Energy consumption analysis method for ultra-low energy consumption building

Abstract

The invention discloses an energy consumption analysis method for an ultra-low energy consumption building, and belongs to the technical field of ultra-low energy consumption buildings, and the method comprises the following steps: S1, collecting and processing the historical data of the ultra-low energy consumption building, and determining the feature data of the ultra-low energy consumption building; s2, performing test optimization on the energy consumption prediction model of the ultra-low energy consumption building, and determining an optimal energy consumption prediction model of the ultra-low energy consumption building; s3, performing energy consumption prediction on the ultra-low energy consumption building, and determining an energy consumption prediction value of the ultra-low energy consumption building; and S4, analyzing the energy consumption predicted value of the ultra-low energy consumption building, and determining an energy consumption analysis result of the ultra-low energy consumption building. The problem that the use effect of the ultra-low energy consumption building is poor due to the fact that effective energy consumption analysis and timely energy consumption management cannot be carried out on the ultra-low energy consumption building in the prior art is solved. According to the invention, effective energy consumption analysis can be carried out on the ultra-low energy consumption building, timely energy consumption management can be carried out on the ultra-low energy consumption building, and the use effect of the ultra-low energy consumption building can be improved.

Description

Technical Field

[0001] The invention relates to the technical field of ultra-low energy consumption buildings, and in particular to an energy consumption analysis method for ultra-low energy consumption buildings. Background Art

[0002] As global climate change and energy crisis intensify, people pay more and more attention to energy efficiency in building design; ultra-low energy consumption buildings are a new type of building concept that aims to achieve efficient energy utilization and reduce environmental impact by adopting advanced energy-saving technologies and design strategies. However, to achieve these goals, it is necessary to analyze and manage the energy consumption of buildings.

[0003] The Chinese patent with publication number CN110708824A discloses an ultra-low energy consumption building system. Through the intelligent dimmer, the brightness of corridor lighting, user lighting and public facility lighting can be adjusted according to different lighting conditions, so as to avoid the increase of power consumption caused by the high power consumption of lighting during the day. Through the power consumption collection module, the power consumption generated by lighting can be quickly collected, and the power consumption can be calculated by the power consumption calculation module to control the lighting of different areas; however, the patent has the following defects:

[0004] The existing technology cannot effectively analyze the energy consumption of ultra-low energy consumption buildings, and cannot timely manage the energy consumption of ultra-low energy consumption buildings, resulting in poor use of ultra-low energy consumption buildings. Summary of the invention

[0005] The purpose of the present invention is to provide an energy consumption analysis method for ultra-low energy consumption buildings, which can effectively analyze the energy consumption of ultra-low energy consumption buildings and timely manage the energy consumption of ultra-low energy consumption buildings, thereby improving the use effect of ultra-low energy consumption buildings and solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An ultra-low energy consumption building energy consumption analysis method comprises the following steps:

[0008] S1: According to the energy consumption analysis requirements of ultra-low energy consumption buildings, collect historical data of ultra-low energy consumption buildings, process the historical data of ultra-low energy consumption buildings, and determine the characteristic data of ultra-low energy consumption buildings;

[0009] S2: Construct an energy consumption prediction model for ultra-low energy consumption buildings, test and optimize the energy consumption prediction model for ultra-low energy consumption buildings, and determine the optimal energy consumption prediction model for ultra-low energy consumption buildings;

[0010] S3: Based on the optimal ultra-low energy consumption prediction model, the energy consumption of ultra-low energy consumption buildings is predicted, and the predicted energy consumption value of ultra-low energy consumption buildings is determined;

[0011] S4: Based on the energy consumption standards for ultra-low energy buildings, analyze the predicted energy consumption values ​​of ultra-low energy buildings and determine the energy consumption analysis results of ultra-low energy buildings.

[0012] Preferably, in S1, historical data of ultra-low energy consumption buildings is collected, and the following operations are performed:

[0013] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the area shape, geographical location, orientation, enclosure materials, window configuration and foundation configuration of ultra-low energy consumption buildings are collected to obtain basic data of ultra-low energy consumption buildings;

[0014] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the temperature, humidity, light and wind speed of the environment in which the ultra-low energy consumption buildings are located are collected to obtain the environmental data of ultra-low energy consumption buildings;

[0015] Among them, the historical data of ultra-low energy consumption buildings are determined based on the basic data of ultra-low energy consumption buildings and the environmental data of ultra-low energy consumption buildings.

