Civil building load calculation method and system based on big data and artificial intelligence

Abstract

The present invention is a method and system for calculating the load of civil buildings based on big data and artificial intelligence. The method includes T10 matching the building energy consumption database, T20 load calculation, and T30 determining the transformer, wherein T10 is the system design input, T20 is the system calculation process, and T30 is the system output. The input of the civil building load calculation includes four parts, namely the building energy consumption database, the total building area and the sub-item building area, the building type, and the area where the building is located; the load calculation output is the transformer configuration. The building energy consumption database is a national database of energy consumption of various types of buildings that has been clustered through big data statistics and artificial intelligence. The present invention changes the existing method of estimating key parameters based on the personal experience of engineers to using artificial intelligence to obtain actual big data of similar projects as the basis for design indicators of new projects. It avoids the wrong selection of transformers in civil buildings and saves operating costs and initial installation costs.

Description

Technical Field

[0001] The present invention belongs to the field of building power supply and distribution design, and in particular relates to a civil building load calculation method and system based on big data and artificial intelligence. Background Art

[0002] Load calculation is the basis of building electrical design and the basis for the selection of industrial and civil electrical equipment. It plays a vital role in the field of building electrical equipment.

[0003] In order to address the problem of inaccurate load calculation for civil buildings, State Grid Zhejiang Company's patent CN202111658345.2 adopts artificial intelligence ideas to predict users' short-term loads to improve the accuracy of holiday load forecasts; State Grid Shandong Company's patent CN201810470018.6 proposes a control method for active distribution networks, which can achieve accurate load forecasts; State Grid Hunan Company's patent CN202011295076.3 proposes a load classification calculation method for the power industry, which achieves a grid load forecasting effect with high reliability, good accuracy and a wide range of applications.

[0004] The load calculation method currently used in the design manual "Industrial and Civilian Power Supply and Distribution Design Manual" has three main shortcomings:

[0005] First, the calculation process is single-purpose. The current demand coefficient method for calculating load capacity uses a method of calculating the transformer backward from the terminal equipment. This involves first counting the terminal equipment to calculate the terminal distribution box, then calculating the floor distribution box, and finally calculating the transformer capacity. This calculation method specifies various parameters such as the demand coefficient, simultaneity coefficient, and dispersion coefficient for each calculation based on the number and type of equipment. However, this method does not account for differences in building type and region. For example, the same office building located in a tropical and a cold zone may have similar equipment configurations but very different user habits. Therefore, it would be unreasonable to obtain the same load calculation results for both.

[0006] Second, key parameters are inaccurate. Current load calculation methods are based on manual statistics. In industrial buildings, where equipment and processes are clearly defined, this method is relatively accurate. However, the spatial usage of civil buildings is highly flexible, making it impossible to provide effective parameters such as demand coefficients through manual statistics. Current civil building load calculation methods lack a basis for most key parameters, such as load indicators and demand coefficients, due to the emergence of new building space types. Electrical engineers are forced to provide their own data. Many parameters are also calculated using interval data. For example, the demand coefficient for electric heating equipment ranges from 0.3 to 0.5, leaving inexperienced electrical engineers unsure whether to choose a higher or lower value. These designs, which are based on empirical estimates by electrical engineers, lack objective calculation results.

[0007] Third, transformer selection is conservative. Because calculation methods are limited and key parameters cannot be accurately selected, electrical engineers often favor conservative choices to mitigate the risk of equipment failure. This escalating pressure leads to even more conservative transformer selection. Extensive research has revealed that transformer load rates in residential buildings are typically only 40% to 50%, significantly below the optimal operating range of 75% to 85%. This not only results in unnecessary waste but also inflates initial installation costs for owners.

[0008] In the research on load calculation of civil buildings that has appeared in recent years, there are still some deficiencies that need to be addressed. There are two main points:

[0009] First, there is a lack of research on power supply and distribution systems. Various load statistics studies are usually initiated by power grid companies, that is, research on the transmission side. These studies have reference value for building power supply and distribution systems, but cannot be directly applied.

