Building parameter automatic calibration method and system based on multi-objective optimization

Abstract

The invention discloses an automatic building parameter calibration method and system based on multi-objective optimization. The method comprises the following steps of determining uncertain building parameters needing to be calibrated and a value range of the uncertain building parameters; building a building simulation model, and embedding the building uncertain parameters into the building simulation model; collecting indoor temperature data and humidity data, and synchronously collecting meteorological data; constructing a multi-objective optimization model; building temperature and humidity under different uncertain parameter values are simulated based on a non-dominated sorting genetic algorithm, the multi-objective optimization model is solved, an optimized non-dominated solution set is output, and a Pareto leading edge is formed; and screening an optimal solution set through Pareto frontier analysis, establishing a calibrated building simulation model, simulating building energy consumption under future weather conditions, and outputting an optimal transformation scheme. According to the method, the calibration requirements of a large-scale building group can be efficiently processed, and the precision and efficiency of building uncertain parameter calibration are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of building parameter calibration, and particularly to an automatic building parameter calibration method and system based on multi-objective optimization. Background Art

[0002] In building environment simulation and energy-saving renovation, accurate parameter calibration is the key to ensuring the reliability of simulation results. However, the existing technologies have the following limitations:

[0003] (1) Isolated parameter calibration: Traditional methods usually calibrate a single parameter (such as the heat transfer coefficient of the building envelope) or a single room, ignoring the multi-parameter coupling effect and the overall spatial difference of the building;

[0004] (2) Dependence on manual experience: The existing calibration process highly depends on the experience of engineers, adjusting parameters through a trial-and-error method, with low efficiency and strong subjectivity;

[0005] (3) Single objective: Existing optimization methods mostly focus on a single objective (such as minimizing the temperature error), and cannot balance multi-dimensional requirements;

[0006] It can be seen that the above defects result in the temperature and humidity simulation errors of existing buildings generally exceeding ±15%, seriously restricting the scientificity and economy of renovation plans. Therefore, there is an urgent need for an automated technology that can achieve multi-parameter collaborative calibration, multi-objective dynamic optimization, and adapt to complex scenarios of multiple rooms. Summary of the Invention

[0007] In order to overcome the defects and deficiencies of the existing technology, the present invention provides an automatic building parameter calibration method based on multi-objective optimization. Through full-parameter collaborative optimization, physical constraint embedding, and dynamic decision support, the present invention significantly improves the temperature and humidity simulation accuracy and provides an efficient and low-cost implementation plan for building renovation.

[0008] To achieve the above object, the present invention adopts the following technical solutions:

[0009] The present invention provides an automatic building parameter calibration method based on multi-objective optimization, including the following steps:

[0010] According to the building type and construction year, referring to the corresponding energy-saving design standards, determine the building uncertain parameters to be calibrated and their value ranges;

[0011] Establish a building simulation model, and embed the building uncertain parameters into the building simulation model by means of text editing;

[0012] Collect indoor temperature data and humidity data, and synchronously collect meteorological data;

[0013] Build a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results in each independent functional area and the measured temperature and humidity data;

[0014] Based on the non-dominated sorting genetic algorithm, simulate the building temperature and humidity under different values of uncertain parameters, solve the multi-objective optimization model, and output the optimized non-dominated solution set to form the Pareto front;

[0015] Through Pareto front analysis, screen the optimal solution set, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan.

[0016] As a preferred technical solution, the building uncertain parameters include: building envelope thermal performance parameters, internal heat source parameters, energy equipment parameters, and ventilation setting parameters.

[0017] As a preferred technical solution, the establishment of the building simulation model specifically includes:

[0018] Build a building geometric model based on building drawings or three-dimensional geometric information and export it as an IDF format file;

[0019] Set the determined fixed parameters through IDF Editor;

[0020] Perform parametric definition on the uncertain parameters, use parametric placeholders to identify the uncertain parameters, and dynamically replace the placeholders through an external script.