[0016] Preferably, in S1, the historical data of ultra-low energy consumption buildings is processed to perform the following operations:

[0017] Obtain historical data on ultra-low energy buildings;

[0018] Clean the historical data of ultra-low energy buildings, including:

[0019] Conduct consistency checks on historical data of ultra-low energy buildings;

[0020] According to the data consistency requirements, check whether there are inconsistent data in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the inconsistent data in the historical data of ultra-low energy consumption buildings;

[0021] Process invalid and missing values ​​in historical data of ultra-low energy consumption buildings;

[0022] According to the requirements of data validity and completeness, check whether there are invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings;

[0023] Identify historical data of ultra-low energy buildings that are useful for energy analysis of ultra-low energy buildings.

[0024] Preferably, in S1, the historical data of ultra-low energy consumption buildings is processed, and the following operations are further performed:

[0025] Obtain historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings;

[0026] Convert and process the historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings;

[0027] Eliminate the dimensional differences between historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, unify the formats of historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, and determine standardized historical data of ultra-low energy consumption buildings;

[0028] Extract features from standardized historical data of ultra-low energy buildings;

[0029] Extract features that can reflect the energy consumption analysis of ultra-low energy buildings;

[0030] Determine the characteristic data of ultra-low energy consumption buildings.

[0031] Preferably, in S2, an ultra-low energy consumption building energy consumption prediction model is constructed, and the following operations are performed:

[0032] Obtaining ultra-low energy consumption building characteristic data;

[0033] Divide the characteristic data of ultra-low energy consumption buildings to determine the training set and test set;

[0034] Select a convolutional neural network model framework suitable for energy consumption prediction of ultra-low energy buildings;

[0035] Based on the training set, the selected convolutional neural network model framework suitable for ultra-low energy consumption building energy consumption prediction is trained to determine the ultra-low energy consumption building energy consumption prediction model.

[0036] Preferably, the ultra-low energy consumption building characteristic data is divided to determine a training set and a test set, including:

[0037] Extracting the amount of invalid values ​​in the historical data of the ultra-low energy consumption building;

[0038] Extracting the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building;

[0039] Obtaining an outlier degree coefficient by using the amount of invalid values ​​in the historical data of ultra-low energy consumption buildings combined with the number of missing data corresponding to the missing values ​​in the historical data of ultra-low energy consumption buildings;

[0040] The outlier degree coefficient is obtained by the following formula:

[0041]

[0042] Among them, E represents the outlier degree coefficient; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C zRepresents the total amount of historical data of ultra-low energy consumption buildings; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N z The total number of data representing the historical data of ultra-low energy buildings;

[0043] Comparing the outlier degree coefficient with a preset coefficient threshold;

[0044] Determine the target ratios of the training set and the test set according to the comparison result between the outlier degree coefficient and the preset coefficient threshold;

[0045] The ultra-low energy consumption building characteristic data are divided according to the target ratios corresponding to the training set and the test set to determine the training set and the test set.

[0046] Preferably, determining the corresponding ratio of the training set and the test set according to the comparison result between the outlier degree coefficient and the preset coefficient threshold comprises:

[0047] Retrieving a comparison result between the abnormal value degree coefficient and a preset coefficient threshold;

[0048] When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is lower than the preset coefficient threshold value, the preset initial ratio value is retrieved as the target ratio;

[0049] When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is not lower than the preset coefficient threshold value, the preset initial ratio value is retrieved;

[0050] The target ratio is obtained by using the abnormal value degree coefficient and the preset initial ratio value, wherein the target ratio is obtained by the following formula:

[0051]

[0052] Where R represents the target ratio; R0 represents the preset initial ratio value; E represents the outlier degree coefficient; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N c represents the number of data corresponding to invalid values ​​in the historical data of the ultra-low energy consumption building; T represents the preset coefficient threshold; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C z Represents the total amount of historical data of ultra-low energy consumption buildings; N z represents the total number of historical data of ultra-low energy consumption buildings; x represents the adjustment coefficient, and the value range of the adjustment coefficient is 0.21-1.88.

[0053] Preferably, in S2, the ultra-low energy consumption building energy consumption prediction model is tested and optimized, and the following operations are performed:

[0054] Obtain energy consumption prediction models for ultra-low energy buildings;

[0055] Based on the test set, the performance of the ultra-low energy consumption building energy consumption prediction model is tested to determine whether the performance of the ultra-low energy consumption building energy consumption prediction model can achieve the expected effect;

[0056] Determine the performance test results based on the ultra-low energy building energy consumption prediction model;

[0057] Based on the performance test results, the energy consumption prediction model of ultra-low energy consumption buildings is mined and analyzed, the parameter configuration of the energy consumption prediction model of ultra-low energy consumption buildings is adjusted, the structural setting of the energy consumption prediction model of ultra-low energy consumption buildings is optimized, the energy consumption prediction model of ultra-low energy consumption buildings is repeatedly iterated and optimized, and the optimal energy consumption prediction model of ultra-low energy consumption buildings is determined.

