Building low-carbon transformation evaluation method based on multi-objective optimization

Abstract

The invention discloses a building low-carbon reconstruction evaluation method based on multi-objective optimization. The method comprises the following steps: acquiring real-time data; obtaining static data; based on the real-time data and the static data, a multi-target evaluation model covering the total carbon emission amount, the transformation cost, the energy-saving income and the investment payback period is constructed, and indoor comfort and project implementation constraint conditions are set for the multi-target evaluation model; an NSGA-II algorithm is adopted to carry out optimization solution on the multi-target evaluation model; and performing automatic screening on the Pareto optimal solution set generated by the NSGA-II algorithm by using an ideal point method to generate a unique recommended transformation scheme, wherein the recommended transformation scheme comprises an optimal low-carbon transformation strategy and various key performance indexes corresponding to the optimal low-carbon transformation strategy. According to the method, the problems of evaluation benchmark distortion, insufficient multi-target collaboration and strong subjectivity of a decision-making process of a current existing building low-carbon reconstruction evaluation method are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building construction, in particular to a building low-carbon reconstruction evaluation method based on multi-objective optimization. BACKGROUND

[0002] In the practice of existing building low-carbon reconstruction, the current evaluation method of existing building low-carbon reconstruction mainly has the following three shortcomings.

[0003] Firstly, the evaluation benchmark is distorted. Traditional evaluation methods mostly use static models based on historical statistical data or design values, which cannot accurately calibrate the evaluation benchmark using real-time operation energy consumption data of the building. This leads to inaccurate judgment of the current performance of the building, thereby causing deviation in the prediction of energy saving benefits of all subsequent reconstruction schemes from the beginning.

[0004] Secondly, the multi-objective coordination is insufficient. Existing methods often analyze indicators such as carbon emissions, reconstruction cost, and energy saving benefits separately, or use single-objective optimization, which makes it difficult to effectively handle the internal conflicts between these objectives.

[0005] Finally, the decision-making process is highly subjective. Even if some methods use multi-objective optimization algorithms to generate a Pareto solution set, the final scheme selection still mostly relies on artificial weight setting or subjective judgment. This approach is inefficient and lacks unified and objective standards, making it difficult to reproduce the decision-making results and lacking scientificity. SUMMARY

[0006] To overcome the defects of the prior art, the present application provides a building low-carbon reconstruction evaluation method based on multi-objective optimization to solve the problems of distorted evaluation benchmark, insufficient multi-objective coordination, and highly subjective decision-making process in the current evaluation method of existing building low-carbon reconstruction.

[0007] To achieve the above-mentioned purpose, a building low-carbon reconstruction evaluation method based on multi-objective optimization is provided, comprising the following steps:

[0008] Obtaining real-time data, including energy consumption data of the building, current market prices, and climate parameters;

[0009] Obtaining static data, including geometric information, envelope structure information, internal load information, and heating, ventilation, and air conditioning system information;

[0010] Based on the real-time data and the static data, a multi-objective evaluation model covering total carbon emissions, reconstruction cost, energy saving benefits, and investment recovery period is constructed, and indoor comfort and project implementation constraints are set for the multi-objective evaluation model;

[0011] The NSGA-II algorithm is used to optimize and solve the multi-objective evaluation model. During the optimization process, an adaptive crossover rate and mutation rate dynamic adjustment mechanism is used to balance the exploration and utilization capabilities of the population in order to maintain population diversity and promote algorithm convergence.

[0012] The Pareto optimal solution set generated by the NSGA-II algorithm is automatically screened using the ideal point method to generate a unique recommended modification scheme. The recommended modification scheme includes the optimal low-carbon modification strategy and its corresponding key performance indicators.

[0013] Furthermore, the energy consumption data includes electricity consumption data, gas consumption data, and water consumption data.

[0014] Furthermore, the current market price includes energy prices and carbon trading prices.

[0015] Furthermore, the climate parameters include outdoor dry-bulb temperature, outdoor relative humidity, total solar radiation intensity, wind speed, and wind direction.

