temperature at the inlet or outlet on the heat source / cold source side; the water flow rate range on the heat source / cold source side is set according to the pipeline design specifications as: G min ≤Data usage≤GB max The operating power of the equipment shall not exceed the rated value.

[0129] The quadratic polynomial model of system energy consumption is as follows:

[0130]

[0131] Where E represents system energy consumption in kW; x1 to x4 represent input variables, which are respectively represented as the temperature difference between the cold and hot water sources, the water flow rate between the cold and hot water sources, the temperature difference between the load side and the water flow rate between the load side; β represents the regression coefficient, which is determined by least squares fitting.

[0132] The significance test for the simulated building model requires a p-value < 0.05 and a goodness of fit R0.05. 2 ≥0.95.

[0133] The steps of the least squares fitting process are as follows:

[0134] (1) Construct the design matrix X: including input variables (temperature difference between cold and heat source side water, flow rate between cold and heat source side water, temperature difference between load side water and flow rate between load side water) and interaction terms and square terms.

[0135] (2) Construct the response vector Y: corresponding to the system energy consumption data output by the simulation.

[0136] (3) Solving for the regression coefficients: using the normal equation β=(X T X) -1 X T Y is used to calculate the optimal regression coefficient.

[0137] (4) Model validation: Residual analysis shows that the data points conform to a normal distribution, and the error between the predicted and measured values ​​is <2%.

[0138] In summary, as Figure 8 As shown, the process of optimizing the operation of a simulated building model's air conditioning system using response surface methodology includes:

[0139] Data collection: First, experimental data were collected by systematically changing the input variables of the simulated building model using experimental design methods. Input variables included the temperature difference between the cold and hot water sources, the water flow rate between the cold and hot water sources, the temperature difference between the load side and the water flow rate on the load side.

[0140] Least squares fitting: Based on the collected experimental data, the least squares method is used to fit these experimental data to establish a preliminary mathematical model, starting from a first-order model, that is, assuming that the relationship between the input variables and the output response is linear.

[0141] Standard deviation estimator: After obtaining the preliminary model, the difference between the predicted values ​​and the actual observed values ​​is calculated, i.e., the residuals. By analyzing the standard deviation of these residuals, the accuracy and reliability of the model can be evaluated.

[0142] Optimize the control scheme and its effect: Based on the preliminary model, propose different control strategies and evaluate their effects. For example, adjust the setpoint of the air conditioning system to achieve energy saving and improve comfort. Then, apply the optimized control strategy to the model and observe its effect.

[0143] Model validation:

[0144] First-order model validation: If the prediction results of the first-order model are consistent with the actual situation and the error is within an acceptable range, the model is considered to be credible and can be directly used to guide the optimization of the air conditioning system.

[0145] Second-order model verification: If the first-order model is not accurate enough, consider building a more complex second-order model (considering the interactions between variables and squared terms). Repeat the verification process above.

[0146] n-order model verification: If the lower-order models cannot meet the requirements, continue to increase the complexity of the models until a model that can accurately describe the system behavior without being too complex is found.

[0147] Model output: Once a model with sufficient credibility (regardless of its order) is found, it can be used to guide the optimal control of the air conditioning system, including but not limited to setting the best operating parameters and formulating energy-saving strategies.

[0148] In step S30:

[0149] Specifically, the dynamic adjustment steps are as follows:

[0150] Real-time data acquisition: Obtain the current operating parameters of the air conditioning system (such as water temperature, flow rate, indoor temperature and humidity) through devices such as the Internet of Things (IoT), i.e., the measured operating data.

[0151] Comparison with the optimized strategy: Deviation analysis was performed between the measured operating data and the preset optimized parameters. The preset optimized parameters are the simulated optimized operating data, such as inlet and outlet water flow rates of 150,000 kg / h on the cold and heat source side and inlet and outlet water temperature difference of 7.5℃ on the cold and heat source side.

[0152] Parameter correction: If the deviation exceeds the threshold (e.g., ±5%), the PID controller is triggered to adjust the pump frequency or valve opening to bring the air conditioning system back to the optimal operating condition, so that the actual operating data of the air conditioning system is restored to the simulated optimized operating data.

[0153] like Figure 9 As shown, the optimization method 1 for the air conditioning system also includes the following steps:

[0154] S40: Apply the optimization results to regionalized case studies for different climate zones and building types to verify their applicability;

[0155] S50: Based on the difference in heating load in winter and cooling load in summer, air source heat pump is used for cooling in summer, and ground source heat pump is used for cooling or heating at other times.