[0058] Preferably, in S3, energy consumption prediction is performed on ultra-low energy consumption buildings by performing the following operations:

[0059] Collect real-time data of ultra-low energy consumption buildings according to the energy consumption analysis requirements of ultra-low energy consumption buildings;

[0060] Input the real-time data of ultra-low energy consumption buildings into the optimal ultra-low energy consumption building energy consumption prediction model;

[0061] Based on the optimal ultra-low energy consumption prediction model, the real-time data of ultra-low energy consumption buildings is mined and analyzed to predict the energy consumption of ultra-low energy consumption buildings;

[0062] Determine the predicted energy consumption of ultra-low energy buildings.

[0063] Preferably, in S4, the predicted energy consumption value of the ultra-low energy consumption building is analyzed, and the following operations are performed:

[0064] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the energy consumption standards of ultra-low energy consumption buildings are set in advance;

[0065] Based on the energy consumption standards for ultra-low energy consumption buildings, the predicted energy consumption values ​​of ultra-low energy consumption buildings are analyzed to determine the energy consumption analysis results of ultra-low energy consumption buildings;

[0066] Among them, if the predicted value of ultra-low energy consumption building energy consumption is within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is that the ultra-low energy consumption building energy consumption is normal;

[0067] Among them, if the predicted value of ultra-low energy consumption building energy consumption is not within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is abnormal ultra-low energy consumption building energy consumption;

[0068] When the energy consumption of ultra-low energy consumption buildings is abnormal, early warning reminders will be issued to guide management personnel to take energy-saving measures in a timely manner, and the design and layout of ultra-low energy consumption buildings will be optimized to ensure normal energy consumption of ultra-low energy consumption buildings.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] 1. According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the present invention collects historical data of ultra-low energy consumption buildings, processes the historical data of ultra-low energy consumption buildings, determines characteristic data of ultra-low energy consumption buildings, divides the characteristic data of ultra-low energy consumption buildings, determines a training set and a test set, and based on the training set, trains a selected convolutional neural network model framework suitable for energy consumption prediction of ultra-low energy consumption buildings, constructs an energy consumption prediction model for ultra-low energy consumption buildings, tests and optimizes the energy consumption prediction model for ultra-low energy consumption buildings, and determines the optimal energy consumption prediction model for ultra-low energy consumption buildings.

[0071] 2. The present invention is based on the optimal ultra-low energy consumption prediction model for ultra-low energy consumption buildings, mines and analyzes the real-time data of ultra-low energy consumption buildings, predicts the energy consumption of ultra-low energy consumption buildings, determines the predicted energy consumption value of ultra-low energy consumption buildings, analyzes the predicted energy consumption value of ultra-low energy consumption buildings based on the ultra-low energy consumption standard of ultra-low energy consumption buildings, determines the energy consumption analysis result of ultra-low energy consumption buildings, and when the energy consumption of ultra-low energy consumption buildings is abnormal, early warning reminders for ultra-low energy consumption buildings are issued to guide management personnel to take energy-saving measures in time, optimize the design and layout of ultra-low energy consumption buildings to make the energy consumption of ultra-low energy consumption buildings normal, effectively analyze the energy consumption of ultra-low energy consumption buildings, and timely manage the energy consumption of ultra-low energy consumption buildings, which can improve the use effect of ultra-low energy consumption buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is an operation flow chart of the energy consumption analysis method of ultra-low energy consumption building of the present invention;

[0073] Figure 2 This is a flow chart of an algorithm for analyzing the predicted energy consumption value of an ultra-low energy consumption building based on the ultra-low energy consumption building energy consumption standard of the present invention. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] In order to solve the existing problem that ultra-low energy consumption buildings cannot be effectively analyzed and timely managed, resulting in poor use of ultra-low energy consumption buildings, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0076] An ultra-low energy consumption building energy consumption analysis method comprises the following steps:

[0077] S1: According to the energy consumption analysis requirements of ultra-low energy consumption buildings, collect historical data of ultra-low energy consumption buildings, process the historical data of ultra-low energy consumption buildings, and determine the characteristic data of ultra-low energy consumption buildings;

[0078] In this embodiment, historical data of ultra-low energy consumption buildings is collected, and the following operations are performed:

[0079] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the area shape, geographical location, orientation, enclosure materials, window configuration and foundation configuration of ultra-low energy consumption buildings are collected to obtain basic data of ultra-low energy consumption buildings;

[0080] It should be noted that the foundation refers to the soil or rock supporting the foundation under the building; the soil layer used as the building foundation is divided into rock, gravel soil, sand, silt, clay and artificial fill. There are two types of foundations: natural foundation and artificial foundation. The natural foundation is a natural soil layer that does not require human reinforcement, while the artificial foundation needs human reinforcement. Common ones include stone chip cushion layer, sand cushion layer, mixed lime soil backfill and compaction, etc.