[0010] Second, there is a lack of research on equipment selection. The key to power supply and distribution system research lies in the selection of transformer equipment; however, current load calculation research focuses on load forecasting without any reference for equipment selection. Summary of the Invention

[0011] To overcome the defects in the existing technology, the present invention proposes a civil building load calculation method and system based on big data and artificial intelligence. The existing method of estimating key parameters based on the personal experience of engineers is replaced by using artificial intelligence to obtain actual big data from similar projects as the basis for design indicators of new projects. This avoids the incorrect selection of transformers in civil buildings and saves operating costs and initial installation costs.

[0012] The present invention is achieved through the following technical solutions:

[0013] A method for calculating loads of civil buildings based on big data and artificial intelligence, comprising the following steps:

[0014] T10: Matching the building energy consumption database, divided into 3 sub-steps:

[0015] Step T11: Matching regional information: Matching data with the same regional tag in the civil building energy consumption database. If data for a specific region is missing, matching the data of the nearest region with data;

[0016] Step T12: Matching building type information: Matching data with the same building type label in the civil building energy consumption database.

[0017] Step T13: Matching sub-item area information: Matching data with the same sub-item area label in the civil building energy consumption database;

[0018] T20: Load calculation, divided into 4 sub-steps:

[0019] Step T21: Calculate the classified energy consumption load S E ;

[0020] Step T22: Calculate the energy consumption load S of each area M ;

[0021] Step T23: Calculate the energy consumption load S of the building type L ;

[0022] Step T24: Compare S E 、S M and S L , take the maximum value to get the total energy consumption of the building S0.

[0023] Step T30: Determine the transformer, which is divided into 3 sub-steps:

[0024] Step T31: Select the confidence coefficient k based on the size of the civil building energy consumption database z , where the confidence coefficient k z According to formula (1), we can get:

[0025]

[0026] Where, is the sample mean; is the critical value of confidence, which is 1.96 when the confidence level is 95%; S is the sample standard deviation; and n is the sample size in the civil building energy consumption database.

[0027] Step T32: Modify the energy consumption configuration according to the confidence coefficient: the final energy consumption configuration S1=k z ×S0;

[0028] Step T33: Calculate the transformer configuration: according to the transformer product sequence and project requirements, make all transformer configuration data greater than or equal to the energy consumption configuration S1.

[0029] Furthermore, the method further includes the step of establishing a civil building energy consumption database, including:

[0030] Step S10: Obtaining energy consumption big data;

[0031] Step S11: Obtain classified energy consumption data;

[0032] Step S12: Obtaining energy consumption data by area;

[0033] Step S13: Obtain building type energy consumption data;

[0034] Step S14: Obtaining a region label;

[0035] Step S20: Data analysis, divided into 4 sub-steps:

[0036] Step S21: Generate classified energy consumption unit data: extract each classified energy consumption generated in step S11, divide it by the building area, and obtain the classified energy consumption unit data;

[0037] Step S22: Generate unit data of energy consumption per item by area: Extract each unit of energy consumption per item generated in step S12 and divide it by the building area to obtain unit data of energy consumption per item by area;

[0038] Step S23: Generate building type energy consumption unit data: Extract the energy consumption of each building type generated in step S13 and divide it by the building area to obtain the building type energy consumption unit data;

[0039] Step S24: clustering using artificial intelligence algorithm.

[0040] Step S30: Data classification, divided into 3 sub-steps:

[0041] Step S31: Generate clustered classified energy consumption unit data, which is k categories in total.

[0042] Step S32: Generate clustered sub-item area energy consumption unit data, which is k categories in total.

[0043] Step S33: Generate clustered building type energy consumption unit data, which is k categories in total.

[0044] Furthermore, the steps of clustering using artificial intelligence algorithms include:

[0045] Step 1: Input the civil building energy consumption database and the number of clusters k;

[0046] Step 2: Initialize k cluster centers;

[0047] Step 3: assign each energy consumption data object to the nearest class;

[0048] Step 4: Recalculate the cluster centers;

[0049] Step 5: Determine whether the clustering target has converged; if not, return to step 3 and reallocate the cluster centers;

[0050] Step 6: If the clustering target has converged, output the result.