[0016] Furthermore, the multi-objective assessment model includes a life-cycle carbon emission model, a total retrofit cost model, an annual comprehensive energy-saving economic benefit model, and a dynamic investment payback period model;

[0017] The life-cycle carbon emission model is as follows:

[0018]

[0019] Among them, C e For total lifecycle carbon emissions, Y represents the building lifecycle being assessed, and E represents the carbon emissions over the entire lifecycle. i Let EF be the predicted annual consumption of the i-th energy source. i Let be the carbon emission factor for the i-th energy source;

[0020] The total renovation cost model is as follows:

[0021]

[0022] Among them, C r For the total renovation cost of the renovation plan, I j C represents the initial investment cost of the j-th technical measure. a Other one-time costs related to the renovation;

[0023] The annual comprehensive energy-saving economic benefit model is as follows:

[0024]

[0025] Among them, B a For the annual comprehensive energy-saving economic benefits, B i R represents the annual consumption of the i-th type of energy in the building before the renovation.i P is the annual consumption of the i-th energy for the retrofitted building i C is the price of the i-th energy, ΔC is the annual operation and maintenance cost change introduced by the retrofit scheme, R c P is the annual carbon emission reduction amount generated by the retrofit scheme, P c C is the carbon trading price;

[0026] The dynamic investment payback period model is:

[0027]

[0028] PBP = k, PBP is the dynamic investment payback period of the retrofit scheme, B a,y C is the annual comprehensive energy-saving economic benefit in the y-th year, C r C is the total retrofit cost of the retrofit scheme.

[0029] Further, the constraint conditions of indoor comfort include thermal comfort constraints, indoor air quality constraints, and light comfort constraints.

[0030] Further, the constraint conditions of project implementation include investment upper limit constraints and construction period constraints.

[0031] Further, the step of optimizing and solving the multi-objective evaluation model by using the NSGA-II algorithm includes:

[0032] A retrofit scheme composed of a series of low-carbon retrofit technical measures is coded into a chromosome;

[0033] A certain number of initial chromosomes are randomly generated to form an initial population;

[0034] The multi-objective evaluation model is called for each chromosome in the current population to calculate the values of the four optimization objective functions;

[0035] For all individuals in the current population that meet the constraints, fast non-dominated sorting is performed according to the objective function values;

[0036] Based on the non-dominated level and the crowding distance, parent individuals are selected from the current population, and crossover and mutation operations are performed according to the crossover probability and the mutation probability to generate offspring individuals. The offspring individuals are combined with the current population to form a temporary population. The individuals in the temporary population are non-dominated sorted and crowdedness calculated, and the optimal individuals are selected to form the next generation population. After iteration termination, the Pareto optimal solution set is output.

[0037] Further, the step of automatically screening the Pareto optimal solution set generated by the NSGA-II algorithm using the ideal point method to generate a unique recommended retrofit scheme includes:

[0038] The linear proportional transformation method is used for normalizing each target value of each scheme in the Pareto optimal solution set, so as to eliminate the influence of different target dimensions and orders of magnitude;

[0039] The weighted Euclidean distance between each normalized scheme and the normalized positive ideal solution and the weighted Euclidean distance between each normalized scheme and the normalized negative ideal solution are calculated;

[0040] The closeness degree of each scheme relative to the ideal solution is calculated;

[0041] All the schemes in the Pareto optimal solution set are arranged in descending order based on the closeness degree, and the scheme with the largest closeness degree is taken as the only recommended reconstruction scheme

[0042] The building low-carbon reconstruction evaluation method based on multi-objective optimization has the advantages that the building benchmark model is accurately calibrated by introducing real-time operation energy consumption data, the starting point of the evaluation is highly consistent with the actual performance of the building, and therefore the accuracy and reliability of benefit prediction of all subsequent reconstruction schemes are greatly improved.

[0043] The building low-carbon reconstruction evaluation method based on multi-objective optimization constructs a multi-objective evaluation model covering core economic and environmental indicators such as carbon emission, reconstruction cost, dynamic energy saving benefit, investment recovery period, and simultaneously considers indoor comfort and project implementation constraints. The NSGA-II algorithm is used for solving, effective trade-off between multiple conflicting objectives can be achieved, and a series of Pareto solutions with better comprehensive benefits are generated.