[0081] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the temperature, humidity, light and wind speed of the environment in which the ultra-low energy consumption buildings are located are collected to obtain the environmental data of ultra-low energy consumption buildings;

[0082] Among them, the historical data of ultra-low energy consumption buildings are determined based on the basic data of ultra-low energy consumption buildings and the environmental data of ultra-low energy consumption buildings.

[0083] It should be noted that by collecting historical data on ultra-low energy consumption buildings, it is convenient to subsequently build an energy consumption prediction model for ultra-low energy consumption buildings, and further predict the energy consumption of ultra-low energy consumption buildings.

[0084] In this embodiment, the historical data of ultra-low energy consumption buildings is processed to perform the following operations:

[0085] Obtain historical data on ultra-low energy buildings;

[0086] Clean the historical data of ultra-low energy buildings, including:

[0087] Conduct consistency checks on historical data of ultra-low energy buildings;

[0088] According to the data consistency requirements, check whether there are inconsistent data in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the inconsistent data in the historical data of ultra-low energy consumption buildings;

[0089] Process invalid and missing values ​​in historical data of ultra-low energy consumption buildings;

[0090] According to the requirements of data validity and completeness, check whether there are invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings;

[0091] Identify historical data of ultra-low energy buildings that are useful for energy analysis of ultra-low energy buildings.

[0092] It should be noted that data cleaning refers to the process of preprocessing data in data governance work, with the aim of correcting or deleting erroneous, duplicated and inconsistent parts of the data to ensure the accuracy and completeness of the data. Through data cleaning, data can be converted into high-quality data to facilitate subsequent data analysis and use.

[0093] Among them, the consistency check is to check whether the data meets the requirements based on the reasonable value range and mutual relationship of each parameter, and to find data that exceeds the normal range, is logically unreasonable or contradictory. For example, a variable measured by a 1-7 scale has a value of 0, and a negative weight has a negative number, which should be regarded as exceeding the normal value range. Computer software such as SPSS, SAS and Excel can automatically identify each variable value that exceeds the range according to the defined value range. Logically inconsistent answers may appear in many forms: for example, many respondents said that they drove to work, but reported that they did not have a car; or the respondents reported that they were heavy buyers and users of a certain brand, but at the same time gave a very low score on the familiarity scale. When inconsistencies are found, the questionnaire number, record number, variable name, error category, etc. should be listed to facilitate further verification and correction.

[0094] Therefore, by cleaning the historical data of ultra-low energy consumption buildings, inconsistent data, invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings can be removed, and the historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings can be determined, which can improve the processing accuracy and efficiency of subsequent historical data of ultra-low energy consumption buildings.

[0095] In this embodiment, the historical data of ultra-low energy consumption buildings is processed, and the following operations are further performed:

[0096] Obtain historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings;

[0097] Convert and process the historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings;

[0098] Eliminate the dimensional differences between historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, unify the formats of historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, and determine standardized historical data of ultra-low energy consumption buildings;

[0099] Extract features from standardized historical data of ultra-low energy buildings;

[0100] Extract features that can reflect the energy consumption analysis of ultra-low energy buildings;

[0101] Determine the characteristic data of ultra-low energy consumption buildings.

[0102] It should be noted that by converting the historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, standardized historical data of ultra-low energy consumption buildings are determined, which is convenient for subsequent feature extraction of standardized historical data of ultra-low energy consumption buildings. By extracting features from standardized historical data of ultra-low energy consumption buildings, characteristic data of ultra-low energy consumption buildings are determined, which is convenient for subsequent construction of energy consumption prediction models for ultra-low energy consumption buildings.

[0103] S2: Construct an energy consumption prediction model for ultra-low energy consumption buildings, test and optimize the energy consumption prediction model for ultra-low energy consumption buildings, and determine the optimal energy consumption prediction model for ultra-low energy consumption buildings;

[0104] In this embodiment, an ultra-low energy consumption building energy consumption prediction model is constructed, and the following operations are performed:

[0105] Obtaining ultra-low energy consumption building characteristic data;

[0106] Divide the characteristic data of ultra-low energy consumption buildings to determine the training set and test set;

[0107] Select a convolutional neural network model framework suitable for energy consumption prediction of ultra-low energy buildings;

[0108] Based on the training set, the selected convolutional neural network model framework suitable for ultra-low energy consumption building energy consumption prediction is trained to determine the ultra-low energy consumption building energy consumption prediction model.

[0109] It should be noted that by constructing an energy consumption prediction model for ultra-low energy consumption buildings, it is convenient to conduct subsequent energy consumption prediction for ultra-low energy consumption buildings and determine the energy consumption prediction value of ultra-low energy consumption buildings.