[0021] As a preferred technical solution, build a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results in each independent functional area and the measured temperature and humidity data, specifically including:

[0022] Calculate the root mean square error variation coefficient between the hourly air temperature measured in each room and the hourly relative temperature and the simulated values respectively;

[0023] Calculate the root mean square error variation coefficient between the hourly air humidity measured in each room and the hourly relative humidity and the simulated values respectively;

[0024] The objective function of building the multi-objective optimization model is to minimize the root mean square error variation coefficient of the air temperature in each room and minimize the root mean square error variation coefficient of the relative humidity in each room.

[0025] As a preferred technical solution, based on the non-dominated sorting genetic algorithm, simulate the building temperature and humidity under different values of uncertain parameters, and solve the multi-objective optimization model, specifically including:

[0026] Generate an initial population based on Latin hypercube sampling;

[0027] Input each individual in the population into the building simulation model, obtain the simulated results of temperature and humidity in each independent functional area, and calculate the objective value of each individual according to the objective function of the multi-objective optimization model;

[0028] Stratify and sort the population, determine the non-dominated rank of individuals, and calculate the crowding distance of individuals within the same non-dominated layer;

[0029] Combine the parent and offspring populations, retain individuals with a higher non-dominated level and a larger crowding degree, and generate a new generation of population through crossover and mutation;

[0030] When it is determined that the iteration termination condition is reached, output the optimized non-dominated solution set to form the Pareto front.

[0031] As a preferred technical solution, screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan, specifically including:

[0032] Screen the solutions that meet the set temperature objective function value and set humidity objective function value from the Pareto optimal solution set, and generate a calibrated building simulation model based on the screening results;

[0033] Generate a future weather file based on the future climate scenario;

[0034] Generate a renovation plan for the thermal performance of the building envelope under future weather conditions based on the calibrated building simulation model;

[0035] Based on the calibrated building simulation model, simulate the building energy consumption under future weather conditions, calculate the energy saving rate by comparing the energy consumption changes under different renovation measures, and output the optimal renovation plan.

[0036] The present invention also provides a building parameter automatic calibration system based on multi-objective optimization, including: a building parameter acquisition module, a building simulation model construction module, an environmental data acquisition module, a multi-objective optimization model construction module, a multi-objective optimization model solving module, and an optimal renovation plan output module;

[0037] The building parameter acquisition module is used to determine the uncertain building parameters to be calibrated and their value ranges according to the building type and construction year with reference to the corresponding energy-saving design standards;

[0038] The building simulation model construction module is used to establish a building simulation model and embed the uncertain building parameters into the building simulation model in a text editing manner;

[0039] The environmental data acquisition module is used to collect indoor temperature data and humidity data, and synchronously collect meteorological data;

[0040] The multi-objective optimization model construction module is used to construct a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data;

[0041] The multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values based on the non-dominated sorting genetic algorithm, solve the multi-objective optimization model, and output the optimized non-dominated solution set to form the Pareto front;

[0042] The optimal renovation plan output module is used to screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan.

[0043] As a preferred technical solution, the multi-objective optimization model construction module is used to construct a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data, specifically including:

[0044] Calculate the root mean square error variation coefficient between the hourly air temperature measured in each room and the hourly relative temperature and the simulated value respectively;

[0045] Calculate the root mean square error variation coefficient between the hourly air humidity measured in each room and the hourly relative humidity and the simulated value respectively;

[0046] The objective function for constructing the multi-objective optimization model is to minimize the root mean square error variation coefficient of the air temperature in each room and to minimize the root mean square error variation coefficient of the relative humidity in each room.

[0047] As a preferred technical solution, the multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values based on the non-dominated sorting genetic algorithm, solve the multi-objective optimization model, specifically including:

[0048] Generate an initial population based on Latin hypercube sampling;

[0049] Input each individual in the population into the building simulation model, obtain the temperature and humidity simulation results of each independent functional area, and calculate the objective value of each individual according to the objective function of the multi-objective optimization model;

[0050] Perform hierarchical sorting on the population, determine the non-dominated rank of the individuals, and calculate the crowding distance of the individuals within the same non-dominated layer;

[0051] Combine the parent and offspring populations, retain the individuals with a higher non-dominated level and a larger crowding degree, and generate a new generation of population through crossover and mutation;

[0052] When it is determined that the iteration termination condition is reached, the optimized non-dominated solution set is output to form the Pareto front.

