Building carbon emission problem determination method and device and storage medium
Through carbon emission characteristic analysis and model training based on building energy consumption data, the problem of inefficient building energy conservation diagnosis is solved, and efficient and accurate identification and diagnosis of building carbon emission problems is achieved.
Patent Information
- Application Number
- CN202411553267.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-01
- Publication Date
- 2025-02-25
AI Technical Summary
The existing technology is difficult to identify building operation characteristics in a timely and accurate manner, resulting in low efficiency and high cost in building energy-saving diagnosis, and lack of methods to deeply integrate with digital technology, resulting in serious lag and poor accuracy in building carbon emission accounting and diagnostic evaluation.
By determining the reference carbon emission characteristics based on the reference energy consumption data of the reference building, identifying multiple carbon emission problem labels, and using these labels and reference energy consumption data, the initial model is trained to obtain the target model. Then, the target energy usage data of the target building is input into the target model and processed to determine the carbon emission problem identification results in the target building.
It has achieved refined energy-saving and carbon reduction diagnosis for buildings, improved the efficiency and accuracy of building energy-saving diagnosis, reduced lag, and solved the problem of unsatisfactory diagnosis efficiency of building energy-saving diagnosis.
Smart Images
Figure CN119515633A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data analysis and the field of building carbon emissions, and specifically, to a method, device and storage medium for determining building carbon emission problems. Background Art
[0002] In recent years, the construction of large-scale new buildings and the operation of existing buildings have generated a large amount of carbon dioxide emissions, which is the second largest source of carbon emissions after energy and electricity. In the entire life cycle of a building, carbon emissions during the building operation phase account for the largest proportion.
[0003] At present, the relevant technologies for building carbon emission accounting and diagnostic assessment are based on years, which cannot identify the building operation characteristics in a timely and accurate manner. Energy-saving diagnosis and carbon efficiency management mainly rely on manual consultation, which is inefficient, costly, and has serious lags. In addition, there is a lack of methods and means for deep integration with digital technology, and the accuracy is poor. Many buildings lack data statistics on energy-using subsystems, making it difficult to carry out refined building energy-saving diagnosis, which is not conducive to targeted energy-saving management of buildings. Therefore, there are technical problems with the efficiency of building energy-saving diagnosis.
[0004] There is currently no effective solution to the above problem tags. Summary of the invention
[0005] The embodiments of the present application provide a method, device and storage medium for determining building carbon emission problems, so as to at least solve the technical problem of unsatisfactory efficiency of building energy-saving diagnosis.
[0006] According to one aspect of an embodiment of the present application, a method for determining building carbon emission problems is provided, comprising: determining a reference carbon emission characteristic based on reference energy consumption data of a reference building, wherein the reference building includes multiple subsystems with different energy consumption demands; determining multiple carbon emission problem labels existing in the reference building based on the reference carbon emission characteristic, wherein the multiple carbon emission problem labels represent the operating status of corresponding subsystems; training an initial model based on the multiple carbon emission problem labels and the reference energy consumption data to obtain a target model; inputting the target energy consumption data of the target building into the target model for processing to determine a carbon emission problem identification result in the target building.
[0007] Optionally, the reference carbon emission characteristics are determined based on the reference energy consumption data of the reference building, including: determining the carbon emission density data of the reference building based on the reference energy consumption data and a predetermined carbon emission factor; when there are multiple reference buildings, the reference carbon emission characteristics are determined by fitting based on the carbon emission density data corresponding to the multiple reference buildings.
[0008] Optionally, the carbon emission density data includes a baseline carbon emission density, a cooling carbon emission density, and a heating carbon emission density, and the fitting is performed based on the carbon emission density data corresponding to multiple reference buildings to determine the reference carbon emission characteristics, including: fitting according to the baseline carbon emission density, the cooling carbon emission density, and the heating carbon emission density corresponding to the multiple reference buildings to obtain a fitting curve characterizing different room temperatures and carbon emission densities; determining the reference carbon emission characteristics based on the segmented fitting trend and turning points in the fitting curve.
[0009] Optionally, determining multiple carbon emission problem labels existing in the reference building based on the reference carbon emission characteristics includes: determining sub-feature parameters corresponding to multiple sub-reference systems included in the reference building in the reference carbon emission characteristics; determining score values of the sub-feature parameters corresponding to the multiple sub-reference systems; among the sub-feature parameters corresponding to the multiple sub-reference systems, determining sub-feature parameters having score values less than a predetermined score threshold as problem labels; and determining the multiple carbon emission problem labels based on the problem labels corresponding to the multiple sub-reference systems.
[0010] Optionally, the target model is a target decision tree, and the initial model is trained based on the multiple carbon emission problem labels and the reference energy consumption data to obtain the target model, including: determining the information gain of the nodes included in the initial decision tree based on the reference energy consumption data; determining predetermined characteristic conditions based on the multiple carbon emission problem labels; and dividing the initial decision tree into nodes according to the information gain and the predetermined characteristic conditions to determine the target decision tree.