[0110] In this embodiment, the ultra-low energy consumption building energy consumption prediction model is tested and optimized, and the following operations are performed:

[0111] Obtain energy consumption prediction models for ultra-low energy buildings;

[0112] Based on the test set, the performance of the ultra-low energy consumption building energy consumption prediction model is tested to determine whether the performance of the ultra-low energy consumption building energy consumption prediction model can achieve the expected effect;

[0113] Determine the performance test results based on the ultra-low energy building energy consumption prediction model;

[0114] Based on the performance test results, the energy consumption prediction model of ultra-low energy consumption buildings is mined and analyzed, the parameter configuration of the energy consumption prediction model of ultra-low energy consumption buildings is adjusted, the structural setting of the energy consumption prediction model of ultra-low energy consumption buildings is optimized, the energy consumption prediction model of ultra-low energy consumption buildings is repeatedly iterated and optimized, and the optimal energy consumption prediction model of ultra-low energy consumption buildings is determined.

[0115] Specifically, the characteristic data of ultra-low energy consumption buildings are divided to determine the training set and test set, including:

[0116] Extracting the amount of invalid values ​​in the historical data of the ultra-low energy consumption building;

[0117] Extracting the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building;

[0118] Obtaining an outlier degree coefficient by using the amount of invalid values ​​in the historical data of ultra-low energy consumption buildings combined with the number of missing data corresponding to the missing values ​​in the historical data of ultra-low energy consumption buildings;

[0119] The outlier degree coefficient is obtained by the following formula:

[0120]

[0121] Among them, E represents the outlier degree coefficient; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C z Represents the total amount of historical data of ultra-low energy consumption buildings; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N z The total number of data representing the historical data of ultra-low energy buildings;

[0122] Comparing the outlier degree coefficient with a preset coefficient threshold;

[0123] Determine the target ratios of the training set and the test set according to the comparison result between the outlier degree coefficient and the preset coefficient threshold;

[0124] The ultra-low energy consumption building characteristic data are divided according to the target ratios corresponding to the training set and the test set to determine the training set and the test set.

[0125] The technical effect of the above technical solution is: by extracting invalid values ​​and missing values ​​from the historical data of ultra-low energy consumption buildings, the technical solution can quantify the degree of abnormality in the data. Invalid values ​​and missing values ​​are common problems in data cleaning, which will affect the accuracy and performance of machine learning models. By calculating the outlier degree coefficient (E), the solution can comprehensively evaluate the overall quality of the data and provide an important reference for subsequent data preprocessing and model training. According to the comparison result of the outlier degree coefficient with the preset coefficient threshold, the solution can dynamically adjust the target ratio of the training set and the test set. This means that when the data quality is high (the outlier degree coefficient is low), more data can be used for model training; when the data quality is low (the outlier degree coefficient is high), the proportion of training data needs to be appropriately reduced to avoid model overfitting or performance degradation. This dynamic adjustment strategy helps to improve the generalization ability and stability of the model under different data quality conditions. By optimizing the division of the training set and the test set, the technical solution can provide more accurate and representative training data and test data for the machine learning model. This helps to improve the accuracy, robustness and generalization ability of the model. At the same time, the solution can also reduce the model performance fluctuations caused by data quality issues and improve the reliability and stability of the model in practical applications. The technical solution realizes the automation and intelligence of data set division by calculating the outlier degree coefficient and dynamically adjusting the training set / test set ratio. This reduces the reliance on manual intervention and improves the efficiency and accuracy of data processing. In addition, the solution also has certain scalability and flexibility, and can adapt to the characteristic data of low-energy buildings of different scales and types.

[0126] In summary, this technical solution provides strong support for the machine learning application of ultra-low energy building characteristic data through technical effects such as quantifying data quality, optimizing data set division and improving model performance.

[0127] Specifically, the ratio of the training set to the test set is determined according to the comparison result between the outlier degree coefficient and the preset coefficient threshold, including:

[0128] Retrieving a comparison result between the abnormal value degree coefficient and a preset coefficient threshold;

[0129] When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is lower than the preset coefficient threshold value, the preset initial ratio value is retrieved as the target ratio;

[0130] When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is not lower than the preset coefficient threshold value, the preset initial ratio value is retrieved;

[0131] The target ratio is obtained by using the abnormal value degree coefficient and the preset initial ratio value, wherein the target ratio is obtained by the following formula:

[0132]

[0133] Where R represents the target ratio; R0 represents the preset initial ratio value; E represents the outlier degree coefficient; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N c represents the number of data corresponding to invalid values ​​in the historical data of the ultra-low energy consumption building; T represents the preset coefficient threshold; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C z Represents the total amount of historical data of ultra-low energy consumption buildings; N z represents the total number of historical data of ultra-low energy consumption buildings; x represents the adjustment coefficient, and the value range of the adjustment coefficient is 0.21-1.88.