[0156] Specifically:

[0157] Regional adaptability analysis: The optimization results are applied to regionalized case studies for different climate regions and building types to verify their applicability.

[0158] Seasonal switching strategy: Based on the difference in heating load in winter and cooling load in summer, switch to air source heat pump cooling from July 1st to September 1st, and use ground source heat pump at other times to flexibly adjust operating parameters to achieve a balance between energy saving and comfort.

[0159] To facilitate understanding of the above embodiments, a specific application scenario of the above embodiments will be used as an example for illustration below:

[0160] 1. Typical daily meteorological data

[0161] (1) Calculation of time-by-time variation coefficient

[0162] To meet the generation requirements of different types of design days, this paper proposes hourly coefficient extraction and design day generation routes with different combinations of meteorological elements. The steps for extracting hourly coefficients are as follows:

[0163] a) Based on 30 years or more of actual meteorological observation data, extract hourly variation coefficients of meteorological elements, and update system data based on the project's own accumulated meteorological data.

[0164] b) Combine the hourly variation trend coefficients of different meteorological elements into an hourly variation trend coefficient matrix and perform cluster analysis;

[0165] c) Generate time-varying coefficients based on the centroids of the time-varying trend coefficients in the clustering results.

[0166] In addition to utilizing publicly available long-term historical data, the system incorporates data recorded by the project itself and updates its database regularly. This helps reflect the latest local weather trends, which is particularly important for rapidly changing urban environments.

[0167] K-means clustering was used to extract the daily hourly variation coefficients of meteorological data from various regions, resulting in hourly variation coefficients that reflect the actual daily changes in meteorological elements (obtained from hourly observations of various regions). The formula for extracting the daily hourly variation trend of meteorological parameters is shown in the table.

[0168] Table 1. Formulas for extracting the hourly variation trends of different design parameters

[0169]

[0170] Since the maximum values ​​of dry-bulb temperature and wet-bulb temperature do not occur simultaneously within a day, the times when their hourly transformation coefficients reach 1 are also different. However, the dry-bulb and wet-bulb temperatures used for calculation during the design day's generation occur simultaneously. Therefore, it is necessary to adjust the hourly transformation coefficients of dry-bulb and wet-bulb temperatures to both reach 1 simultaneously to ensure that the design values ​​of dry-bulb and wet-bulb temperatures are at the same time in the design day parameters. The specific method is: divide the hourly variation trend coefficient of wet-bulb temperature by the hourly variation trend coefficient of wet-bulb temperature corresponding to the time when the hourly transformation coefficient of dry-bulb temperature is 1, and finally use the transformed hourly variation trend coefficient of wet-bulb temperature as the hourly transformation coefficient of wet-bulb temperature. The same processing method is adopted for the hourly transformation coefficients of main and auxiliary parameters in other parameter combinations, and the mathematical description is shown in the following formula:

[0171]

[0172] In the formula: —Auxiliary parameter time-sequential coefficient; β vice,i — Typical trend coefficient of auxiliary parameter; β vice,I —The auxiliary parameter trend coefficient corresponding to the moment when the typical trend coefficient of the main parameter is 1.

[0173] The daily hourly variation coefficients for the design, taking Beijing as an example, are generated using the above method, as shown in the table below.

[0174] Table 2. Hourly Variation Coefficients of Beijing Design Days

[0175]

[0176]

[0177] (2) Generation of design parameters for the new standard

[0178] The design day's hourly value is obtained by inversely solving the formula for extracting the hourly change trend coefficient, and the resulting formula is shown in Table 3.

[0179] Table 3 Formulas for Generating Design Days

[0180]

[0181] In the formula: X i —Hourly values ​​of the main parameters for the design day; X design —Design value of main parameters on the design date; β main,i —Time-sequential coefficients of the main parameters; X p —Daily average of main parameters on the design date; X range —Daily variation of the main parameters on the design date; Y i —Design daily auxiliary parameter hourly values; Y design —Design values ​​for auxiliary parameters; —Time-sequential coefficients of auxiliary parameters; Y p —Daily average of design auxiliary parameters; Y range —Daily variation of auxiliary parameters in design.

[0182] Taking Beijing as an example, the seven new standard design day parameters calculated using the above method are shown in the table.