[0053] As an optimal technical solution, the optimal renovation plan output module is used to screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan, specifically including:

[0054] Screen the solutions that meet the set temperature objective function value and the set humidity objective function value from the Pareto optimal solution set, and generate a calibrated building simulation model based on the screening results;

[0055] Generate a future weather file based on the future climate scenario;

[0056] Generate a renovation plan for the thermal performance of the building envelope under future weather conditions based on the calibrated building simulation model;

[0057] Taking the calibrated building simulation model as a benchmark, simulate the building energy consumption under future weather conditions, calculate the energy saving rate by comparing the energy consumption changes under different renovation measures, and output the optimal renovation plan.

[0058] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0059] (1) By combining the multi-objective optimization algorithm with building energy consumption simulation, the present invention significantly improves the accuracy and efficiency of building uncertain parameter calibration. Compared with the traditional calibration methods that rely on manual experience or single-objective optimization, the present invention can optimize multiple objective functions (such as temperature and humidity simulation errors) simultaneously to ensure that the calibration results are more scientific and comprehensive.

[0060] (2) By integrating the Latin Hypercube Sampling (LHS) and the NSGA-II multi-objective optimization algorithm, the present invention realizes an automated and continuous parameter calibration process. The system can automatically generate the initial population, call EnergyPlus for simulation, calculate the objective function values, and output the Pareto optimal solution set through iterative optimization. This process not only reduces the time and effort of manual intervention but also supports parallel computing, enabling efficient processing of the calibration requirements for large-scale building complexes.

[0061] (3) By introducing a three-dimensional visualization platform, the present invention can dynamically render the Pareto front and the calibrated building simulation model, providing intuitive decision-making support for users, supporting the simulation and comparison of building energy consumption under future climate conditions, and helping users quickly screen the optimal renovation plan through interactive analysis tools. This intelligent visualization function greatly reduces the technical threshold, making it easy for non-professionals to understand and use. Description of the Drawings

[0062] Figure 1Schematic diagram of the overall process of the building parameter automatic calibration method based on multi-objective optimization of the present invention;

[0063] Figure 2 Schematic diagram of the value range of the building uncertain parameters of the present invention;

[0064] Figure 3 Schematic diagram of the Pareto front of the present invention;

[0065] Figure 4 Comparison chart of building energy consumption before and after renovation under future weather conditions of the present invention. Detailed implementation manners

[0066] In order 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 the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0067] Embodiment 1

[0068] As Figure 1 shown, this embodiment provides a building parameter automatic calibration method based on multi-objective optimization, including the following steps:

[0069] S1: According to the building type and construction age, referring to the corresponding energy-saving design standards, determine the building uncertain parameters to be calibrated and their value ranges;

[0070] In this embodiment, the building uncertain parameters can be classified into the following categories:

[0071] (1) Building envelope thermal performance parameters: wall heat transfer coefficient, wall thermal mass, external wall solar radiation absorption coefficient, roof heat transfer coefficient, roof thermal mass, roof solar radiation absorption coefficient, floor heat transfer coefficient, floor thermal mass, external window heat transfer coefficient, external window solar heat gain coefficient, etc.;

[0072] (2) Internal heat source parameters: personnel density, lighting power density, electrical power density and personnel activity schedule, etc.;

[0073] (3) Energy equipment parameters: air conditioner energy efficiency ratio, air conditioner usage time, air conditioner set temperature, etc.;

[0074] (4) Ventilation setting parameters: indoor ventilation mode, air infiltration rate, fresh air volume, window opening area and time, etc.;

[0075] The above uncertain parameter ranges can be determined according to the energy-saving design standards corresponding to the building construction age, and at the same time, other effective information provided by the owner and the building constructor also needs to be comprehensively considered for corresponding adjustment, such as Figure 2As shown, the value ranges of various uncertain parameters can be obtained by referring to the "Energy Efficiency Design Standard for Residential Buildings in Hot Summer and Warm Winter Zones" (JGJ75-2003).