[0134] The technical effect of the above technical solution is: by comparing the outlier degree coefficient with the preset coefficient threshold, the technical solution can dynamically adjust the ratio of the training set and the test set. This dynamic adjustment mechanism ensures that under different data quality conditions, the machine learning model can be provided with appropriate proportions of training and test data. When the outlier degree coefficient is low, it indicates that the data quality is high. At this time, the preset initial ratio value is used as the target ratio, and the high-quality data can be fully utilized for model training, thereby improving the accuracy and generalization ability of the model. When the outlier degree coefficient is not lower than the preset coefficient threshold, it indicates that there are many outliers in the data. At this time, the target ratio calculated by the formula will be adjusted accordingly to reduce the impact of outliers on model training and further improve the robustness of the model. The technical solution introduces an adjustment coefficient x, whose value range is between 0.21-1.88. By adjusting the value of the adjustment coefficient, the range of variation of the target ratio can be flexibly controlled to adapt to the characteristic data of low-energy buildings of different scales and types. In addition, the scheme also considers a variety of data quality indicators (such as the amount of invalid data, the number of missing data, etc.), making the technical solution more comprehensive and flexible. By calculating the outlier degree coefficient and dynamically adjusting the training set / test set ratio, this technical solution realizes the automation and intelligence of data set division. This reduces the reliance on manual intervention and improves the efficiency and accuracy of data processing. In the case of high data quality, by maintaining the original training set ratio, limited data resources can be used more effectively and the training efficiency of the model can be improved. In the case of low data quality, by increasing the training set ratio and correspondingly increasing the test set ratio, it can ensure that the model can more comprehensively evaluate its performance during the testing phase, thereby avoiding the problem of overfitting or underfitting.

[0135] In summary, this technical solution provides strong support for the machine learning application of ultra-low energy building characteristic data through dynamic adjustment of data set ratio, improvement of model generalization ability, flexibility and scalability, automation and intelligence, and optimization of resource utilization.

[0136] S3: Based on the optimal ultra-low energy consumption prediction model, the energy consumption of ultra-low energy consumption buildings is predicted, and the predicted energy consumption value of ultra-low energy consumption buildings is determined;

[0137] In this embodiment, energy consumption prediction is performed for ultra-low energy consumption buildings by performing the following operations:

[0138] Collect real-time data of ultra-low energy consumption buildings according to the energy consumption analysis requirements of ultra-low energy consumption buildings;

[0139] Input the real-time data of ultra-low energy consumption buildings into the optimal ultra-low energy consumption building energy consumption prediction model;

[0140] Based on the optimal ultra-low energy consumption prediction model, the real-time data of ultra-low energy consumption buildings is mined and analyzed to predict the energy consumption of ultra-low energy consumption buildings;

[0141] Determine the predicted energy consumption of ultra-low energy buildings.

[0142] It should be noted that by mining and analyzing the real-time data of ultra-low energy consumption buildings through the optimal ultra-low energy consumption prediction model, the energy consumption of ultra-low energy consumption buildings can be predicted, and the energy consumption prediction value of ultra-low energy consumption buildings can be determined, which is convenient for subsequent analysis of the energy consumption of ultra-low energy consumption buildings and also convenient for subsequent timely management of the energy consumption of ultra-low energy consumption buildings.

[0143] S4: Based on the energy consumption standards for ultra-low energy buildings, analyze the predicted energy consumption values ​​of ultra-low energy buildings and determine the energy consumption analysis results of ultra-low energy buildings.

[0144] In this embodiment, the predicted energy consumption value of the ultra-low energy consumption building is analyzed, and the following operations are performed:

[0145] According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the energy consumption standards of ultra-low energy consumption buildings are set in advance;

[0146] Based on the energy consumption standards for ultra-low energy consumption buildings, the predicted energy consumption values ​​of ultra-low energy consumption buildings are analyzed to determine the energy consumption analysis results of ultra-low energy consumption buildings;

[0147] Among them, if the predicted value of ultra-low energy consumption building energy consumption is within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is that the ultra-low energy consumption building energy consumption is normal;

[0148] Among them, if the predicted value of ultra-low energy consumption building energy consumption is not within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is abnormal ultra-low energy consumption building energy consumption;

[0149] When the energy consumption of ultra-low energy consumption buildings is abnormal, early warning reminders will be issued to guide management personnel to take energy-saving measures in a timely manner, and the design and layout of ultra-low energy consumption buildings will be optimized to ensure normal energy consumption of ultra-low energy consumption buildings.

