WO2023005120A1

WO2023005120A1 - Energy consumption prediction method and apparatus for building, and computer device and storage medium

Info

Publication number: WO2023005120A1
Application number: PCT/CN2021/139906
Authority: WO
Inventors: 杨志科; 蒋秋明; 王兴荣; 董孔益
Original assignee: 上海上实龙创智能科技股份有限公司
Priority date: 2021-07-27
Filing date: 2021-12-21
Publication date: 2023-02-02
Also published as: CN113505537A

Abstract

Disclosed in the present application are an energy consumption prediction method and apparatus for a building, and a computer device and a storage medium. The energy consumption prediction method for a building comprises: acquiring energy consumption association data of a building within a target time period and energy consumption measurement data of the building within a historical time period; and inputting the energy consumption association data and the energy consumption measurement data into an energy consumption prediction model, so as to obtain energy consumption prediction data, within the target time period, which is output by the energy consumption prediction model.

Description

Building energy consumption prediction method, device, computer equipment and storage medium

This application claims priority to a Chinese patent application with application number 202110849274.8 filed with the China Patent Office on July 27, 2021, the entire contents of which are incorporated herein by reference.

technical field

The embodiments of the present application relate to image processing technologies, for example, to a building energy consumption prediction method, device, computer equipment and storage medium.

Background technique

In modern cities, building economy has become an important part of economic development in the central business district of a megacity. How to reasonably manage and control the energy consumption of buildings is an important means to achieve optimal building economic development.

In related technologies, building energy consumption is usually detected on the same day.

However, the timeliness of the energy consumption test results obtained from the daily test is poor.

Contents of the invention

Embodiments of the present application provide a building energy consumption prediction method, device, computer equipment, and storage medium, so as to realize the prediction of building energy consumption and improve the prediction accuracy of building energy consumption.

In the first aspect, the embodiment of the present application provides a method for predicting building energy consumption, including:

Obtain the energy consumption related data of the building target time period, and the energy consumption measurement data of the historical time period;

The energy consumption correlation data and the energy consumption measurement data are input into an energy consumption prediction model to obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.

In the second aspect, the embodiment of the present application also provides a building energy consumption prediction device, including:

The data acquisition module is configured to acquire the energy consumption related data of the building target time period, and the energy consumption measurement data of the historical time period;

The energy consumption prediction module is configured to input the energy consumption related data and the energy consumption measurement data into the energy consumption prediction model, and obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.

In a third aspect, the embodiment of the present application also provides a computer device, the computer device comprising:

at least one processor;

a storage device configured to store at least one program;

When the at least one program is executed by the at least one processor, the at least one processor implements the building energy consumption prediction method provided in the embodiment of the present application.

In the fourth aspect, the embodiment of the present application also provides a storage medium including computer-executable instructions, and the computer-executable instructions are used to execute the method for predicting building energy consumption as provided in the embodiment of the present application when executed by a computer processor .

Description of drawings

Fig. 1 is a flowchart of a method for predicting building energy consumption in an embodiment of the present application;

Fig. 2 is a flowchart of a method for predicting building energy consumption in another embodiment of the present application;

Fig. 3 is a schematic diagram of a model training in an embodiment of the present application;

Fig. 4 is a flowchart of a method for predicting building energy consumption in another embodiment of the present application;

Fig. 5 is a schematic structural diagram of a building energy consumption prediction device in an embodiment of the present application;

Fig. 6 is a schematic structural diagram of a computer device in an embodiment of the present application.

Detailed ways

The application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only some structures related to the present application are shown in the drawings but not all structures.

Figure 1 is a flow chart of a method for predicting building energy consumption provided by an embodiment of the present application. This embodiment is applicable to the situation of predicting building energy consumption. The method can be executed by a building energy consumption predicting device, which can Realized by software and/or hardware, configured in a computer device, the computer device can be a server device and a client device, for example, the client device can be a mobile phone, a tablet computer, a vehicle terminal or a desktop computer, etc. The method for predicting building energy consumption in an embodiment of the present application includes the following steps:

S110. Obtain the energy consumption related data of the building target time period and the energy consumption measurement data of the historical time period.

The target time period refers to the time period of the energy consumption to be predicted. The energy consumption associated data may refer to values of parameters used to predict building energy consumption. The historical time period is the time period prior to the target time period. Wherein, the duration of the historical time period is longer than the duration of the target time period, so that energy consumption trends within a period of time can be calculated and energy consumption can be predicted more accurately. Energy consumption measurement data is the real value of building energy consumption in the historical time period. Among them, the energy consumption related data may include temperature data, humidity data, illumination data, wind data, location data and time type data, etc.