[0076] S2: Establish a building simulation model in the EnergyPlus software according to the building drawings, and at the same time embed the above-mentioned uncertain parameters into the simulation model by means of text editing.

[0077] In this embodiment, the specific steps for establishing a building simulation model including uncertain parameters in step S2 are as follows:

[0078] (1) Based on the building drawings or three-dimensional geometric information, construct a building geometric model in the SketchUp software through the OpenStudio plug-in, and export it as an IDF format file recognizable by EnergyPlus.

[0079] (2) According to other valid information provided by the owner and the house builder, set the determined fixed parameters in EnergyPlus through the IDFEditor, including: thermal performance parameters of the building envelope (such as fixing the heat transfer coefficient and solar heat gain coefficient if the external window type is known); internal heat source parameters (such as fixing the lighting and electrical equipment power density); energy equipment parameters (such as fixing the air-conditioning usage schedule); ventilation setting parameters (such as fixing the fresh air volume).

[0080] (3) Define the uncertain parameters parametrically: In the IDF file, use parametric placeholder identifiers for the uncertain parameters in the Text Editor of the EnergyPlus software, and then dynamically replace the placeholders through external scripts (such as R language, Python) to achieve batch simulation of multi-parameter combinations.

[0081] S3: Hourly monitor the actual temperature and humidity data through the temperature and humidity sensors deployed indoors, and simultaneously collect the meteorological data of this area during the monitoring period, and upload the above data to the calibration system.

[0082] In this embodiment, the meteorological data collection and indoor sensor arrangement specifically include the following steps:

[0083] (1) Indoor temperature and humidity sensor arrangement: According to the building function zoning and thermal environment characteristics, determine the sensor arrangement positions, and at least 1 temperature and humidity sensor should be arranged in each independent functional area (such as office, meeting room, corridor) in the building. The sensor arrangement height is 1.2m - 1.5m to simulate the temperature and humidity conditions in the personnel activity area.

[0084] (2) Meteorological data collection: Obtain the meteorological data during the monitoring period through the local meteorological station or building meteorological sensors, including dry bulb temperature, relative humidity, solar radiation intensity and wind speed.

[0085] (3) Data processing and synchronization: Organize the meteorological data into a meteorological file in epw format, and synchronously upload it to the calibration system together with the indoor monitoring data for driving EnergyPlus simulation and error calculation;

[0086] S4: Construct a multi-objective optimization model with the optimization direction of minimizing the errors between the simulated temperature and humidity results of EnergyPlus in each room and the measured temperature and humidity data;

[0087] In this embodiment, constructing the multi-objective optimization model specifically includes the following steps:

[0088] (1) Calculate the root mean square error (RMSE) between the hourly air temperature (Ta) or hourly relative humidity (RH) measured in each room and the simulated values respectively, as shown in the following formula:

[0089]

[0090] Among them, represents the simulated value, y i represents the measured value, i is the i-th hour, and n is the total number of hours:

[0091] (2) Calculate the coefficient of variation of the root mean square error (CV(RMSE)) between the hourly air temperature or hourly relative humidity measured in each room and the simulated values respectively, as shown in the following formula:

[0092]

[0093] Among them, represents the measured average value within the calibration time period:

[0094] (3) Finally, the objective functions 1 and 2 are to minimize the coefficient of variation of the root mean square error of the air temperature and relative humidity in each room. Among them, Ta is the air temperature, RH is the relative humidity, z is the z-th room, and m is the total number of rooms:

[0095]

[0096] S5: Use the non-dominated sorting genetic algorithm (NSGA-II) to call the EnergyPlus software to simulate the building temperature and humidity under different uncertain parameter values, so as to solve the multi-objective optimization model;

[0097] In this embodiment, the specific steps of the genetic algorithm solution include:

[0098] (1) Parameter initialization and population generation: After clarifying the value ranges of the uncertain parameters of each building, use Latin hypercube sampling (LHS) to generate the initial population to ensure the diversity and uniformity of the initial solutions;

[0099] In this embodiment, the population size of each generation is set to four times the uncertain parameter, and the maximum number of iterations is set to 50;