[0183] Table 4 Parameters for Design Days in Beijing

[0184]

[0185]

[0186] 2. Building Model and Load Simulation

[0187] (1) Establishment of architectural model

[0188] Using the SketchUp plugin in TRNSYS 18, a 3D model of the building was constructed based on the building floor plans and thermal parameters of the building envelope. The building space was divided into multiple independent thermal zones to ensure that the heat load characteristics of each zone could be accurately captured. Through the establishment of the building's heat balance model, hourly load operation data was calculated, including cooling and heating load demands for different time periods throughout the year.

[0189] (2) Construction of the air conditioning system model

[0190] Based on the building load model, and combined with the designed underground pipe parameters and cooling tower parameters, a fan coil unit + fresh air system model was further constructed, with the ground source heat pump as the main heat source and the cooling tower as the auxiliary cooling source. The model fully considers the dynamic coupling relationship between the ground source heat pump, cooling tower, fan coil unit and other equipment to ensure the accuracy of the simulation results.

[0191] 3. System Operation Analysis

[0192] (1) Energy efficiency comparison between air source heat pumps and ground source heat pumps

[0193] Through simulation analysis, the operational data for standalone air-source heat pumps and standalone ground-source heat pumps were obtained throughout the year. The results are as follows: Figures 10 to 12 As shown.

[0194] Depend on Figures 10 to 12 It can be seen that the annual cooling and heating capacity of air source heat pumps are higher than those of ground source heat pumps. Although the annual operating energy consumption of both air source and ground source heat pumps shows an upward trend, the operating energy consumption of air source heat pumps is significantly higher, while the annual energy saving rate of ground source heat pumps is approximately 6%. Comparing the coefficient of performance (COP) of air source and ground source heat pumps, it was found that during the heating period (corresponding hours in the figure: 0-2496, 7656-8760), the COP of ground source heat pumps is consistently higher than that of air source heat pumps. During the cooling period, i.e., from May 1st to July 1st (corresponding hours in the figure: 2880-5088), the COP of ground source heat pumps is higher, while during the period from July 1st to September 1st (corresponding hours in the figure: 5088-7296), the COP of air source heat pumps is actually higher. Based on the annual operation analysis, it is recommended to use air source heat pumps for cooling from July 1st to September 1st, and ground source heat pumps for cooling / heating during the rest of the time, which can achieve a more efficient energy-saving operation strategy.

[0195] 4. Response Surface Optimization Analysis

[0196] (1) Experimental Design and Modeling

[0197] A response surface methodology (RSM) was employed, with the temperature difference and flow rate of the inlet and outlet water on the cold and heat source sides, and the temperature difference and flow rate of the inlet water on the load side, as independent variables (i.e., input variables), and system energy consumption as the response variable, to establish a quadratic polynomial model of system energy consumption. The Box-Behnken experimental design (BBD) was used for the experiments, and data samples were obtained through simulation to fit the relationship model between system energy consumption and the independent variables.

[0198] The factor levels and codes for the Box-Behnken experiment are shown in Table 5 below, and the design values ​​and response values ​​are shown in Table 6 below.

[0199] Table 5. Factor levels and codes in the Box-Behnken experiment.

[0200]

[0201] Table 6. Box-Behnken experimental design and response values

[0202]

[0203]

[0204] (2) Optimization results

[0205] Using MATLAB, a multi-parameter least squares fitting was performed. The response result is a linear model with variances all less than 0.05. The final model equation is as follows:

[0206] Sqrt(E)=1.0437×A+1.9945×B+3.0000×C+4.0044×D+5

[0207] The result generated by MATLAB is as follows: Figures 13A to 13D As shown, Figure 13A The graph shows the normal distribution of the residuals. The residual points are closer to the straight line, which indicates that the residuals conform to the normal distribution characteristics. This means that the error term of the model satisfies the normality assumption. No abnormal residuals that deviate significantly from the normal distribution appeared during the model fitting process, which verifies the rationality of the model error structure. Figure 13B The graph shows the correspondence between the residuals and the predicted values ​​of the equation. The residuals are scattered and irregular, indicating that there is no obvious functional relationship or trend (such as linear or curvilinear trend) between the residuals and the predicted values. This means that the model has fully captured the information in the data and there are no unconsidered systematic errors or omitted variables, which further proves that the model has a good fit. Figure 13C This is a graph showing the correspondence between predicted values ​​and actual experimental values. Points close to the same straight line indicate a higher consistency between the model's predicted values ​​and the actual measured values. This means that the model's prediction accuracy for the response values ​​meets the requirements and can effectively reflect the true relationship between variables, thus verifying the model's reliability and practicality. Figure 13D It is a three-dimensional scatter plot of the first two principal components obtained through principal component analysis. Parameter 1 is the interaction term of A and B (A×B), and parameter 2 is the interaction term of C and D (C×D). The scatter plots show a specific clustering or dispersion trend, which can help analyze the comprehensive influence of the interaction terms (A×B and C×D) on the response value (y), and provide a visual basis for judging the significance of the interaction of factors and the direction of optimization.