[0011] Optionally, determining the information gain of the nodes in the initial decision tree based on the reference energy usage data includes: determining the information entropy and conditional entropy of the reference energy usage data; and determining the information gain based on the information entropy and the conditional entropy.
[0012] Optionally, the method also includes: obtaining initial energy consumption data of the reference building, wherein the initial energy consumption data includes: cooling data, heating data, electricity consumption data, and gas consumption data; when there are differences in sampling time among the cooling data, the heating data, the electricity consumption data, and the gas consumption data, data alignment processing is performed on the cooling data, the heating data, the electricity consumption data, and the gas consumption data, respectively, to obtain the reference energy consumption data.
[0013] According to another aspect of an embodiment of the present application, a device for determining building carbon emission problems is provided, including: a feature determination module, used to determine a reference carbon emission feature based on reference energy consumption data of a reference building, wherein the reference building includes multiple subsystems with different energy consumption demands; a label determination module, used to determine a plurality of carbon emission problem labels existing in the reference building based on the reference carbon emission feature, wherein the plurality of carbon emission problem labels represent the operating status of corresponding subsystems; a model acquisition module, used to train an initial model based on the plurality of carbon emission problem labels and the reference energy consumption data to obtain a target model; and a result determination module, used to input the target energy consumption data of the target building into the target model for processing to determine the carbon emission problem identification result in the target building.
[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is provided, wherein the non-volatile storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing any one of the methods for determining building carbon emission problems.
[0015] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising: one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any one of the methods for determining building carbon emission problems.
[0016] In the embodiment of the present application, a building carbon emission problem determination method is adopted, and a reference carbon emission feature is determined based on the reference energy consumption data of a reference building, wherein the reference building includes multiple subsystems with different energy consumption requirements; based on the reference carbon emission feature, multiple carbon emission problem labels existing in the reference building are determined, wherein the multiple carbon emission problem labels represent the operating status of the corresponding subsystems; based on the multiple carbon emission problem labels and the reference energy consumption data, the initial model is trained to obtain a target model; the target energy consumption data of the target building is input into the target model for processing to determine the carbon emission problem identification result in the target building. The purpose of conducting refined energy-saving and carbon-reduction diagnosis for buildings is achieved, and the technical effect of improving the efficiency and accuracy of building energy-saving diagnosis and reducing hysteresis is achieved, thereby solving the technical problem of unsatisfactory efficiency of building energy-saving diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0018] Figure 1is a flow chart of an optional method for determining building carbon emission problems provided according to an embodiment of the present application;
[0019] Figure 2 is a first schematic flow chart of an optional method for determining a building carbon emission problem provided according to an embodiment of the present application;
[0020] Figure 3 is a second schematic flow chart of an optional method for determining a building carbon emission problem provided according to an embodiment of the present application;
[0021] Figure 4 It is a schematic diagram of an optional device for determining building carbon emission problems provided according to an embodiment of the present application. DETAILED DESCRIPTION
[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.
[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0024] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:
[0025] Information entropy refers to the uncertainty of possible events in the information source. The more chaotic the information source, the greater the entropy. Entropy is the weighted average of the amount of information.
[0026] Conditional Entropy is the uncertainty of another random variable under the condition that a certain random variable is known.
[0027] Information Gain is used to measure the difference between two probability distributions P and Q. Information Gain describes the difference when encoding with Q and then encoding with P. Usually P represents the distribution of samples or observations, or it may be a theoretical distribution that is precisely calculated. Q represents a description or approximation of P.
[0028] According to an embodiment of the present application, a method embodiment for determining a building carbon emission problem is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0029] Figure 1 is a flowchart of an optional method for determining building carbon emission problems provided in an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:
[0030] Step S102, determining a reference carbon emission characteristic based on reference energy consumption data of a reference building, wherein the reference building includes a plurality of subsystems with different energy consumption demands;
[0031] It can be understood that the reference building includes multiple subsystems with different energy demands. The reference carbon emission characteristics are determined based on the reference energy data of the reference building. By analyzing the energy demands of different subsystems in the reference building, the energy consumption patterns and carbon emission characteristics of each subsystem can be more accurately identified, and the main sources of building carbon emissions can be finely identified, providing a more accurate basis for the formulation of emission reduction measures.
[0032] In an optional embodiment, a reference carbon emission characteristic is determined based on the reference energy consumption data of the reference building, including: determining the carbon emission density data of the reference building based on the reference energy consumption data and a predetermined carbon emission factor; in the case of multiple reference buildings, fitting is performed based on the carbon emission density data corresponding to the multiple reference buildings to determine the reference carbon emission characteristic.