[0044] The building low-carbon reconstruction evaluation method based on multi-objective optimization adopts an automatic screening mechanism based on preset rules, objectively recommends a unique reconstruction scheme from the Pareto optimal solution set, avoids the subjectivity and inefficiency of traditional manual screening, and provides a scientific and standardized decision support tool for decision makers. BRIEF DESCRIPTION OF DRAWINGS

[0045] Other features, objects and advantages of the application will become more apparent after reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0046] Figure 1 The flow chart of the building low-carbon reconstruction evaluation method based on multi-objective optimization of the embodiments of the application.

[0047] Figure 2 The flow chart of the embodiments of the application in which the NSGA-II algorithm is used to optimize and solve the multi-objective evaluation model. DETAILED DESCRIPTION

[0048] The application will be described in further detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related application, and not to limit the application. In addition, it should be noted that only parts related to the application are shown in the drawings for ease of description.

[0049] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and embodiments.

[0050] Referring to Figure 1 The application provides a building low-carbon reconstruction evaluation method based on multi-objective optimization, comprising the following steps:

[0051] S1, acquiring real-time data, the real-time data including building energy consumption data, current market price, and climate parameters.

[0052] In this embodiment, the energy consumption data includes power consumption data, gas consumption data, and water consumption data.

[0053] The current market price includes energy prices and carbon trading prices.

[0054] The climate parameters include outdoor dry-bulb temperature, outdoor relative humidity, total solar radiation intensity, wind speed, and wind direction.

[0055] Preferably, the various energy consumption data of the building is collected in real time through Internet of Things sensors, and the current market prices such as time-of-use electricity prices and carbon trading prices, and the climate parameters of the meteorological database are obtained through integrated application program interfaces.

[0056] Specifically, the energy consumption data collected through the Internet of Things sensors includes power consumption data, gas consumption data, and water consumption data.

[0057] The power consumption data is collected in real time by installing a smart electricity meter or deploying a current transformer and a multifunctional power meter on a key power distribution circuit.

[0058] The gas consumption data is obtained by installing a smart gas meter or a flow meter in a building using a gas boiler for heating / heat water or having a centralized gas kitchen.

[0059] The water consumption data is obtained by installing a smart water meter.

[0060] The current market price includes energy prices and carbon trading prices.

[0061] Among them, by connecting the API of energy suppliers such as power companies and gas companies, real-time or preset time-of-use electricity prices, step gas prices, peak-valley electricity prices, etc. can be obtained.

[0062] Through the API of the national or local carbon emission trading market, real-time or recent carbon quota trading prices can be obtained for evaluating the economic value of carbon emission reduction.

[0063] Meteorological data refers to meteorological data service API of the national or local meteorological bureau, or small meteorological stations locally deployed by the building to obtain hourly outdoor meteorological parameters. Meteorological parameters include outdoor dry-bulb temperature, outdoor relative humidity, total solar radiation intensity, wind speed, and wind direction. These data will be used for building energy consumption simulation and indoor comfort evaluation.

[0064] S2, obtain static data, including geometric information, envelope information, internal load information, and HVAC system information.

[0065] In this embodiment, static data is preferentially extracted or exported from BIM.

[0066] Static data includes geometric information, envelope information, internal load information, and HVAC system information.

[0067] Specifically, the geometric information includes the total area of the building, the area of each floor, the height of the building, the orientation of the building, and the window-wall ratio.

[0068] The envelope information includes the material type, thickness, construction level, and thermal parameters of the wall, roof, floor, and exterior doors and windows.

[0069] The internal load information includes the personnel density, lighting device power density, office equipment power density, and their hourly usage rates.

[0070] The HVAC system information includes the type and capacity of the cold and heat sources, the type and number of terminal equipment, the system control method, and the design energy efficiency parameters.

[0071] If there is no BIM model, or the BIM model information is incomplete or inaccurate, static data needs to be obtained and supplemented through the following methods:

[0072] Consult building design drawings, as-built drawings, historical renovation materials, and equipment account books, etc.

[0073] Use three-dimensional laser scanning and other technologies to assist in obtaining accurate building geometric information;

[0074] Refer to the relevant national and local building energy-saving design standards, specifications, or typical building databases to reasonably estimate or value some parameters that are difficult to obtain directly.