[0100] (2) Objective function evaluation: Input each individual (parameter combination) in the population into the building simulation model, call EnergyPlus for simulation, obtain the simulation results of the temperature and relative humidity of each room, and calculate the objective value of each individual according to the objective function 1 (minimizing the coefficient of variation of the root mean square error of the air temperature in each room) and objective function 2 (minimizing the coefficient of variation of the root mean square error of the relative humidity in each room) defined by the model output;

[0101] (3) Non-dominated sorting and crowding degree calculation (core of NSGA-II): Stratify and sort the population to determine the non-dominated rank (Pareto front level) of the individuals. Within the same non-dominated layer, calculate the crowding distance of the individuals to ensure the diversity of the solutions;

[0102] (4) Selection, crossover and mutation: Combine the parent and offspring populations, retain the individuals with a high non-dominated level and a large crowding degree, and generate a new generation of population through crossover (simulating gene recombination) and mutation (introducing random perturbations);

[0103] (5) Termination condition judgment: If the preset maximum number of generations (Max.generations) is reached, terminate the optimization; if the termination condition is not reached, return to step (2) to continue iterative optimization;

[0104] (6) Output the final result: As Figure 3 shown, output the optimized non-dominated solution set (Final population), which constitutes the Pareto front;

[0105] S6: Screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, and propose an enclosure structure renovation plan to adapt to future climate change.

[0106] In this embodiment, the specific steps for generating the enclosure structure renovation priority plan include:

[0107] (1) Screen the solutions that meet the following conditions from the Pareto optimal solution set: the temperature objective function value ≤ 2%, the humidity objective function value ≤ 7%, and generate a calibrated building simulation model based on the screening results;

[0108] In this embodiment, the calibration results are: the heat transfer coefficient of the wall is 1.48 W / m 2 K, the thermal mass of the wall is 285 KJ / m 2 K, the heat transfer coefficient of the roof is 0.86 W / m 2 K, the thermal mass of the wall is 204 KJ / m 2K, the heat transfer coefficient of the external window is 3.8 W / m 2 K, and the solar heat gain coefficient of the external window is 0.75;

[0109] (2) Generate future weather files based on future climate scenarios (such as IPCC SSP5-8.5), with a time range from 2030 to 2080, covering typical heatwave events;

[0110] (3) Based on the calibrated building simulation model, generate renovation plans for the thermal performance of the building envelope under future weather conditions, specifically including:

[0111] Renovation of the roof and external walls: Use heat-insulating and reflective coatings (solar reflectance ≥ 0.7) or add ventilation and heat-insulating layers to reduce the heat transfer coefficient to 80%-90% of the calibrated value;

[0112] Renovation of the external windows: Replace low-emissivity glass with a solar heat gain coefficient (SHGC) ≤ 70% of the calibrated value; Set adjustable external sunshade devices (such as venetian blinds, sunshade grilles) to reduce solar radiation heat gain by 30%-50%;

[0113] (4) Analysis of improvement effects and optimization of the plan: Based on the calibrated building simulation model, simulate the building energy consumption under future weather conditions; Compare the energy consumption changes under different renovation measures, calculate the energy-saving rate (such as the air-conditioning energy consumption reduction ≥ 25%), and output the optimal renovation plan in combination with the cost;

[0114] Such as Figure 4 shown, obtain the energy consumption changes under the renovation plans for the roof, external walls, and external windows. Compared with renovating the roof and external walls, it is recommended to prioritize the renovation of the external windows, select low-transmittance low-e glass with an SHGC lower than 0.45, and set adjustable external sunshade devices (such as venetian blinds, sunshade grilles). In 2030, the air-conditioning energy consumption of the building before renovation is 22.5 kwh / m 2 , and it drops to 20.9 kwh / m 2 after renovation, with an energy-saving rate as high as 7.1%; Although the building air-conditioning energy consumption increases year by year with climate change, the energy-saving rate after renovation also increases year by year. In 2080, the air-conditioning energy consumption of the building before renovation is 25 kwh / m 2 , and the air-conditioning energy consumption of the renovated building is 22.3 kwh / m 2 , with an energy-saving rate of 10.8%.