[0033] It can be understood that the carbon emission density data of the reference building is determined based on the reference energy consumption data and the predetermined carbon emission factor. When there are multiple reference buildings, the carbon emission density data corresponding to the multiple reference buildings are fitted to determine the reference carbon emission characteristics. By determining the carbon emission characteristics with reference energy consumption data and carbon emission factors, the carbon emissions of the building under specific conditions can be accurately calculated.
[0034] Optionally, there may be multiple methods for determining the reference carbon emission characteristics, for example, Figure 2 is a first schematic flow chart of an optional method for determining building carbon emission problems provided in an embodiment of the present application, such as Figure 2 As shown, with temperature as the horizontal axis and building carbon emission density as the vertical axis, the carbon emission density trend line (i.e., carbon emission density trend line) is fitted piecewise by the least squares method to establish a building carbon emission characteristic parameter model (i.e., carbon emission characteristic parameter model) with a time scale of month or year.
[0035] In an optional embodiment, the carbon emission density data includes a baseline carbon emission density, a cooling carbon emission density, and a heating carbon emission density, and fitting is performed based on the carbon emission density data corresponding to multiple reference buildings to determine the reference carbon emission characteristics, including: fitting according to the baseline carbon emission density, cooling carbon emission density, and heating carbon emission density corresponding to the multiple reference buildings to obtain a fitting curve characterizing different room temperatures and carbon emission densities; determining the reference carbon emission characteristics based on the segmented fitting trend and turning points in the fitting curve.
[0036] It can be understood that the carbon emission density data includes the baseline carbon emission density, cooling carbon emission density, and heating carbon emission density. The carbon emission densities corresponding to multiple reference buildings are fitted to obtain a fitting curve that characterizes the relationship between different room temperatures and carbon emission densities, identify the segmented fitting trend and turning points in the fitting curve, and determine the reference carbon emission characteristics. By analyzing the fitting trend of carbon emissions at different room temperatures, it is helpful to identify the peak and inefficiency points of energy use, thereby providing more accurate guidance for building energy management.
[0037] Optionally, there can be multiple methods for calculating the above-mentioned baseline carbon emission density, cooling carbon emission density, and heating carbon emission density. For example, based on the processed historical energy consumption data, the carbon emission factor is input to calculate the building's daily baseline carbon emission density (it can also be other predetermined periods, such as weekly, monthly), cooling carbon emission density, and heating carbon emission density.
[0038]
[0039] Among them, y i is the carbon emission density, expressed in kilograms of carbon dioxide per square meter per day, kgCO2 / (m 2 ·day), which is the carbon emission density; i is the carbon emission density category; φ is the identifier of the energy type, including electricity and gas; l represents the total number of energy types; u φ is the energy consumption density, in kWh / (m 2 day), v φ It is the dynamic carbon emission factor, and its unit is kilogram of carbon dioxide per kilojoule (kgCO2 / kJ).
[0040] Optionally, the above-mentioned reference carbon emission characteristics can be multiple, for example, according to the segmented fitting trend line and its intersection, the building subsystem carbon emission characteristic parameters (i.e., carbon emission characteristic parameters) are extracted, including the benchmark carbon emission density b0 (kgCO2 / m 2), carbon exhaust hot start point h1 (degrees Celsius, ℃), carbon exhaust cold start point c1 (℃), carbon exhaust cold sensitivity c0, carbon exhaust hot sensitivity h0.
[0041] In an optional embodiment, the method also includes: obtaining initial energy consumption data of a reference building, wherein the initial energy consumption data includes: cooling data, heating data, electricity data, and gas data; when there are differences in sampling time lengths among the cooling data, heating data, electricity data, and gas data, data alignment processing is performed on the cooling data, heating data, electricity data, and gas data to obtain reference energy consumption data.
[0042] It can be understood that the initial energy consumption data of the reference building is obtained. The initial energy consumption data includes: cooling data, heating data, electricity data, and gas data. If there is a difference in sampling time between the data, the data is aligned to obtain the reference energy consumption data. Through data alignment, the deviation introduced by different sampling frequencies or cycles can be eliminated, ensuring that when analyzing and comparing different types of energy consumption data, all data are based on the same time base, enhancing data consistency and comparability.
[0043] Optionally, there can be multiple data alignment methods. For example, there are differences in the sampling times of cold, hot, electricity, and gas data (i.e., cooling data, heating data, electricity consumption data, and gas consumption data). The data needs to be aligned to keep them consistent in time, and the difference or resampling method is used to achieve data alignment.
[0044] Optionally, there can be multiple data preprocessing methods, for example, cleaning the raw data of building energy subsystems such as cooling, heating, electricity, and gas collected in the database, removing possible outliers, noise, and interference, and using Kalman filtering to smooth the data and remove noise.