Energy consumption associated data can be obtained by sending a request to the service interface. For example, temperature data, humidity data, and wind data can be obtained by sending requests to the service interface of the meteorological system. The illumination data can be obtained by sending a request to the service interface of the server that provides the humidity detection service. The time type data may be determined according to the pre-acquired time of the target time period and the corresponding relationship between time and time type. The energy consumption measurement data in the historical time period may be determined by querying the data in the historical time period in the energy consumption measurement data database. In addition, there are other ways to obtain it, which can be set as needed.

S120. Input the energy consumption related data and the energy consumption measurement data into an energy consumption prediction model, and obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.

The energy consumption prediction model is set to determine the energy consumption prediction data in the target time period according to the energy consumption correlation data in the target time period and the energy consumption measurement data in the historical time period. The energy consumption prediction model may be a machine learning model, and the training samples may include energy consumption related data in the first time period, energy consumption measurement data in the second time period, and energy consumption prediction data in the first time period, wherein the second time period The timing of is before the first time period. Wherein, the machine learning model may include a convolutional neural network model, a support vector machine model, or a recurrent neural network model.

The energy consumption prediction data is the energy consumption data in the target time period. Typically, the target time period includes a future time from the current time. For example, the current time is July 8, the target time period is July 9, and the historical time period may be July 1-July 8. In fact, energy consumption-related data is not the real value obtained by real-time measurement, but the data obtained by prediction.

The embodiment of the present application determines the energy consumption prediction data based on the data related to energy consumption and the historical energy consumption measurement data, and can predict the energy consumption data based on the historical data of energy consumption and the current data of the factors related to energy consumption, avoiding the It can only detect the energy consumption in real time, realize the prediction of energy consumption data based on existing factors and historical energy consumption trends, and take into account the data of different time periods and energy consumption trends, increase the predictive factors of energy consumption, and improve energy consumption. consumption forecast accuracy.

Fig. 2 is a flow chart of a building energy consumption prediction method provided by another embodiment of the present application. The technical solution of this embodiment is further refined on the basis of the above technical solution. The energy consumption prediction model is a Stacking integrated model. The Stacking integrated model includes a first layer model and a second layer model, the first layer model includes at least one strong learning model, and the second layer model includes a weak learning model. The method includes:

S210. Obtain the energy consumption related data of the building target time period and the energy consumption measurement data of the historical time period.

S220. Input the energy consumption related data and the energy consumption measurement data into an energy consumption prediction model, and obtain the energy consumption prediction data output by the energy consumption prediction model for the target time period, and the energy consumption prediction model It is a Stacking integrated model, the Stacking integrated model includes a first layer model and a second layer model, the first layer model includes at least one strong learning model, and the second layer model includes a weak learning model.

In fact, Stacking is a layered model integration framework. Taking two layers as an example, the first layer model is composed of multiple base learners whose input is the original training set, and the second layer model uses the output of the first layer base learner as the training set for retraining to obtain a complete The Stacking model. The energy consumption related data and energy consumption measurement data are input into the first layer model, and the characteristics output by the first layer model are obtained, and the characteristics are input into the second layer model, and the output of the second layer model is obtained, which is determined as the energy consumption prediction model The output is the energy consumption prediction data for the target time period.

Among them, the strong learning model has a good prediction effect, while the weak learning model has a strong focus. Through the strong learning model to improve the prediction accuracy, combined with the weak learning model to improve the feature learning ability, the prediction accuracy of the Stacking integrated model is greatly improved. The first layer model uses a variety of regression models to learn and predict the input data, and the prediction results are used as the input of the second layer model. The second layer model uses a relatively simple regression model to reduce the risk of overfitting and obtain the final forecast result.

The strong learning models include support vector machine models, extreme gradient boosting models, and backpropagation models, and the weak learning models include linear regression models.

Among them, Support Vector Machines (Support Vector Machines, SVM) is a supervised learning model and related learning algorithms that use classification and regression analysis to analyze data. Given a set of training samples, each training sample is labeled as belonging to one or the other of the two categories. The training algorithm for support vector machines creates a model that assigns new samples to one of two classes, making it a non-probabilistic binary linear classifier. The support vector machine model represents samples as points in a map in space, so that samples with a single class can be separated as clearly as possible. The support vector machine adopts the structural risk minimization criterion to design the learning machine, which compromises the empirical risk and the confidence range, and has good generalization ability; the support vector machine is specially aimed at the limited sample situation, and its goal is to obtain the optimal information under the existing information. solution, not just the optimal solution when the number of samples tends to infinity, but the number of samples in this application is limited.

The extreme gradient boosting model (eXtreme Gradient Boosting, xgboost) is a gradient boosting algorithm. Gradient boosting algorithms refer to a class of ensemble machine learning algorithms that can be used for classification or regression predictive modeling problems. Boosting is built from a decision tree model. Trees are added to the ensemble one at a time and adjusted to correct for prediction errors caused by previous models. Compared with other gradient boosting algorithms, xgboost has fast execution speed and good model performance. xgboost is the loss value after learning before each iteration of learning, and the predicted value of the base classifier is added to predict the result.