[0075] S3, based on real-time data and static data, construct a multi-objective evaluation model covering total carbon emissions, retrofit cost, energy-saving benefits, and investment payback period, and set indoor comfort and project implementation constraints for the multi-objective evaluation model.

[0076] The basis of the multi-objective evaluation model is to construct a low-carbon renovation technology measure library which not only lists technical options, but also contains key parameters required for quantitative evaluation.

[0077] The technology library used in the method of the present application covers common technical categories in low-carbon renovation of existing buildings.

[0078] Table 1 below shows, by way of example, some technical categories and example values of their key parameters.

[0079] It should be emphasized that the specific values in the table are only examples, and in actual application, they should be filled in and updated according to the specific circumstances of the project, local market conditions, the latest product information and technical standards.

[0080] Those skilled in the art can obtain these parameters through routine technical research, product manual review, market inquiry, engineering quota, and public databases and standards.

[0081] Table 1: Parameter representation examples of low-carbon renovation technology measure library

[0082]

[0083]

[0084] In addition to the core parameters listed in Table 1, each measure in the technology library should also be associated with changes in operation and maintenance costs, construction period impacts, applicability and constraint associations, etc.

[0085] For each potential renovation scheme composed of specific combinations of different technical measures in the low-carbon renovation technology measure library shown in Table 1, the present application predicts its energy consumption performance by using EnergyPlus.

[0086] Before that, the energy consumption baseline model needs to be calibrated.

[0087] First, based on the extracted static data, an initial building energy consumption baseline model is established in EnergyPlus.

[0088] Then, using at least one week or more of real-time operating energy consumption data collected, the initial model is calibrated, and the uncertainty parameters in the model are adjusted until the deviation between the model's simulation energy consumption and the actual collected energy consumption data meets the preset accuracy requirement. The calibrated baseline model accurately reflects the current actual performance of the building and is the core basis for all subsequent evaluations.

[0089] The multi-objective evaluation model includes a life cycle carbon emission model, a total renovation cost model, an annual comprehensive energy-saving economic benefit model, and a dynamic investment payback period model.

[0090] The full life cycle carbon emission model (C e ) is:

[0091]

[0092] Wherein, C e is the full life cycle carbon emission, unit is kgCO2;

[0093] Y is the evaluated building life cycle year, unit is year;

[0094] E i is the predicted consumption of the i-th energy in a year, calculated by EnergyPlus, unit is / year;

[0095] EF i is the carbon emission factor of the i-th energy, unit is kgCO2 / unit.

[0096] The total transformation cost model (C r ) is:

[0097]

[0098] Wherein, C r is the total transformation cost of the transformation scheme, unit is yuan;

[0099] I j is the initial investment cost of the j-th technical measure, unit is yuan;

[0100] C a is other one-off fees related to the transformation, unit is yuan.

[0101] The annual comprehensive energy-saving economic benefit model (B a ) is:

[0102]

[0103] Wherein, B a is the annual comprehensive energy-saving economic benefit, unit is yuan / year;

[0104] B i is the annual consumption of the i-th energy of the building before transformation, unit is / year;

[0105] R i is the annual consumption of the i-th energy of the building after transformation, unit is / year;

[0106] P i is the price of the i-th energy, unit is yuan / unit;

[0107] ΔC is the annual operation and maintenance cost change introduced by the transformation scheme, unit is yuan / year;

[0108] R c annual carbon emission reduction generated by the retrofit scheme, in kgCO2 / year;

[0109] P c carbon trading price, in yuan / kgCO2.

[0110] The dynamic payback period model (PBP) is:

[0111]

[0112] wherein PBP is the dynamic payback period of the retrofit scheme, in years;

[0113] B a,y annual comprehensive energy-saving economic benefit of the yth year, in yuan;

[0114] C r total retrofit cost of the retrofit scheme, in yuan.

[0115] find the smallest year k (1≤k≤Y) such that PBP=k.

[0116] In this embodiment, the constraint conditions of indoor comfort include thermal comfort constraint, indoor air quality constraint, and light comfort constraint.

[0117] The thermal comfort constraint is set according to ASHRAE Standard 55 or GB / T 50785 “Standard for Evaluation of Indoor Thermal and Humidity Environment of Civil Buildings”, the indoor air quality constraint is set according to GB / T 18883 “Standard for Indoor Air Quality”, and the light comfort constraint is set according to GB 50034 “Standard for Architectural Lighting Design”.