[0045] Step S104, based on the reference carbon emission characteristics, determining a plurality of carbon emission problem tags existing in the reference building, wherein the plurality of carbon emission problem tags represent the operating status of the corresponding subsystems;
[0046] It can be understood that multiple carbon emission problem labels existing in the reference building are determined based on the reference carbon emission characteristics. The multiple carbon emission problem labels represent the operating status of the corresponding subsystems. By analyzing the operating status of the subsystems, refined energy management is achieved.
[0047] In an optional embodiment, based on a reference carbon emission characteristic, multiple carbon emission problem labels existing in a reference building are determined, including: in the reference carbon emission characteristic, sub-feature parameters corresponding to multiple sub-reference systems included in the reference building are determined; score values of the sub-feature parameters corresponding to the multiple sub-reference systems are determined; among the sub-feature parameters corresponding to the multiple sub-reference systems, sub-feature parameters having score values less than a predetermined score threshold are determined as problem labels; and multiple carbon emission problem labels are determined based on the problem labels corresponding to the multiple sub-reference systems.
[0048] It can be understood that the scoring values of the sub-feature parameters corresponding to the multiple sub-reference systems included in the reference building are determined, and the sub-feature parameters with scoring values less than the predetermined scoring threshold are determined as problem labels. Based on the problem labels corresponding to the multiple sub-reference systems, multiple carbon emission problem labels are determined. By referring to the carbon emission characteristics and determining multiple carbon emission problem labels, the performance of the sub-reference system in carbon emissions can be accurately identified, thereby achieving refined diagnosis of carbon emissions.
[0049] Optionally, there may be multiple methods for extracting the carbon emission characteristic parameters of the above-mentioned subsystems (i.e., the sub-characteristic parameters corresponding to the sub-reference systems respectively). For example, by extracting the carbon emission characteristic parameters of the building energy subsystem, ranking and evaluating each parameter, setting a threshold, and identifying the poor characteristic parameters of each building, problems existing in the operation of each subsystem of the building can be identified based on the characteristic parameter labels.
[0050] Optionally, there may be multiple methods for determining the problem labels. For example, by determining the characteristic parameters of each building subsystem (i.e., the sub-characteristic parameters corresponding to the sub-reference systems respectively) and the operating data of the subsystem (i.e., the sub-reference system), a correspondence between the subsystem characteristic parameters and the existing problems is established, and a problem label is assigned to each characteristic parameter. For example, the label corresponding to the baseline carbon emission density is: the lighting and socket system load is high, and the lighting and socket system works too long; the label corresponding to the carbon emission thermal sensitivity is: the heating system efficiency is low, the heating system water supply temperature is too high, and the heating system works too long; the label corresponding to the carbon emission cold sensitivity is: the air-conditioning system has low cooling efficiency, and the air-conditioning system works too long; the label corresponding to the carbon emission cold start point or hot start point is: the building maintenance structure has poor thermal insulation performance, etc.
[0051] Step S106, training the initial model based on multiple carbon emission problem labels and reference energy consumption data to obtain a target model;
[0052] It can be understood that the initial model is trained using multiple carbon emission problem labels and reference energy consumption data to obtain a target model. By training the initial model, the accuracy and generalization ability of the target model in carbon emission analysis can be improved.
[0053] Optionally, the above initial model can be multiple, for example, random forest, neural network, unsupervised learning, etc. Random forest is an integrated learning method that makes predictions by building multiple decision trees. For the identification of carbon emission problems, random forest can process a large amount of data and features, and has good generalization ability, and can effectively identify the patterns and laws of carbon emissions. Neural network is a computational model that simulates the interconnection between neurons, and learns and predicts through the combination of multiple layers of neurons. For the identification of carbon emission problems, neural networks can build models by learning a large amount of data, and can discover complex patterns and associations in the data, thereby improving the accuracy of carbon emission identification. Unsupervised learning is a learning method that does not require labeled data, and learns by discovering hidden patterns and structures in the data. For the identification of carbon emission problems, unsupervised learning can help discover potential laws and clusters in the data, thereby identifying patterns and trends in carbon emissions, which is helpful for the analysis and prediction of carbon emission problems.
[0054] In an optional embodiment, the target model is a target decision tree, and the initial model is trained based on multiple carbon emission problem labels and reference energy consumption data to obtain the target model, including: determining the information gain of the nodes included in the initial decision tree based on the reference energy consumption data; determining predetermined characteristic conditions based on multiple carbon emission problem labels; dividing the initial decision tree into nodes according to the information gain and the predetermined characteristic conditions to determine the target decision tree.
[0055] It can be understood that the target model is determined to be a target decision tree. Based on the reference energy consumption data and multiple carbon emission problem labels, the information gain and predetermined characteristic conditions of the nodes in the initial decision tree are determined. Based on the information gain and the predetermined characteristic conditions, the initial decision tree is node-divided to determine the target decision tree. By combining information gain with characteristic conditions, the target decision tree model can automatically optimize its structure, accurately locate and reduce carbon emissions.