[0051] A civil building load calculation system based on big data and artificial intelligence, wherein the system is stored in a computer-readable storage medium in the form of computer-executable instructions. When a processor executes the computer-executable instructions, the civil building load calculation method based on big data and artificial intelligence as described above is implemented.

[0052] Due to the adoption of the above technical solution, the beneficial effect achieved by the present invention is that the method of estimating key parameters based on the personal experience of engineers in the existing technology is replaced by using artificial intelligence to obtain actual big data of similar projects as the basis for design indicators of new projects, thereby avoiding the wrong selection of transformers in civil buildings and saving operating costs and initial installation costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 Schematic diagram of load calculation in an embodiment of the present invention.

[0055] Figure 2 A flow chart is provided for establishing an energy consumption database in an embodiment of the present invention.

[0056] Figure 3 This is a flow chart of the artificial intelligence clustering algorithm in an embodiment of the present invention.

[0057] Figure 4 This is a load calculation flow chart in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following is a further description of specific embodiments of the present invention in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is intended to facilitate understanding of the present invention and does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0059] The invention belongs to the field of building power supply and distribution design, and in particular relates to a civil building load calculation system.

[0060] like Figure 1 As shown in the figure, the input of civil building load calculation includes four parts, namely building energy consumption database, total building area and sub-item building area, building type, and the area where the building is located; the output of load calculation is transformer configuration.

[0061] The building energy consumption database, a nationwide database of energy consumption for various types of buildings compiled through big data statistics and artificial intelligence clustering, is the core of this invention. The total building area refers to the building's reported building area; the sub-item building area is the area after the total building area is further divided into functional areas such as corridors, offices, restrooms, foyers, and elevator halls, as specified in Article 2.2.1 of the "Regulations on the Depth of Preparation of Building Engineering Design Documents (2016 Edition)." Building types refer to the 31 civil building types specified in Appendix A of GB 51348-2019 "Civil Building Electrical Design Standard," such as office buildings, theaters, cinemas, exhibition halls, shopping malls, airports, hotels, schools, hospitals, and residential buildings. The building's location should be specified at the prefecture-level administrative unit. Transformer configuration refers to the transformer capacity selection for the building's power supply and distribution system.

[0062] like Figure 2 As shown in Figure 1, the flow chart for establishing a civil building energy consumption database includes three steps, each of which includes three to four sub-steps. The specific steps are explained below.

[0063] Step S10: Obtaining Big Energy Consumption Data. This data is based on the energy consumption data generated by a large number of existing buildings over many years of operation. This data has been classified according to national standards and can be used for load calculations for similar projects. Step S10 is divided into four sub-steps.

[0064] Step S11: Obtain categorized energy consumption data. According to Article 10.5.1 of JGJT 229-2010, "Code for Green Design of Civil Buildings," building energy consumption systems are categorized by lighting, air conditioning, electricity, and electric vehicles. Each existing building generates a large amount of this data, which is then entered into the database.

[0065] Step S12: Obtaining energy consumption data by area. According to Article 2.2.1 of the "Regulations on the Depth of Preparation of Construction Engineering Design Documents (2016 Edition)", the total building area must be divided into functional areas, which are reflected in the architectural plan. Each existing building generates a large amount of total energy consumption data, which is then allocated to generate energy consumption data by area and recorded in the database.

[0066] Step S13: Obtain building type energy consumption data. According to Appendix A of GB 51348-2019, "Standard for Electrical Design of Civil Buildings," civil building types can be divided into 31 categories. Extract the building type information for each existing building and combine it with the generated total energy consumption data to enter it into the database.

[0067] Step S14: Obtaining a regional tag: Extracting the regional information of the existing building that generates each piece of energy consumption data, which must be specific to the prefecture-level administrative unit.