[0118] In this embodiment, the constraint conditions of project implementation include investment upper limit constraint and construction period constraint.

[0119] The investment upper limit constraint is to ensure that the total retrofit cost (C r ) is less than the preset investment upper limit of the project, and the construction period constraint is to ensure that the construction period corresponding to the retrofit scheme obtained according to the low-carbon retrofit technical measure library is less than the preset construction period of the project.

[0120] S4, the NSGA-II algorithm is used to optimize and solve the multi-objective evaluation model. In the optimization process, the dynamic adjustment mechanism of the adaptive crossover rate and the mutation rate is used to balance the exploration and utilization ability of the population to maintain the population diversity and promote the convergence of the algorithm.

[0121] Specifically, as shown in FIG. 4, step S4 includes: Figure 2

[0122] ​S41, encode a chromosome for the retrofit scheme composed of a series of low-carbon retrofit technical measures.

[0123] First, encode the population initialization, that is, encode a chromosome for the retrofit scheme composed of a series of low-carbon retrofit technical measures.

[0124] In an embodiment of the present application, binary encoding can be used, where each bit of the chromosome corresponds to a basic technical measure option in the technical library, and "1" indicates that the measure is adopted, and "0" indicates that the measure is not adopted.

[0125] For technical measures with continuous parameters, real number encoding or discretization into different levels represented by multiple binary bits can be used.

[0126] S42, randomly generate a certain number of initial chromosomes to form an initial population.

[0127] A certain number of initial chromosomes are randomly generated to form an initial population P0. The size of the population size N can be set according to the complexity of the problem and the computing resources.

[0128] S43, call the multi-objective evaluation model for each chromosome in the current population to calculate the values of the four optimization objective functions.

[0129] Call the multi-objective evaluation model constructed above for each chromosome in the current population P t (t is the current iteration number).

[0130] Specifically, it is necessary to determine whether the chromosome satisfies the indoor comfort constraint and the project implementation constraint, mark the scheme that does not satisfy the constraint as infeasible, and calculate the values of the four optimization objective functions for the chromosome that satisfies the constraint: the total life cycle carbon emission (C e ), the total retrofit cost (C r ), the annual comprehensive energy-saving economic benefit (B a ), and the dynamic payback period (PBP).

[0131] S44, for all individuals in the current population that satisfy the constraints, perform fast non-dominated sorting according to the objective function values.

[0132] For all individuals in the current population P tAll individuals in the population that satisfy the constraints are ranked according to their objective function values. The population is divided into different non-dominated levels (F1, F2,..., Fk). F1 contains all non-dominated solutions, F2 contains the non-dominated solutions among the individuals remaining after F1 is removed, and so on. For individuals in the same non-dominated level, their crowding distances are calculated. The crowding distance is used to measure the sparsity of an individual in its neighborhood, aiming to maintain the diversity of the population in the objective space and avoid the solutions from concentrating in certain regions.

[0133] S45, based on the non-dominated level and the crowding distance, tournament selection method is used to select parent individuals from the current population, and cross and mutation operations are performed to generate offspring individuals according to the cross probability and the mutation probability, and then the offspring individuals are combined with the current population to form a temporary population. The individuals in the temporary population are non-dominated sorted and the crowding distance is calculated, and then the optimal individuals are selected to form the next generation population. After the iteration is terminated, the Pareto optimal solution set is output.

[0134] Offspring population Q t is generated, including selection, cross and mutation.

[0135] Selection, that is, using the elitist strategy and combining the tournament selection method based on the non-dominated level and the crowding distance, selects individuals with lower non-dominated level and larger crowding distance from the parent population P t to participate in reproduction.