[0056] Optionally, the above method for constructing a decision tree at least includes: selecting a feature with the largest information gain as a partition feature of the current node according to a calculation formula of information gain, and further constructing a decision tree.
[0057] In an optional embodiment, determining the information gain of the nodes included in the initial decision tree based on the reference energy usage data includes: determining the information entropy and conditional entropy of the reference energy usage data; and determining the information gain based on the information entropy and conditional entropy.
[0058] It can be understood that the information entropy and conditional entropy of the reference energy consumption data are determined to determine the information gain. By calculating the information gain, the features that have the greatest impact on the classification results can be efficiently screened out, which helps to build a more accurate model.
[0059] Optionally, there can be multiple methods for calculating the above-mentioned information gain. For example, the total electricity and gas energy consumption data of each building in the database (i.e., reference energy consumption data) and building operation problems (i.e., multiple carbon emission problem labels) are used as input to carry out decision tree model training, calculate the impurity index information entropy, and select the best features for node splitting.
[0060] The mathematical expression for calculating information entropy is:
[0061] Entropy(D)=-∑(p(i)*log2(p(i)))
[0062] Among them, Entropy(D) represents the information entropy of the data set D (i.e., reference energy consumption data), p(i) represents the proportion of samples of the i-th category in D, the type of sample can be the data of subsystems with different energy demands, and i can be understood as the carbon emission density category.
[0063] The mathematical expression of conditional entropy is:
[0064] Entropy(D|A)=∑(|Dv| / |D|)*Entropy(Dv)
[0065] Among them, Entropy(D|A) represents the conditional entropy of the data set D under the given condition of feature A, |Dv| represents the number of samples in D when the value of feature A is v (predetermined value), and feature A can be a feature condition obtained based on the carbon emission feature.
[0066] Information gain = entropy of the entire data set - weighted average of the entropy of the divided subsets
[0067] The mathematical expression of information gain is:
[0068] Gain(D,A)=Entropy(D)-Entropy(D|A)
[0069] Among them, Gain(D,A) represents the information gain obtained by partitioning the data set D using feature A. According to the calculation formula of information gain, the feature with the largest information gain can be selected as the partition feature of the current node to further construct a decision tree.
[0070] Optionally, there may be multiple methods for diagnosing and analyzing the new building (i.e., the target building). For example, after the decision tree is constructed, the decision tree may be used to diagnose and analyze the new building. During diagnosis, the sample (energy consumption data of the target building) is traversed downward step by step from the root node according to the characteristic conditions until a leaf node is reached. The category corresponding to the leaf node is the diagnosis result.
[0071]
[0072] Where y′ is the diagnostic problem category; f(x) is the prediction function of the decision tree model, x represents the energy consumption data; N is the number of leaf nodes; w j is the weight of the jth node; R j is the region of the jth leaf node; I(x∈R j ) is the indicator function; argmax c Represents the category c obtained by maximizing the calculation results of each node.
[0073] Step S108, inputting the target energy consumption data of the target building into the target model for processing to determine the carbon emission problem identification result in the target building.
[0074] It can be understood that by inputting the target energy consumption data of the target building into the target model for processing, the carbon emission problem identification result in the target building can be obtained. The energy consumption data of the target building is input into the target model for processing to identify the carbon emission problem, thus realizing the refined energy-saving and carbon-reduction diagnosis of the building.
[0075] Through the above step S102, based on the reference energy consumption data of the reference building, the reference carbon emission characteristics are determined, wherein the reference building includes multiple subsystems with different energy consumption requirements; step S104, based on the reference carbon emission characteristics, multiple carbon emission problem labels existing in the reference building are determined, wherein multiple carbon emission problem labels represent the operating status of the corresponding subsystems; step S106, based on multiple carbon emission problem labels and reference energy consumption data, the initial model is trained to obtain the target model; step S108, the target energy consumption data of the target building is input into the target model for processing to determine the carbon emission problem identification result in the target building. The purpose of carrying out refined energy-saving and carbon-reduction diagnosis for buildings can be achieved, and the technical effect of improving the efficiency and accuracy of building energy-saving diagnosis and reducing hysteresis is achieved, thereby solving the technical problem of unsatisfactory efficiency of building energy-saving diagnosis.
[0076] Based on the above embodiments and optional embodiments, the present application proposes an optional implementation manner. Figure 3 is a second schematic flow chart of an optional method for determining building carbon emission problems provided in an embodiment of the present application, such as Figure 3 shown.
[0077] Step S1, preprocessing the collected historical data to convert the energy consumption data into carbon emission data.
[0078] The historical data of cooling, heating and electricity collected from the building energy subsystem are processed by data cleaning, data alignment, normalization and data partitioning to build a building operation energy database. The raw data of cooling, heating, electricity, gas and other building energy subsystems collected in the database are cleaned to remove possible outliers, noise and interference, and Kalman filtering is used to smooth the data and remove noise.