[0208] A comparison between actual measured values ​​and equation-predicted values ​​is as follows: Figure 14 Based on the formula (relative error = (measured value - true value) / true value), the relative error is within 2%, which meets the actual requirements and can be used to guide system optimization or control strategy design. In this study, to ensure room terminal temperature control, the heat exchange on the fixed cold / heat source side and the load side is consistent. 7235 sets of data meeting the criteria were selected. Substituting these data into the fitted formula, the trend is as follows: Figure 15 .

[0209] Depend on Figure 15It can be seen that the highest optimized energy consumption of the system is approximately 1,353,181.253 kW, with the inlet and outlet water temperature difference on both sides (i.e., the cold and heat source side and the load side) maintained at around 5.6℃, and the flow rate at around 200,000 kg / h. The lowest optimized energy consumption is approximately 443,099.773 kW, at which point the inlet and outlet water temperature difference and flow rate on the cold and heat source side are the same as those on the load side, with the inlet and outlet water temperature difference on both sides maintained at around 10℃, and the flow rate at around 110,000 kg / h. Compared with the highest energy consumption, the energy saving rate is approximately 67.2%.

[0210] The above specific application scenarios have verified the applicability and superiority of the air conditioning system optimization method 1 of the present invention, which has the following advantages:

[0211] Dynamic optimization: Based on the building's hourly load data and system response characteristics, the operating strategy is dynamically adjusted, resulting in a significant improvement in energy efficiency;

[0212] Significant energy-saving effect: Through the switching strategy of cold and heat sources and parameter optimization, the annual energy saving rate reached 67.2%, which fully demonstrates the effective reduction of system operating costs by optimized design;

[0213] Wide applicability: It has good adaptability to different building types, climate conditions and load requirements, and can be widely applied to more engineering cases.

[0214] The following are apparatus embodiments corresponding to the above method embodiments. This embodiment can be implemented in conjunction with the above embodiments. The relevant technical details mentioned in the above embodiments remain valid in this embodiment, and will not be repeated here to reduce repetition. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the above embodiments.

[0215] In another embodiment, such as Figure 16 The present invention also provides an air conditioning system optimization device 2, which employs the above-described air conditioning system optimization method 1, wherein the air conditioning system optimization device 2 includes at least:

[0216] The data acquisition and processing module 20 is used to collect the annual operation data and measured operation data of the air conditioning system, generate typical meteorological daily data for the whole year using the annual operation data, and generate typical meteorological annual data based on meteorological data from recent years.

[0217] Model building module 21 is used to build a simulated building model in TRNSYS. It inputs meteorological condition data, building information, personnel lighting equipment parameters and actual operating conditions of the air conditioning system, and generates simulated optimized operating data of the air conditioning system. The meteorological condition data includes typical meteorological year data and typical meteorological day data throughout the year.

[0218] The adjustment module 22 is used to perform deviation analysis between the measured operating data and the simulated optimized operating data. When the deviation exceeds the preset threshold, the air conditioning system automatically adjusts the equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

[0219] In this embodiment, as Figure 17 As shown, the data acquisition and processing module 20 also includes:

[0220] The first processing unit 200 is used to standardize meteorological data from recent years, extract key features using principal component analysis, and generate typical meteorological year data through orthogonal transformation.

[0221] The second processing unit 201 is used to perform time series clustering on the annual operating data using the K-means algorithm to generate typical daily meteorological data for the whole year.

[0222] In this embodiment, as Figure 18 As shown, the model building module 21 includes:

[0223] The building unit 210 is used to create a simulated building model in TRNSYS and input information such as building envelope characteristics, interior layout, equipment operating parameters and human activity patterns.

[0224] The simulation data setting unit 211 is used to input typical meteorological year data and meteorological parameters corresponding to the seven standard design days to be generated into the TRNSYS model, and set the simulation conditions, including indoor temperature and humidity setpoints, personnel density, and lighting equipment usage modes.