[0136] Cross, that is, for the selected parent individuals, cross operation is performed according to the cross probability P c determined at the current iteration number t to generate offspring individuals. The cross probability P c is mainly adjusted in a non-linear decreasing manner according to the iteration process of the algorithm. In the early stage of the algorithm iteration, a higher cross probability is needed to promote the population to conduct extensive search and explore more possible solution space. As the iteration proceeds, when the algorithm gradually converges, the cross probability needs to be reduced to protect the discovered excellent genotypes and promote the algorithm to fine convergence to the optimal solution. Its calculation formula is:

[0137]

[0138] Wherein, P c (t) is the cross probability of the tth generation;

[0139] P c,max is the preset maximum cross probability, generally 0.9;

[0140] P c,min is the preset minimum cross probability, generally 0.4;

[0141] t is the current iteration number;

[0142] Gmax is the preset maximum iteration number;

[0143] β is a nonlinear adjustment factor for controlling the speed of the crossover probability decrease, generally 2.

[0144] Variation is to mutate the offspring individuals generated by crossover, or directly select individuals from the parents, according to the mutation probability P m (t) determined at the current iteration number t. The mutation probability P m is mainly dynamically adjusted according to the current population diversity level. When the population diversity is high, it means that the population is still exploring widely, and a lower mutation rate is needed. When the population diversity decreases to a certain extent, it means that the algorithm may fall into local optimum or tend to premature convergence, and the mutation rate needs to be increased to introduce new genes to increase population diversity and help the algorithm jump out of local optimum. Its calculation formula is:

[0145]

[0146] In the formula, P m (t) is the mutation probability of the tth generation;

[0147] P m,max is the preset maximum mutation probability, generally 0.2;

[0148] γ is an adjustment factor for controlling the sensitivity of the mutation probability to the change of diversity, generally 1;

[0149] t is the current iteration number;

[0150] D(t) is the average crowding distance of the first non-dominated front of the tth generation;

[0151] D max is the maximum average crowding distance that has occurred from the beginning of the algorithm to the t-1th generation.

[0152] Then, merging and elitist selection, that is, merging the parent population P t and the offspring population Q t generated by the above genetic operations to form a temporary population R t with a size of 2N. All individuals in R t are non-dominated sorted and crowdedness calculated. Then, the best N individuals are selected from them (preferentially selecting individuals with lower non-dominated levels, and preferentially selecting individuals with larger crowding distances in the same level) to form the next generation parent population P t+1 .

[0153] Finally, the iteration termination judgment is performed. The above fitness evaluation is repeated until the iteration process of merging and elite selection is terminated, until the preset iteration termination condition is met, such as reaching the maximum iteration number G max, or the Pareto optimal frontier is not significantly improved in a plurality of consecutive generations. After the algorithm is terminated, the individuals in F1 in the final output P Gmax constitute the Pareto optimal solution set obtained in this round of optimization.

[0154] S5, the Pareto optimal solution set generated by the NSGA-II algorithm is automatically screened by the ideal point method to generate a unique recommended modification scheme, and the recommended modification scheme includes an optimal low-carbon modification strategy and corresponding key performance indicators.

[0155] Specifically, step S5 includes:

[0156] S51, the target values of each scheme in the Pareto optimal solution set are normalized by linear scaling transformation method to eliminate the influence of different target dimensions and orders of magnitude.

[0157] S52, the weighted Euclidean distance between each normalized scheme and the normalized positive ideal solution and the weighted Euclidean distance between each normalized scheme and the normalized negative ideal solution are calculated.

[0158] S53, the fitting degree of each scheme relative to the ideal solution is calculated.

[0159] S54, all schemes in the Pareto optimal solution set are ranked in descending order based on the fitting degree, and the scheme with the largest fitting degree is taken as the unique recommended modification scheme.

[0160] In this embodiment, the automatic screening of the generated Pareto optimal solution set means using the ideal point method (TOPSIS) to generate a unique recommended modification scheme.

[0161] The ideal point method needs to find the optimal value and the worst value of each optimization target (minimization of total life cycle carbon emissions Ce, minimization of total modification cost Cr, maximization of annual comprehensive energy-saving economic benefit Ba, minimization of dynamic payback period PBP) of all schemes in the Pareto optimal solution set. Among them, the virtual solution composed of the optimal value of each target is the positive ideal solution (PIS), and the virtual solution composed of the worst value of each target is the negative ideal solution (NIS).