[0079] There are differences in the sampling time of cooling, heating, electricity, and gas data (i.e., cooling data, heating data, electricity consumption data, and gas consumption data). The data needs to be aligned to keep them consistent in time. The difference or resampling method is used to achieve data alignment.
[0080] Step S2, extracting carbon emission characteristic parameters of building subsystems through correlation analysis between carbon emission data and temperature data based on historical building energy consumption data in the database;
[0081] With temperature as the horizontal axis and building carbon emission density as the vertical axis, the carbon emission density trend line (i.e., carbon emission density trend line) is fitted piecewise using the least squares method, and a building carbon emission characteristic parameter model (i.e., carbon emission characteristic parameter model) with a time scale of month or year is established.
[0082] According to the segmented fitting trend line and its intersection, the carbon emission characteristic parameters of the building subsystem (i.e., carbon emission characteristic parameters) are extracted, including the baseline carbon emission density b0 (kgCO2 / m2), carbon emission hot start point h1 (℃), carbon emission cold start point c1 (℃), carbon emission cold sensitivity c0, and carbon emission hot sensitivity h0.
[0083] Step S3, manually analyzing the characteristic parameters of each building subsystem and the subsystem operation data, assigning subsystem operation problem labels that match the characteristic parameters of the building subsystem, and identifying the problems existing in the building;
[0084] By manually analyzing the characteristic parameters of each building subsystem (i.e. the sub-characteristic parameters corresponding to the sub-reference system respectively) and the operating data of the subsystem (i.e. the sub-reference system), the corresponding relationship between the subsystem characteristic parameters and the existing problems is established, and each characteristic parameter is assigned a problem label. For example, the corresponding label for the baseline carbon emission density is: the lighting and socket system load is high, and the lighting and socket system works too long; the corresponding label for the carbon emission thermal sensitivity is: the heating system efficiency is low, the heating system water supply temperature is too high, and the heating system works too long; the corresponding label for the carbon emission cold sensitivity is: the air-conditioning system has low cooling efficiency and the air-conditioning system works too long; the corresponding label for the carbon emission cold start point or hot start point is: the thermal insulation performance of the building maintenance structure is poor, etc.
[0085] By extracting the carbon emission characteristic parameters of the building energy subsystem, ranking and evaluating each parameter, setting thresholds, and identifying the poor characteristic parameters of each building, the problems existing in the operation of each subsystem of the building can be identified based on the characteristic parameter labels.
[0086] Step S4, obtaining the total amount of electricity and gas data of each building in the database, as well as the problems of each building, and using the subsystem operation problems and the total energy consumption data of the building to train a decision tree model (i.e., an initial model);
[0087] According to the calculation formula of information gain, the feature with the largest information gain is selected as the partition feature of the current node to further construct the decision tree.
[0088] The total electricity and gas energy consumption data of each building in the database (i.e., reference energy consumption data) and building operation problems (i.e., multiple carbon emission problem labels) are used as input to carry out decision tree model training, calculate the information entropy of the impurity index, and select the best features for node splitting.
[0089] The mathematical expression for calculating information entropy is:
[0090] Entropy(D)=-∑(p(i)*log2(p(i)))
[0091] Among them, Entropy(D) represents the information entropy of the data set D (i.e., reference energy consumption data), p(i) represents the proportion of the i-th type of samples in D, and the type of samples can be data of subsystems with different energy consumption requirements.
[0092] The mathematical expression of conditional entropy is:
[0093] Entropy(D|A)=∑(|Dv| / |D|)*Entropy(Dv)
[0094] Among them, Entropy(D|A) represents the conditional entropy of the data set D under the given condition of feature A, |Dv| represents the number of samples in D when the value of feature A is v, and feature A can be a feature condition obtained based on the carbon emission feature.
[0095] Information gain = entropy of the entire data set - weighted average of the entropy of the divided subsets
[0096] The mathematical expression of information gain is:
[0097] Gain(D,A)=Entropy(D)-Entropy(D|A)
[0098] Among them, Gain(D,A) represents the information gain obtained by partitioning the data set D using feature A. According to the calculation formula of information gain, the feature with the largest information gain can be selected as the partition feature of the current node to further construct a decision tree.
[0099] After the decision tree is built, the decision tree (i.e., the target model) can be used to diagnose and analyze the new building. During diagnosis, the sample (energy consumption data of the target building) starts from the root node and traverses downward step by step according to the characteristic conditions until it reaches the leaf node. The category corresponding to the leaf node is the diagnosis result.
[0100]
[0101] Where y′ is the diagnostic problem category; f(x) is the prediction function of the decision tree model, x represents the energy consumption data; N is the number of leaf nodes; w j is the weight of the jth node; R j is the region of the jth leaf node; I(x∈R j ) is the indicator function; argmax c It means taking the category c that maximizes c.
[0102] Step S5, input the total energy consumption data of buildings outside the database. For new buildings that only have total electricity and gas consumption, input their monthly, quarterly, and annual total electricity and gas consumption data, and diagnose the existing problems through the decision tree model.