[0115] In this embodiment, the full process automation can be achieved through the data acquisition stage, the parameter calibration stage, and the decision-making output stage. In the data acquisition stage, the monitoring instrument module uploads the indoor temperature and humidity data to the cloud server in real time through the wireless transmission protocol; in the parameter calibration stage, the cloud server calls the multi-objective optimization algorithm engine (NSGA-II) to run the EnergyPlus simulation in parallel; accelerates the parameter optimization process through a distributed computing framework (such as Apache Spark); updates the parameter constraint database in real time, stores the "Energy Efficiency Design Standard for Public Buildings" (GB 50189-2015), the "Code for Thermal Design of Civil Buildings" (GB 50176-2016), and the "General Code for Building Energy Efficiency and Renewable Energy Utilization" (GB 55015-2021) to ensure that the calibration results comply with the specifications, provides a parameterized placeholder replacement function, and supports batch simulation and optimization; in the decision-making output stage, the calibration results and the renovation plan are dynamically displayed through a 3D visualization platform. Specifically, the 3D visualization platform dynamically renders the three-objective Pareto front and the renovation plan effect, provides a comparison chart of the building energy consumption changes of different renovation plans under future climate conditions, and supports interactive analysis.

[0116] The automatic calibration method for building parameters based on multi-objective optimization of the present invention is applicable to the collaborative calibration of the thermal parameters of the building envelope structure and the internal heat source parameters of existing buildings (especially old buildings), aiming to solve the problem of indoor temperature and humidity simulation deviation caused by parameter uncertainty, and provide data-driven decision support for building energy efficiency renovation.

[0117] Embodiment 2

[0118] This embodiment provides an automatic calibration system for building parameters based on multi-objective optimization, which is used to implement the automatic calibration method for building parameters based on multi-objective optimization in the above-mentioned Embodiment 1. The system includes: a building parameter acquisition module, a building simulation model construction module, an environmental data acquisition module, a multi-objective optimization model construction module, a multi-objective optimization model solving module, and an optimal renovation plan output module;

[0119] In this embodiment, the building parameter acquisition module is used to determine the uncertain building parameters to be calibrated and their value ranges according to the building type and construction year, with reference to the corresponding energy efficiency design standards;

[0120] In this embodiment, the building simulation model construction module is used to establish a building simulation model and embed the uncertain building parameters into the building simulation model in a text editing manner;

[0121] In this embodiment, the environmental data acquisition module is used to collect indoor temperature data and humidity data, and synchronously collect meteorological data;

[0122] Specifically, based on the temperature and humidity sensors to collect indoor temperature data and humidity data, high-precision temperature sensors (accuracy: ±0.5°C) and humidity sensors (accuracy: ±2%RH) are deployed in typical functional areas of the building, supporting wireless transmission protocols (such as LoRaWAN, ZigBee), and the data upload frequency ≤ 10 minutes;

[0123] In this embodiment, the multi-objective optimization model construction module is used to construct a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data;

[0124] In this embodiment, the multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values based on the non-dominated sorting genetic algorithm, solve the multi-objective optimization model, and output the optimized non-dominated solution set to form the Pareto front;

[0125] In this embodiment, the optimal renovation plan output module is used to screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan.

[0126] In this embodiment, the multi-objective optimization model construction module is used to construct a multi-objective optimization model, with the optimization direction of minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data, specifically including:

[0127] Calculate the hourly air temperature measured in each room, as well as the root mean square error variation coefficient between the hourly relative temperature and the simulated value;

[0128] Calculate the hourly air humidity measured in each room, as well as the root mean square error variation coefficient between the hourly relative humidity and the simulated value;

[0129] The objective function for constructing the multi-objective optimization model is to minimize the root mean square error variation coefficient of the air temperature in each room and minimize the root mean square error variation coefficient of the relative humidity in each room.