[0162] After the positive and negative ideal solutions are determined, the ideal point method needs to calculate the distance between each scheme and the positive and negative ideal solutions. Specifically, the target values of each scheme in the Pareto optimal solution set are normalized by linear scaling transformation method to eliminate the influence of different target dimensions and orders of magnitude. Then, the weighted Euclidean distance (D j +) and the weighted Euclidean distance (D j - ), the calculation formula is as follows:

[0163]

[0164] wherein,

[0165] v jk is the kth normalized target value of the jth scheme;

[0166] V k + and v k - are the positive ideal value and the negative ideal value of the kth normalized target respectively;

[0167] w k is the weight, which is set by the decision maker according to the importance of each target;

[0168] m is the number of targets.

[0169] After calculating the distance, the ideal point method needs to calculate the closeness degree C j * of each scheme j relative to the ideal solution, the calculation formula is as follows:

[0170]

[0171] C j * The value range of C j * is between 0 and 1, the closer to 1 indicates that the scheme is closer to the positive ideal solution and farther away from the negative ideal solution, and the more comprehensive the performance is. According to the calculated relative closeness C j * , all the schemes in the Pareto optimal solution set are ranked in descending order. The scheme with the maximum value of C

[0172] The building low-carbon reconstruction evaluation method based on multi-objective optimization of the application firstly collects various energy consumption data of the building in real time through the Internet of Things sensor, and integrates API interface to obtain current market prices such as time-of-use electricity price, carbon trading price and climate parameters of meteorological database; secondly, static data such as building area and structure type are extracted by using BIM and other technical means, and the upper limit of investment and construction period are preset as constraint conditions according to project requirements; thirdly, based on the integrated real-time data and extracted static data, a multi-objective evaluation model covering total carbon emission, reconstruction cost, energy saving benefit and investment payback period is constructed, and indoor comfort and project implementation constraints are set for the model; then, the improved NSGA-II algorithm is used to optimize and solve the multi-objective evaluation model, and the dynamic adjustment mechanism of adaptive crossover rate and mutation rate is used to balance the exploration and utilization ability of the population in the optimization process, so as to maintain the diversity of the population and promote the convergence of the algorithm; then, based on the preset decision rule, the Pareto optimal solution set generated by the NSGA-II algorithm is automatically screened to generate a unique recommended reconstruction scheme; finally, the recommended optimal low-carbon reconstruction strategy and its corresponding key performance indicators are output, including specific reconstruction measure combination, predicted total carbon emission, reconstruction cost, energy saving benefit and investment payback period.

[0173] The building low-carbon reconstruction evaluation method based on multi-objective optimization of the application accurately calibrates the building benchmark model by introducing real-time operation energy consumption data, which ensures that the starting point of the evaluation is highly consistent with the actual performance of the building, thereby greatly improving the accuracy and reliability of the benefit prediction of all subsequent reconstruction schemes.

[0174] The building low-carbon reconstruction evaluation method based on multi-objective optimization of the application constructs a multi-objective evaluation model covering core economic and environmental indicators such as carbon emission, reconstruction cost, dynamic energy saving benefit and investment payback period, and simultaneously considers indoor comfort and project implementation constraints. By using the improved NSGA-II algorithm for solving, the algorithm can effectively balance multiple conflicting objectives and generate a series of Pareto solutions with better comprehensive benefits.

[0175] The building low-carbon reconstruction evaluation method based on multi-objective optimization of the application adopts an automatic screening mechanism based on preset rules to objectively recommend a unique reconstruction scheme from the Pareto optimal solution set, avoiding the subjectivity and inefficiency of traditional manual screening, and providing a scientific and standardized decision support tool for decision makers.

[0176] The above description is only the preferred embodiment of the present application and the explanation of the technical principles. It should be understood by those skilled in the art that the scope of the protection of the present application is not limited to the technical solutions formed by the specific combinations of the above technical features. It should also cover other technical solutions formed by the combinations of the above technical features or their equivalent features without departing from the concept of the present application. For example, the technical solutions formed by the mutual replacements of the above features and the technical features disclosed in the present application (but not limited to) with similar functions.