[0103] The above optional implementation methods can achieve at least the following effects: low diagnostic requirements. For buildings that lack subsystem operation data, the energy consumption and carbon emission problem diagnosis at the building subsystem level can be realized based on the total electricity and gas data, which is more refined; the result accuracy is high. By coupling the building characteristic parameters through the artificial intelligence algorithm (decision tree) model, data + mechanism drive, the diagnosis and analysis of only the total electricity and gas consumption can be realized, and the problems in the building energy and carbon operation can be identified. Compared with traditional manual analysis, the accuracy of building energy and carbon diagnosis can be greatly improved; the time granularity is fine. The building can obtain the building subsystem characteristic parameters and existing problems through monthly and quarterly data to realize decision tree model training, so the time scale of new building operation diagnosis can be increased from annual to monthly or quarterly.
[0104] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0105] In this embodiment, a device for determining building carbon emission problems is also provided, which is used to implement the above embodiments and preferred implementations, and will not be repeated here. As used below, the terms "module" and "device" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0106] According to an embodiment of the present application, a device embodiment for implementing a method for determining a building carbon emission problem is also provided. Figure 4 is a schematic diagram of an optional building carbon emission problem determination device provided according to an embodiment of the present application, such as Figure 4 As shown, the above-mentioned building carbon emission problem determination device includes a feature determination module 402, a label determination module 404, a model acquisition module 406, and a result determination module 408. The device is described below.
[0107] A characteristic determination module 402 is used to determine a reference carbon emission characteristic based on reference energy consumption data of a reference building, wherein the reference building includes a plurality of subsystems with different energy consumption requirements;
[0108] The tag determination module 404 is connected to the feature determination module 402 and is used to determine a plurality of carbon emission problem tags existing in the reference building based on the reference carbon emission feature, wherein the plurality of carbon emission problem tags represent the operating status of the corresponding subsystem;
[0109] The model acquisition module 406 is connected to the label determination module 404 and is used to train the initial model based on multiple carbon emission problem labels and reference energy consumption data to obtain a target model;
[0110] The result determination module 408 is connected to the model acquisition module 406 and is used to input the target energy consumption data of the target building into the target model for processing to determine the carbon emission problem identification result in the target building.
[0111] In a device for determining a building carbon emission problem provided by an embodiment of the present application, a feature determination module 402 is set to determine a reference carbon emission feature based on reference energy consumption data of a reference building, wherein the reference building includes multiple subsystems with different energy consumption requirements; a label determination module 404 is connected to the feature determination module 402, and is used to determine multiple carbon emission problem labels existing in the reference building based on the reference carbon emission feature, wherein multiple carbon emission problem labels represent the operating status of the corresponding subsystems; a model acquisition module 406 is connected to the label determination module 404, and is used to train the initial model based on multiple carbon emission problem labels and reference energy consumption data to obtain a target model; a result determination module 408 is connected to the model acquisition module 406, and is used to input the target energy consumption data of the target building into the target model for processing, and determine the carbon emission problem identification result in the target building. The purpose of carrying out refined energy-saving and carbon-reduction diagnosis for buildings is achieved, and the technical effect of improving the efficiency and accuracy of building energy-saving diagnosis and reducing hysteresis is achieved, thereby solving the technical problem of unsatisfactory efficiency of building energy-saving diagnosis.
[0112] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.
[0113] It should be noted that the above-mentioned feature determination module 402, label determination module 404, model acquisition module 406, and result determination module 408 correspond to steps S102 to S108 in the embodiment, and the examples and application scenarios implemented by the above-mentioned modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment. It should be noted that the above-mentioned modules can be run in a computer terminal as part of the device.
[0114] It should be noted that the optional or preferred implementation of this embodiment can refer to the relevant description in the embodiment, which will not be repeated here.
[0115] The above-mentioned building carbon emission problem determination device may also include a processor and a memory, and the feature determination module 402, the label determination module 404, the model acquisition module 406, the result determination module 408, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize the corresponding functions.
[0116] The processor includes a kernel, which retrieves the corresponding program unit from the memory. There can be one or more kernels. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one storage chip.
[0117] An embodiment of the present application provides a non-volatile storage medium having a program stored thereon, which implements a method for determining a building carbon emission problem when the program is executed by a processor.
[0118] The embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: based on the reference energy consumption data of a reference building, a reference carbon emission feature is determined, wherein the reference building includes multiple subsystems with different energy consumption requirements; based on the reference carbon emission feature, multiple carbon emission problem labels existing in the reference building are determined, wherein the multiple carbon emission problem labels represent the operating status of the corresponding subsystems; based on the multiple carbon emission problem labels and the reference energy consumption data, an initial model is trained to obtain a target model; the target energy consumption data of the target building is input into the target model for processing to determine the carbon emission problem identification result in the target building. The device in this article can be a server, a PC, etc.