[0068] Step S20: Data analysis, which is divided into 4 sub-steps.

[0069] Step S21: Generate classified energy consumption unit data. Extract each classified energy consumption generated in step S11 and divide it by the building area to obtain the classified energy consumption unit data.

[0070] Step S22: Generate unit data of sub-item area energy consumption. Extract each sub-item area energy consumption generated in step S12 and divide it by the building area to obtain the unit data of sub-item area energy consumption.

[0071] Step S23: Generate building type energy consumption unit data. Extract the energy consumption of each building type generated in step S13 and divide it by the building area to obtain the building type energy consumption unit data.

[0072] Step S24: clustering using artificial intelligence algorithm. The clustering process is detailed in Figure 3 shown.

[0073] Step S30: Data classification, which is divided into 3 sub-steps.

[0074] Step S31: Generate clustered classified energy consumption unit data, which is k categories in total.

[0075] Step S32: Generate clustered sub-item area energy consumption unit data, which is k categories in total.

[0076] Step S33: Generate clustered building type energy consumption unit data, which is k categories in total.

[0077] like Figure 3 The figure shows the flow chart of artificial intelligence clustering algorithm.

[0078] Step 1: Input the energy consumption database and the number of clusters k. Typically, k = 3, which divides the energy consumption data into three categories: high, medium, and low, for the owner to choose from.

[0079] Step 2: Initialize k cluster centers.

[0080] Step 3: Assign each energy consumption data object to the nearest class.

[0081] Step 4: Recalculate the cluster centers.

[0082] Step 5: Determine whether the clustering target has converged. If not, return to step 3 and reallocate the cluster centers.

[0083] Step 6: If the clustering target has converged, output the result.

[0084] like Figure 4As shown in Figure 1, the civil building load calculation process consists of three steps, each of which includes three to four sub-steps. T10 represents the design input, T20 represents the calculation process, and T30 represents the system output. The specific steps are explained below.

[0085] Step T10: Matching the building energy consumption database is divided into 3 sub-steps.

[0086] Step T11: Match regional information. Match data with the same regional tag in the building energy consumption database. Similar to Section 17.6.3 of the Industrial and Civilian Power Supply and Distribution Design Manual (4th Edition), if data for a specific region is missing, match it with the nearest region that has data.

[0087] Step T12: Match building type information. Match data with the same building type tag in the building energy consumption database.

[0088] Step T13: Matching sub-item area information: Match data with the same sub-item area label in the building energy consumption database.

[0089] Step T20: Load calculation, which is divided into 4 sub-steps.

[0090] Step T21: Calculate the classified energy consumption load S E . Select the building grade according to the building purpose, corresponding Figure 3 The number of clusters k is usually selected from the three levels of high, medium and low. According to the classified energy consumption unit data matched in step S31, multiply it by the building area of the building to be calculated to obtain the four types of classified energy consumption calculation loads of lighting, air conditioning, electricity, and electric vehicles, and sum them to obtain the total load S E .

[0091] Step T22: Calculate the energy consumption load S of each area M Select the building purpose and building grade, and multiply the sub-item area data matched in step T13 and the sub-item area energy consumption unit data matched in step S32 by the building area to be calculated to obtain the calculation load of each sub-item area, and sum them to obtain the total load S M .

[0092] Step T23: Calculate the energy consumption load S of the building type L Select the building grade based on the building purpose, and multiply the building type information matched in step T12 and the building type energy consumption unit data matched in step S33 by the building area to be calculated to obtain the total load S L .

[0093] Step T24: Compare the three energy consumptions and take the larger one. E 、S M and S L, take the larger one to get the total energy consumption S0 of the building.

[0094] Step T30: Determine the transformer, which is divided into 3 sub-steps.

[0095] Step T31: Select the confidence coefficient k based on the database size z Among them, the confidence coefficient k z Calculated according to formula (1).

[0096]

[0097] Where, is the sample mean; is the critical value of confidence, which is 1.96 when the confidence level is 95%; S is the sample standard deviation; and n is the sample size in the civil building energy consumption database.