[0130] In this embodiment, the multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values based on the non-dominated sorting genetic algorithm, solve the multi-objective optimization model, specifically including:

[0131] Generate an initial population based on Latin hypercube sampling;

[0132] Input each individual in the population into the building simulation model, obtain the temperature and humidity simulation results of each independent functional area, and calculate the objective value of each individual according to the objective function of the multi-objective optimization model;

[0133] Stratify and sort the population to determine the non-dominated rank of individuals, and calculate the crowding distance of individuals within the same non-dominated layer;

[0134] Combine the parent and offspring populations, retain the individuals with higher non-dominated levels and larger crowding degrees, and generate a new generation of population through crossover and mutation;

[0135] When it is determined that the iteration termination condition is reached, output the optimized non-dominated solution set to form the Pareto front.

[0136] In this embodiment, the optimal renovation plan output module is used to screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation plan, specifically including:

[0137] Screen the solutions that meet the set temperature objective function value and set humidity objective function value from the Pareto optimal solution set, and generate a calibrated building simulation model based on the screening results;

[0138] Generate a future weather file based on the future climate scenario;

[0139] Based on the calibrated building simulation model, generate a renovation plan for the thermal performance of the building envelope under future weather conditions;

[0140] Taking the calibrated building simulation model as a benchmark, simulate the building energy consumption under future weather conditions, calculate the energy saving rate by comparing the energy consumption changes under different renovation measures, and output the optimal renovation plan.

[0141] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A method for automatic calibration of building parameters based on multi-objective optimization, characterized in that: The steps include: According to the building type and construction year, refer to the corresponding energy-saving design standards to determine the building uncertainty parameters that need to be calibrated and their value ranges; Establish a building simulation model and embed the building uncertain parameters into the building simulation model through text editing; Collect indoor temperature and humidity data, and simultaneously collect meteorological data; A multi-objective optimization model was constructed to minimize the error between the simulated temperature and humidity results and the measured temperature and humidity data of each independent functional area. Based on the non-dominated sorting genetic algorithm, the building temperature and humidity under different uncertain parameter values ​​are simulated, the multi-objective optimization model is solved, and the optimized non-dominated solution set is output to form the Pareto frontier. The optimal solution set is screened through Pareto front analysis, a calibrated building simulation model is established, the building energy consumption under future weather conditions is simulated, and the optimal renovation plan is output.

2. The automatic calibration method of building parameters based on multi-objective optimization according to claim 1 is characterized in that: The uncertain building parameters include: thermal performance parameters of the building envelope structure, internal heat source parameters, energy equipment parameters, and ventilation setting parameters.

3. The automatic calibration method of building parameters based on multi-objective optimization according to claim 1 is characterized in that: The building simulation model is established, specifically comprising: Build architectural geometric models based on architectural drawings or 3D geometric information and export them as IDF format files; Set the determined fixed parameters through IDF Editor; For uncertain parameters, parameterized definitions are performed, and parameterized placeholders are used to identify uncertain parameters. The placeholders are dynamically replaced through external scripts.

4. The automatic calibration method of building parameters based on multi-objective optimization according to claim 1 is characterized in that: A multi-objective optimization model is constructed to minimize the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data, including: Calculate the hourly air temperature measured in each room and the root mean square error coefficient of variation between the hourly relative temperature and the simulated value; Calculate the hourly air humidity measured in each room and the root mean square error coefficient of variation between the hourly relative humidity and the simulated value; The objective function of constructing the multi-objective optimization model is to minimize the root mean square error coefficient of variation of the air temperature in each room and to minimize the root mean square error coefficient of variation of the relative humidity in each room.

5. The automatic calibration method of building parameters based on multi-objective optimization according to claim 1 is characterized in that: Based on the non-dominated sorting genetic algorithm, the building temperature and humidity under different uncertain parameter values ​​are simulated, and the multi-objective optimization model is solved, including: Generate the initial population based on Latin hypercube sampling; Each individual in the population is input into the building simulation model to obtain the temperature and humidity simulation results of each independent functional area, and the target value of each individual is calculated according to the objective function of the multi-objective optimization model; The population is stratified and sorted, the non-dominated levels of individuals are determined, and the crowding distance of individuals in the same non-dominated layer is calculated; Combine the parent and offspring populations, retain individuals with high non-dominated levels and high crowding, and generate a new generation of populations through crossover and mutation; When it is determined that the iteration termination condition is reached, the optimized non-dominated solution set is output to form the Pareto front.