[0225] Simulation unit 212 is used for TRNSYS simulation and real-time adjustment of simulation parameters;

[0226] The optimization unit 213 is used to optimize the simulated building model through response surface design method. It includes: systematically changing the input variables of the simulated building model to collect experimental data by using experimental design method; setting the equipment operation limits of the air conditioning system, and using the TRNSYS simulation output data to minimize system energy consumption as the optimization objective, describing the relationship between system energy consumption and input variables through a quadratic polynomial model; and solving the simulated building model using optimization algorithm.

[0227] like Figure 19 As shown, the air conditioning system optimization device 2 also includes:

[0228] The regional analysis module 23 is used to apply the optimization results to regionalized case studies for different climate regions and building types to verify their applicability;

[0229] Seasonal switching module 24 is used to adjust the operating load based on the difference between winter heating and summer cooling, using an air source heat pump for cooling in summer and a ground source heat pump for cooling or heating during the rest of the year.

[0230] Specifically, the seasonal switching module 24 is used to switch to air source heat pump cooling from July 1st to September 1st based on the difference in heating load in winter and cooling load in summer, and to activate ground source heat pump at other times to flexibly adjust operating parameters and achieve a balance between energy saving and comfort.

[0231] Furthermore, those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the air conditioning system optimization device 2 described herein can be referred to the corresponding process in the aforementioned embodiment of the air conditioning system optimization method 1, and will not be repeated here.

[0232] Figure 20 This is a schematic diagram of the structure of an electronic device 3 provided in an embodiment of the present invention. Specifically, the electronic device 3 may include at least one processor 30 and at least one memory 31. The memory 31 stores a computer program 310, which is loaded and executed by the processor 30 to implement the relevant steps in the optimization method of the air conditioning system disclosed in any of the foregoing embodiments, such as... Figures 1A to 10 The steps are shown.

[0233] In addition, the memory 31, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, and the storage method can be temporary storage or permanent storage.

[0234] In addition to including a computer program capable of performing the air conditioning system optimization method 1 executed by the electronic device 3 as disclosed in any of the foregoing embodiments, the computer program 310 may further include a computer program capable of performing other specific tasks.

[0235] Another embodiment of the present invention discloses a computer storage medium storing computer-executable instructions. When the computer-executable instructions are loaded and executed by a processor, they implement step 1 of the air conditioning system optimization method disclosed in any of the foregoing embodiments, such as... Figures 1A to 9 The steps are shown.

[0236] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms fall within the scope of protection of the present invention.

Claims

1. A method for optimizing an air conditioning system, characterized in that, Includes the following steps: Collect the annual operating data and measured operating data of the air conditioning system, use the annual operating data to generate typical meteorological day data for the whole year, and generate typical meteorological year data based on meteorological data from recent years; A simulated building model is established in TRNSYS. Meteorological condition data, building information, personnel, lighting, equipment parameters, and the actual multimodal operating conditions of the air conditioning system are input. Simulated optimized operating data of the air conditioning system is generated. The meteorological condition data includes the typical meteorological year data and the typical meteorological day data for the whole year. The measured operating data and the simulated optimized operating data are analyzed for deviation. When the deviation exceeds a preset threshold, the air conditioning system automatically adjusts the system and equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

2. The method for optimizing an air conditioning system according to claim 1, characterized in that, The process of establishing a simulated building model in TRNSYS, inputting meteorological condition data, building information, occupant lighting, equipment parameters, and the actual multimodal operating conditions of the air conditioning system, and generating simulated optimized operating data for the air conditioning system, further includes: optimizing the simulated building model using response surface methodology, including: Experimental design methods were used to systematically change the input variables of the simulated building model in order to collect experimental data; The equipment operation limits of the air conditioning system are set. Based on the TRNSYS simulation output data, with minimizing system energy consumption as the optimization objective, the relationship between system energy consumption and input variables is described by a quadratic polynomial model. The simulated building model is solved using an optimization algorithm.

3. The method for optimizing an air conditioning system according to claim 2, characterized in that, The setting of equipment operating limits for the air conditioning system further includes: Set the water temperature range for the heat source and load sides; The water flow range on the cold and heat source side and the load side shall be designed in accordance with the pipeline design specifications; Set the rated operating power of the cold and heat source equipment.