[0098] Step T32: Modify the energy consumption configuration according to the confidence coefficient. The final energy consumption configuration S1=k z ×S0.

[0099] Step T33: Calculate the transformer configuration. According to the transformer product sequence and project requirements, all transformer configuration data are greater than or equal to the energy consumption configuration S1.

[0100] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. It is apparent to those skilled in the art that various changes, modifications, substitutions, and variations to these embodiments may be made without departing from the principles and spirit of the present invention, and these changes and modifications still fall within the scope of protection of the present invention.

Claims

1. A method for calculating loads on civil buildings based on big data and artificial intelligence, characterized in that: Including steps: T10: Matching the building energy consumption database, divided into 3 sub-steps: Step T11: Matching regional information: Matching data with the same regional tag in the civil building energy consumption database. If data for a specific region is missing, matching the data of the nearest region with data; Step T12: Matching building type information: Matching data with the same building type label in the civil building energy consumption database; Step T13: Matching sub-item area information: Matching data with the same sub-item area label in the civil building energy consumption database; T20: Load calculation, divided into 4 sub-steps: Step T21: Calculate the classified energy consumption load S E ; Step T22: Calculate the energy consumption load S of each area M ; Step T23: Calculate the energy consumption load S of the building type L ; Step T24: Compare S E 、S M and S L , take the maximum value to get the total energy consumption of the building S0; T30: Determine the transformer, divided into 3 sub-steps: Step T31: Select the confidence coefficient k based on the size of the civil building energy consumption database z , where the confidence coefficient k z According to formula (1), we can get: Where, is the sample mean; z * is the critical value of confidence, when the confidence level is 95%, z * is 1.96; S is the sample standard deviation; n is the sample size in the civil building energy consumption database; Step T32: Modify the energy consumption configuration according to the confidence coefficient: the final energy consumption configuration S1 = k z ×S0; Step T33: Calculate the transformer configuration: according to the transformer product sequence and project requirements, make all transformer configuration data greater than or equal to the energy consumption configuration S1; It also includes the steps for establishing a civil building energy consumption database, including: Step S10: Obtaining energy consumption big data; Step S11: Obtain classified energy consumption data; Step S12: Obtaining energy consumption data by area; Step S13: Obtain building type energy consumption data; Step S14: Obtaining a region label; Step S20: Data analysis, divided into 4 sub-steps: Step S21: Generate classified energy consumption unit data: extract each classified energy consumption generated in step S11, divide it by the building area, and obtain the classified energy consumption unit data; Step S22: Generate unit data of energy consumption per item by area: Extract each unit of energy consumption per item generated in step S12 and divide it by the building area to obtain unit data of energy consumption per item by area; Step S23: Generate building type energy consumption unit data: Extract the energy consumption of each building type generated in step S13 and divide it by the building area to obtain the building type energy consumption unit data; Step S24: clustering using artificial intelligence algorithm; Step S30: Data classification, divided into 3 sub-steps: Step S31: Generate clustered classified energy consumption unit data, which is divided into k categories; Step S32: Generate clustered sub-item area energy consumption unit data, which is divided into k categories in total; Step S33: Generate clustered building type energy consumption unit data, which is k categories in total.

2. The method for calculating loads of civil buildings based on big data and artificial intelligence according to claim 1 is characterized in that: The steps of clustering using artificial intelligence algorithms include: Step 1: Input the civil building energy consumption database and the number of clusters k; Step 2: Initialize k cluster centers; Step 3: assign each energy consumption data object to the nearest class; Step 4: Recalculate the cluster centers; Step 5: Determine whether the clustering target has converged; if not, return to step 3 and reallocate the cluster centers; Step 6: If the clustering target has converged, output the result.

3. A civil building load calculation system based on big data and artificial intelligence, characterized in that: The system is stored in a computer-readable storage medium in the form of computer-executable instructions. When a processor executes the computer-executable instructions, the civil building load calculation method based on big data and artificial intelligence as described in any one of claims 1 to 2 is implemented.

Images