[0150] It should be noted that based on the energy consumption standards for ultra-low energy consumption buildings, the predicted energy consumption values ​​of ultra-low energy consumption buildings are analyzed, and the energy consumption analysis results of ultra-low energy consumption buildings are shown in Table 1:

[0151] Table 1: Energy consumption analysis results of ultra-low energy consumption buildings

[0152]

[0153]

[0154] Therefore, based on the energy consumption standards of ultra-low energy consumption buildings, the predicted energy consumption values ​​of ultra-low energy consumption buildings are analyzed to determine the energy consumption analysis results of ultra-low energy consumption buildings. Based on the energy consumption analysis results of ultra-low energy consumption buildings, an energy consumption control plan for ultra-low energy consumption buildings is formulated. Based on the energy consumption control plan for ultra-low energy consumption buildings, the energy consumption of ultra-low energy consumption buildings is controlled, such as: conducting energy consumption early warning reminders for ultra-low energy consumption buildings, guiding management personnel to take energy-saving measures in a timely manner, and optimizing the design and layout of ultra-low energy consumption buildings to ensure normal energy consumption of ultra-low energy consumption buildings.

[0155] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device.

[0156] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing energy consumption of ultra-low energy consumption buildings, characterized in that: The steps include: S1: According to the energy consumption analysis requirements of ultra-low energy consumption buildings, collect historical data of ultra-low energy consumption buildings, process the historical data of ultra-low energy consumption buildings, and determine the characteristic data of ultra-low energy consumption buildings; S2: Construct an energy consumption prediction model for ultra-low energy consumption buildings, test and optimize the energy consumption prediction model for ultra-low energy consumption buildings, and determine the optimal energy consumption prediction model for ultra-low energy consumption buildings; S3: Based on the optimal ultra-low energy consumption prediction model, the energy consumption of ultra-low energy consumption buildings is predicted, and the predicted energy consumption value of ultra-low energy consumption buildings is determined; S4: Based on the energy consumption standards for ultra-low energy buildings, analyze the predicted energy consumption values ​​of ultra-low energy buildings and determine the energy consumption analysis results of ultra-low energy buildings.

2. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 1, characterized in that: In S1, historical data of ultra-low energy consumption buildings is collected, and the following operations are performed: According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the area shape, geographical location, orientation, enclosure materials, window configuration and foundation configuration of ultra-low energy consumption buildings are collected to obtain basic data of ultra-low energy consumption buildings; According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the temperature, humidity, light and wind speed of the environment in which the ultra-low energy consumption buildings are located are collected to obtain the environmental data of ultra-low energy consumption buildings; Among them, the historical data of ultra-low energy consumption buildings are determined based on the basic data of ultra-low energy consumption buildings and the environmental data of ultra-low energy consumption buildings.

3. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 2, characterized in that: In S1, the historical data of ultra-low energy consumption buildings is processed, and the following operations are performed: Obtain historical data on ultra-low energy buildings; Clean the historical data of ultra-low energy buildings, including: Conduct consistency checks on historical data of ultra-low energy buildings; According to the data consistency requirements, check whether there are inconsistent data in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the inconsistent data in the historical data of ultra-low energy consumption buildings; Process invalid and missing values ​​in historical data of ultra-low energy consumption buildings; According to the requirements of data validity and completeness, check whether there are invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings that are useless for the energy consumption analysis of ultra-low energy consumption buildings, and remove the invalid values ​​and missing values ​​in the historical data of ultra-low energy consumption buildings; Identify historical data of ultra-low energy buildings that are useful for energy analysis of ultra-low energy buildings.

4. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 3 is characterized in that: In S1, the historical data of ultra-low energy consumption buildings is processed, and the following operations are performed: Obtain historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings; Convert and process the historical data of ultra-low energy buildings that are useful for energy consumption analysis of ultra-low energy buildings; Eliminate the dimensional differences between historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, unify the formats of historical data of ultra-low energy consumption buildings that are useful for energy consumption analysis of ultra-low energy consumption buildings, and determine standardized historical data of ultra-low energy consumption buildings; Feature extraction of standardized ultra-low energy building historical data; Extract features that can reflect the energy consumption analysis of ultra-low energy buildings; Determine the characteristic data of ultra-low energy consumption buildings.

5. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 4, characterized in that: In S2, an ultra-low energy consumption building energy consumption prediction model is constructed, and the following operations are performed: Obtaining ultra-low energy consumption building characteristic data; Divide the characteristic data of ultra-low energy consumption buildings to determine the training set and test set; Select a convolutional neural network model framework suitable for energy consumption prediction of ultra-low energy buildings; Based on the training set, the selected convolutional neural network model framework suitable for ultra-low energy consumption building energy consumption prediction is trained to determine the ultra-low energy consumption building energy consumption prediction model.

6. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 5, characterized in that: The characteristic data of ultra-low energy consumption buildings are divided to determine the training set and test set, including: Extracting the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; Extracting the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; Obtaining an outlier degree coefficient by using the amount of invalid values ​​in the historical data of ultra-low energy consumption buildings combined with the number of missing data corresponding to the missing values ​​in the historical data of ultra-low energy consumption buildings; The outlier degree coefficient is obtained by the following formula: Among them, E represents the outlier degree coefficient; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C z Represents the total amount of historical data of ultra-low energy consumption buildings; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N z The total number of data representing the historical data of ultra-low energy buildings; Comparing the outlier degree coefficient with a preset coefficient threshold; Determine the target ratios of the training set and the test set according to the comparison result between the outlier degree coefficient and the preset coefficient threshold; The ultra-low energy consumption building characteristic data are divided according to the target ratios corresponding to the training set and the test set to determine the training set and the test set.

7. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 6, characterized in that: Determining the corresponding ratio of the training set and the test set according to the comparison result between the outlier degree coefficient and the preset coefficient threshold, including: Retrieving a comparison result between the abnormal value degree coefficient and a preset coefficient threshold; When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is lower than the preset coefficient threshold value, the preset initial ratio value is retrieved as the target ratio; When the comparison result between the abnormal value degree coefficient and the preset coefficient threshold value indicates that the abnormal value degree coefficient is not lower than the preset coefficient threshold value, the preset initial ratio value is retrieved; The target ratio is obtained by using the abnormal value degree coefficient and the preset initial ratio value, wherein the target ratio is obtained by the following formula: Where R represents the target ratio; R0 represents the preset initial ratio value; E represents the outlier degree coefficient; N x represents the number of missing data corresponding to the missing values ​​in the historical data of the ultra-low energy consumption building; N c represents the number of data corresponding to invalid values ​​in the historical data of the ultra-low energy consumption building; T represents the preset coefficient threshold; C w represents the amount of invalid values ​​in the historical data of the ultra-low energy consumption building; C z Represents the total amount of historical data of ultra-low energy consumption buildings; N z represents the total number of historical data of ultra-low energy consumption buildings; x represents the adjustment coefficient, and the value range of the adjustment coefficient is 0.21-1.

88.

8. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 5, characterized in that: In S2, the energy consumption prediction model of ultra-low energy consumption buildings is tested and optimized, and the following operations are performed: Obtain energy consumption prediction models for ultra-low energy buildings; Based on the test set, the performance of the ultra-low energy consumption building energy consumption prediction model is tested to determine whether the performance of the ultra-low energy consumption building energy consumption prediction model can achieve the expected effect; Determine the performance test results based on the ultra-low energy building energy consumption prediction model; Based on the performance test results, the energy consumption prediction model of ultra-low energy consumption buildings is mined and analyzed, the parameter configuration of the energy consumption prediction model of ultra-low energy consumption buildings is adjusted, the structural setting of the energy consumption prediction model of ultra-low energy consumption buildings is optimized, the energy consumption prediction model of ultra-low energy consumption buildings is repeatedly iterated and optimized, and the optimal energy consumption prediction model of ultra-low energy consumption buildings is determined.

9. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 8, characterized in that: In S3, energy consumption prediction is performed on ultra-low energy consumption buildings by performing the following operations: Collect real-time data of ultra-low energy consumption buildings according to the energy consumption analysis requirements of ultra-low energy consumption buildings; Input the real-time data of ultra-low energy consumption buildings into the optimal ultra-low energy consumption building energy consumption prediction model; Based on the optimal ultra-low energy consumption prediction model, the real-time data of ultra-low energy consumption buildings is mined and analyzed to predict the energy consumption of ultra-low energy consumption buildings; Determine the predicted energy consumption of ultra-low energy buildings.

10. The method for analyzing energy consumption of ultra-low energy consumption buildings according to claim 9, characterized in that: In S4, the predicted energy consumption value of the ultra-low energy consumption building is analyzed, and the following operations are performed: According to the energy consumption analysis requirements of ultra-low energy consumption buildings, the energy consumption standards of ultra-low energy consumption buildings are set in advance; Based on the energy consumption standards for ultra-low energy consumption buildings, the predicted energy consumption values ​​of ultra-low energy consumption buildings are analyzed to determine the energy consumption analysis results of ultra-low energy consumption buildings; Among them, if the predicted value of ultra-low energy consumption building energy consumption is within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is that the ultra-low energy consumption building energy consumption is normal; Among them, if the predicted value of ultra-low energy consumption building energy consumption is not within the ultra-low energy consumption building energy consumption standard range, the ultra-low energy consumption building energy consumption analysis result is abnormal ultra-low energy consumption building energy consumption; When the energy consumption of ultra-low energy consumption buildings is abnormal, early warning reminders will be issued to guide management personnel to take energy-saving measures in a timely manner, and the design and layout of ultra-low energy consumption buildings will be optimized to ensure normal energy consumption of ultra-low energy consumption buildings.

Images