[0119] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: determining a reference carbon emission characteristic based on reference energy consumption data of a reference building, wherein the reference building includes multiple subsystems with different energy consumption demands; determining multiple carbon emission problem labels existing in the reference building based on the reference carbon emission characteristic, wherein the multiple carbon emission problem labels represent the operating status of the corresponding subsystems; training an initial model based on the multiple carbon emission problem labels and the reference energy consumption data to obtain a target model; inputting the target energy consumption data of the target building into the target model for processing to determine the carbon emission problem identification result in the target building.
[0120] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0121] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0122] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0124] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0125] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0127] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0128] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0129] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.
Claims
1. A method for determining building carbon emission problems, characterized in that: include: Determining a reference carbon emission characteristic based on reference energy consumption data of a reference building, wherein the reference building includes a plurality of subsystems with different energy consumption demands; Based on the reference carbon emission characteristics, determining a plurality of carbon emission problem tags existing in the reference building, wherein the plurality of carbon emission problem tags represent the operating status of corresponding subsystems; Based on the multiple carbon emission problem labels and the reference energy consumption data, an initial model is trained to obtain a target model; The target energy consumption data of the target building is input into the target model for processing to determine the carbon emission problem identification result in the target building.
2. The method according to claim 1, characterized in that The reference carbon emission characteristics are determined based on the reference energy consumption data of the reference building, including: Determining carbon emission density data of the reference building based on the reference energy consumption data and a predetermined carbon emission factor; In the case where there are multiple reference buildings, the reference carbon emission characteristics are determined by performing fitting based on the carbon emission density data corresponding to the multiple reference buildings.
3. The method according to claim 2, characterized in that The carbon emission density data includes a baseline carbon emission density, a cooling carbon emission density, and a heating carbon emission density. The fitting based on the carbon emission density data corresponding to the multiple reference buildings to determine the reference carbon emission characteristics includes: Fitting is performed based on the benchmark carbon emission densities, the cooling carbon emission density, and the heating carbon emission density respectively corresponding to the multiple reference buildings to obtain a fitting curve representing different room temperatures and carbon emission densities; The reference carbon emission characteristics are determined based on the segmented fitting trend and turning points in the fitting curve.
4. The method according to claim 1, characterized in that: The step of determining a plurality of carbon emission problem labels existing in the reference building based on the reference carbon emission characteristics includes: In the reference carbon emission characteristics, sub-characteristic parameters corresponding to a plurality of sub-reference systems included in the reference building are determined; Determine the score values of the sub-feature parameters respectively corresponding to the multiple sub-reference systems; Among the sub-feature parameters respectively corresponding to the multiple sub-reference systems, determining a sub-feature parameter having a score value less than a predetermined score threshold as a question label; The multiple carbon emission problem labels are determined based on the problem labels respectively corresponding to the multiple sub-reference systems.
5. The method according to claim 1, characterized in that: The target model is a target decision tree. Based on the multiple carbon emission problem labels and the reference energy consumption data, the initial model is trained to obtain the target model, including: Based on the reference energy consumption data, determining the information gain of the nodes included in the initial decision tree; Determining predetermined characteristic conditions according to the multiple carbon emission problem labels; According to the information gain and the predetermined characteristic condition, the initial decision tree is divided into nodes to determine the target decision tree.
6. The method according to claim 5, characterized in that The determining, based on the reference energy consumption data, information gains of nodes in the initial decision tree comprises: Determining information entropy and conditional entropy of the reference energy consumption data; The information gain is determined based on the information entropy and the conditional entropy.
7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: Acquire initial energy consumption data of the reference building, wherein the initial energy consumption data includes: cooling data, heating data, electricity consumption data, and gas consumption data; When there are differences in sampling durations among the cooling data, the heating data, the electricity consumption data, and the gas consumption data, data alignment is performed on the cooling data, the heating data, the electricity consumption data, and the gas consumption data to obtain the reference energy consumption data.
8. A device for determining building carbon emission problems, characterized in that: include: A feature determination module, configured to determine a reference carbon emission feature based on reference energy consumption data of a reference building, wherein the reference building includes a plurality of subsystems with different energy consumption requirements; A tag determination module, configured to determine a plurality of carbon emission problem tags existing in the reference building based on the reference carbon emission characteristics, wherein the plurality of carbon emission problem tags represent the operating status of corresponding subsystems; A model acquisition module is used to train an initial model based on the multiple carbon emission problem labels and the reference energy consumption data to obtain a target model; The result determination module is used to input the target energy consumption data of the target building into the target model for processing, and determine the carbon emission problem identification result in the target building.
9. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the building carbon emission problem determination method as described in any one of claims 1 to 7.
10. An electronic device, characterized in that: include: One or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the building carbon emission problem determination method as described in any one of claims 1 to 7.