Claims

1. A method for evaluating low-carbon retrofitting of buildings based on multi-objective optimization, characterized in that, Includes the following steps: Acquire real-time data, including building energy consumption data, current market prices, and climate parameters; Acquire static data, which includes geometric information, building envelope information, internal load information, and HVAC system information; Based on the real-time data and the static data, a multi-objective evaluation model is constructed, covering total carbon emissions, renovation costs, energy-saving benefits and investment payback period, and indoor comfort and project implementation constraints are set for the multi-objective evaluation model. The NSGA-II algorithm is used to optimize and solve the multi-objective evaluation model. During the optimization process, an adaptive crossover rate and mutation rate dynamic adjustment mechanism is used to balance the exploration and utilization capabilities of the population in order to maintain population diversity and promote algorithm convergence. The Pareto optimal solution set generated by the NSGA-II algorithm is automatically screened using the ideal point method to generate a unique recommended modification scheme. The recommended modification scheme includes the optimal low-carbon modification strategy and its corresponding key performance indicators.

2. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The energy consumption data includes electricity consumption data, gas consumption data, and water consumption data.

3. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The current market prices mentioned include energy prices and carbon trading prices.

4. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The climate parameters include outdoor dry-bulb temperature, outdoor relative humidity, total solar radiation intensity, wind speed, and wind direction.

5. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The multi-objective assessment model includes a life-cycle carbon emission model, a total retrofit cost model, an annual comprehensive energy-saving economic benefit model, and a dynamic investment payback period model. The life-cycle carbon emission model is as follows: Among them, C e For total lifecycle carbon emissions, Y represents the building lifecycle being assessed, and E represents the carbon emissions over the entire lifecycle. i Let EF be the predicted annual consumption of the i-th energy source. i Let be the carbon emission factor for the i-th energy source; The total renovation cost model is as follows: Among them, C r For the total renovation cost of the renovation plan, I j C represents the initial investment cost of the j-th technical measure. a Other one-time costs related to the renovation; The annual comprehensive energy-saving economic benefit model is as follows: Among them, B a For the annual comprehensive energy-saving economic benefits, B i R represents the annual consumption of the i-th type of energy in the building before the renovation. i P represents the annual consumption of the i-th type of energy in the renovated building. i Let R be the price of the i-th energy source, ΔC be the annual change in operation and maintenance costs introduced by the retrofit plan, and R be the price of the ith energy source. c The annual carbon emission reduction generated by the renovation plan, P c For carbon trading prices; The dynamic payback period model is as follows: Where, PBP = k, PBP is the dynamic payback period of the renovation plan, and B a,y For the comprehensive energy-saving economic benefits in year y, C r This represents the total cost of the renovation plan.

6. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The constraints on indoor comfort include thermal comfort constraints, indoor air quality constraints, and light comfort constraints.

7. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The constraints on the implementation of the project include an investment ceiling and a construction period constraint.

8. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 1, characterized in that, The steps for optimizing the multi-objective evaluation model using the NSGA-II algorithm include: The transformation plan, which consists of a series of low-carbon transformation technologies, is encoded into a chromosome. A certain number of initial chromosomes are randomly generated to form the initial population; For each chromosome in the current population, the multi-objective evaluation model is invoked to calculate the values ​​of the four optimization objective functions for the activity; For all individuals in the current population that satisfy the constraints, perform a fast non-dominated sort based on the objective function value; The tournament selection method based on non-dominated hierarchy and crowding distance selects parent individuals from the current population, and performs crossover and mutation operations based on crossover and mutation probabilities to generate offspring individuals. The offspring individuals are then merged with the current population to form a temporary population. The individuals in the temporary population are sorted by non-dominated hierarchy and crowding degree is calculated. The optimal individuals are then selected to form the next generation population. After the iteration terminates, the Pareto optimal solution set is output.

9. The building low-carbon retrofit assessment method based on multi-objective optimization according to claim 8, characterized in that, The step of automatically filtering the Pareto optimal solution set generated by the NSGA-II algorithm using the ideal point method to generate a unique recommended modification scheme includes: The objective values ​​of each scheme in the Pareto optimal solution set are normalized by the linear scaling transformation method to eliminate the influence of different objective dimensions and orders of magnitude. Calculate the weighted Euclidean distance between each normalized scheme and the normalized positive ideal solution, as well as the weighted Euclidean distance between each scheme and the normalized negative ideal solution; Calculate the patching progress relative to the ideal solution for each scheme; Based on the patching progress, all solutions in the Pareto optimal solution set are sorted in descending order, and the solution with the largest patching progress is selected as the sole recommended modification solution.

Images