4. The method for optimizing an air conditioning system according to claim 2, characterized in that, The quadratic polynomial model is as follows: Where E represents the system energy consumption in kW; i, j = 1, 2, 3, 4; x1 to x4 represent the input variables, which are respectively represented as the temperature difference between the cold and hot source sides, the water flow rate between the cold and hot source sides, the water temperature difference between the load side, and the water flow rate between the load side; β represents the regression coefficient, which is determined by least squares fitting.

5. The method for optimizing an air conditioning system according to claim 1, characterized in that, The process of generating typical meteorological year data based on meteorological data from recent years also includes the following steps: The meteorological data from recent years were standardized, and key features were extracted using principal component analysis. Orthogonal transformation was then used to generate typical meteorological year data.

6. The method for optimizing an air conditioning system according to claim 5, characterized in that, The standardization process of the meteorological data from recent years, the extraction of key features using principal component analysis, and the generation of typical meteorological year data through orthogonal transformation further include: The meteorological data from recent years, including dry-bulb temperature, wet-bulb temperature, solar radiation, and wind speed, are standardized according to the following formula. Where X represents the meteorological data from the past few years, X norm The data represents standardized meteorological data from recent years, where μ is the mean and σ is the standard deviation. Calculate the covariance matrix of the standardized meteorological data from recent years; Extract the eigenvalues ​​of the covariance matrix and the corresponding eigenvectors of the eigenvalues, and select the top m principal components with a cumulative contribution rate ≥ 85% from the eigenvalues, where m is a positive integer; The meteorological data from recent years are projected onto a space composed of m principal components through orthogonal transformation to generate a reconstructed dataset, wherein the meteorological data of a typical year is the reconstructed dataset.

7. The method for optimizing an air conditioning system according to claim 1, characterized in that, The process of generating typical meteorological data for the whole year using the annual operational data also includes the following steps: The annual operational data is clustered using the K-means algorithm to generate typical daily meteorological data for the entire year.

8. The method for optimizing an air conditioning system according to claim 1, characterized in that, The step of performing time-series clustering on the annual operational data using the K-means algorithm to generate the annual typical meteorological daily data further includes the following steps: Extract hourly variation trend coefficients of meteorological elements based on actual meteorological observation data of the past 30 years or more, and update meteorological data in real time; The hourly variation trend coefficients of different meteorological elements are combined into an hourly variation trend coefficient matrix. Cluster analysis is performed on the hourly variation trend coefficient matrix, and the centroids of the clusters are extracted as hourly variation coefficients. By combining the hourly coefficients with the daily average and daily difference of the main parameters of the design day, the hourly values ​​of the design day are generated in reverse. The hourly values ​​of the design day are used to generate meteorological parameter variation curves corresponding to various standard design day types.

9. The method for optimizing an air conditioning system according to claim 8, characterized in that, The various standard design days include summer air conditioning, dehumidification, and fresh air, and winter air conditioning, heating, humidification, and fresh air; the main parameters of the design days include dry bulb temperature, moisture content, and enthalpy.

10. The method for optimizing an air conditioning system according to claim 1, characterized in that, It also includes the following steps: For different climate zones and building types, the optimization results will be applied to regional case studies to verify their applicability; Based on the difference in heating load in winter and cooling load in summer, air source heat pumps are used for cooling in summer, while ground source heat pumps are used for cooling or heating at other times.

11. An optimization device for an air conditioning system, characterized in that, The air conditioning system optimization method according to any one of claims 1 to 10, wherein the air conditioning system optimization device comprises at least: The data acquisition and processing module is used to collect the annual operation data and measured operation data of the air conditioning system, generate typical daily meteorological data for the whole year using the annual operation data, and generate typical annual meteorological data based on meteorological data from recent years. The model building module is used to build a simulated building model in TRNSYS. It inputs meteorological condition data, building information, personnel lighting equipment parameters and the actual operating conditions of the air conditioning system, and generates simulated optimized operating data of the air conditioning system. The meteorological condition data includes the typical meteorological year data and the typical meteorological day data for the whole year. The adjustment module is used to perform deviation analysis between the measured operating data and the simulated optimized operating data. When the deviation exceeds a preset threshold, the air conditioning system automatically adjusts the equipment operating parameters to restore the actual operating data of the air conditioning system to the simulated optimized operating data.

12. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the optimization method of the air conditioning system as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, Used to store a computer program; when the computer program is executed by a processor, it implements the optimization method of the air conditioning system as described in any one of claims 1 to 10.

Images