6. The automatic calibration method of building parameters based on multi-objective optimization according to claim 1 is characterized in that: Through Pareto frontier analysis, the optimal solution set is selected, a calibrated building simulation model is established, the building energy consumption under future weather conditions is simulated, and the optimal renovation plan is output, including: Select solutions that meet the set temperature objective function value and the set humidity objective function value from the Pareto optimal solution set, and generate a calibrated building simulation model based on the screening results; Generate future weather files based on future climate scenarios; Generate thermal performance improvement plans for building envelopes under future weather conditions based on the calibrated building simulation model; Based on the calibrated building simulation model, the building energy consumption under future weather conditions is simulated, the energy consumption changes under different renovation measures are compared to calculate the energy saving rate, and the optimal renovation plan is output.

7. An automatic calibration system for building parameters based on multi-objective optimization, characterized in that: include: Building parameter acquisition module, building simulation model construction module, environmental data acquisition module, multi-objective optimization model construction module, multi-objective optimization model solution module, and optimal transformation plan output module; The building parameter acquisition module is used to determine the building uncertain parameters to be calibrated and their value ranges according to the building type and construction year, with reference to the corresponding energy-saving design standards; The building simulation model construction module is used to establish a building simulation model, and embed the building uncertain parameters into the building simulation model through text editing; The environmental data acquisition module is used to collect indoor temperature data and humidity data, and simultaneously collect meteorological data; The multi-objective optimization model construction module is used to construct a multi-objective optimization model, with minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data as the optimization direction; The multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values ​​based on the non-dominated sorting genetic algorithm, solve the multi-objective optimization model, and output the optimized non-dominated solution set to form the Pareto frontier; The optimal renovation solution output module is used to screen the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation solution.

8. The automatic calibration system of building parameters based on multi-objective optimization according to claim 7 is characterized in that: The multi-objective optimization model construction module is used to construct a multi-objective optimization model, with minimizing the error between the simulated temperature and humidity results of each independent functional area and the measured temperature and humidity data as the optimization direction, specifically including: Calculate the hourly air temperature measured in each room and the root mean square error coefficient of variation between the hourly relative temperature and the simulated value; Calculate the hourly air humidity measured in each room and the root mean square error coefficient of variation between the hourly relative humidity and the simulated value; The objective function of constructing the multi-objective optimization model is to minimize the root mean square error coefficient of variation of the air temperature in each room and to minimize the root mean square error coefficient of variation of the relative humidity in each room.

9. The automatic calibration system of building parameters based on multi-objective optimization according to claim 7 is characterized in that: The multi-objective optimization model solving module is used to simulate the building temperature and humidity under different uncertain parameter values ​​based on the non-dominated sorting genetic algorithm, and solve the multi-objective optimization model, specifically including: Generate the initial population based on Latin hypercube sampling; Each individual in the population is input into the building simulation model to obtain the temperature and humidity simulation results of each independent functional area, and the target value of each individual is calculated according to the objective function of the multi-objective optimization model; The population is stratified and sorted, the non-dominated levels of individuals are determined, and the crowding distance of individuals in the same non-dominated layer is calculated; Combine the parent and offspring populations, retain individuals with high non-dominated levels and high crowding, and generate a new generation of populations through crossover and mutation; When it is determined that the iteration termination condition is reached, the optimized non-dominated solution set is output to form the Pareto front.

10. The automatic calibration system of building parameters based on multi-objective optimization according to claim 7, characterized in that: The optimal renovation solution output module is used to select the optimal solution set through Pareto front analysis, establish a calibrated building simulation model, simulate the building energy consumption under future weather conditions, and output the optimal renovation solution, specifically including: Select solutions that meet the set temperature objective function value and the set humidity objective function value from the Pareto optimal solution set, and generate a calibrated building simulation model based on the screening results; Generate future weather files based on future climate scenarios; Generate thermal performance improvement plans for building envelopes under future weather conditions based on the calibrated building simulation model; Based on the calibrated building simulation model, the building energy consumption under future weather conditions is simulated, the energy consumption changes under different renovation measures are compared to calculate the energy saving rate, and the optimal renovation plan is output.

Images