The backpropagation model (Backpropagation, BP) is a multilayer feedforward neural network trained according to the error backpropagation algorithm. The backpropagation model uses the gradient descent method for U-shaped energy connection, and uses the gradient search technology to minimize the mean square error between the actual output value and the expected output value of the network.

A linear regression model is a mathematical regression model that determines the correlation between variables.

By adopting the support vector machine model, extreme gradient boosting model and backpropagation model as the strong learning model, and using the linear regression model as the weak learning model, the advantages of various algorithms can be obtained, and the data space and structure can be observed from different angles to avoid Local optimality occurs when using a single model, and the algorithm fusion with high degree of difference and strong learning ability is used for optimization, so that the prediction effect of Stacking model fusion can reach the best.

The acquisition of energy consumption related data of the building target time period includes at least one of the following: temperature data, humidity data, illumination data and time type data.

The temperature data may refer to the predicted temperature of the building for a target time period. The humidity data may refer to the predicted humidity of the building for a target time period. The lighting data may refer to the predicted lighting of the building during the time period when the lighting exists in the target time period. The time type data may refer to a date type corresponding to a target time period, a corresponding season type, and the like. Among them, the date type includes working day type and non-working day type; season type corresponds to spring, summer, autumn and winter and other types. In addition, the energy-related data may also include wind power data, which is the predicted wind power of the building within the target time period. The energy consumption-associated data may also include location data and climate data, etc. The location data may refer to the geographic location of the building and the climate corresponding to the geographic location.

By configuring the energy consumption associated data as at least one of temperature data, humidity data, illumination data, and time type data, the range of factors affecting energy consumption can be enriched, and the accuracy of energy consumption prediction can be improved.

The temperature data includes the average temperature of the building in the target time period and the instantaneous temperature of each unit time period in the target time period, and the unit time period is the time obtained by dividing the target time period part.

The unit time period is a unit obtained by dividing the target time period. Exemplarily, the duration of the target time period is one day, and the unit time period may be 1 hour; for another example, the duration of the target time period is one month or one week, and the unit time period may be 1 day. In one example, the target time period is January 1st and the unit time period is one hour of January 1st. The temperature data may include the predicted average temperature for a day on January 1, and the predicted temperature for 24 hours. The humidity data may be the average humidity of a day on January 1st.

By configuring more detailed temperature data, the features related to energy consumption can be subdivided, the dimension of features can be increased, feature information can be enriched, and the accuracy of energy consumption prediction can be improved.

The energy consumption related data includes temperature data and humidity data; obtaining the energy consumption related data of the building target time period includes: sending a weather forecast data acquisition request to the meteorological system; receiving the temperature data and humidity of the target time period fed back by the meteorological system data.

Temperature data and humidity data can be obtained through meteorological systems that provide weather forecast services. The weather forecast data acquisition request is used to request the weather system to acquire the temperature data and humidity data of the building in the target time period. Wherein, the meteorological forecast data acquisition request includes the geographical location of the building, the forecast time, and the type of forecast data.

Obtain the predicted temperature and predicted humidity of the target time period through the meteorological system, improve the accuracy of temperature and humidity, and thus improve the accuracy of energy consumption prediction.

The energy consumption measurement data of the historical time period includes the energy consumption measurement data of each unit time period of the building in the historical time period; the energy consumption forecast data of the target time period includes the building in the Energy consumption forecast data for each unit time period within the target time period.

The unit time period is used to divide the target time period, and can also divide the historical time period. By configuring and refining the energy consumption measurement data and predicting the energy consumption prediction data of the building in a unit time period, the fine-grained energy consumption prediction data can be increased, and the prediction accuracy can be flexibly controlled.

Inputting the energy consumption associated data and the energy consumption measurement data into the energy consumption prediction model includes: forming data of a specified structure according to the energy consumption associated data and the energy consumption measurement data, and performing normalization processing , and input the normalized data into the energy consumption prediction model.

In an example, the training process of the Stacking integrated model may include: collecting energy consumption related data in a target time period, energy consumption measurement data in a historical time period, and energy consumption measurement data in a target time period. Wherein, the energy consumption measurement data in the target time period is used as the real value of the energy consumption prediction data in the target time period. According to the energy consumption correlation data and energy consumption measurement data, generate the sample data of the specified structure, each sample data includes the predicted average temperature in the first time period, the predicted instantaneous temperature in each hour in the first time period, and the predicted temperature in the first time period. Forecasted average humidity, date type for the first time period, forecast sunlight for each hour in the hours of light in the first time period and power consumption measurements for each hour in the second time period, and the first time period Hourly power consumption measurement data. For example, the first time period is the current day, and the second time period is 10 days before the current day. Wherein, the predicted instantaneous temperature for each hour in the first time period includes T0, T1, T2...T23. The power consumption measurement data of each hour in the second time period includes E0, E1...E229.

Based on the following formula: normalize the generated sample data.

in,

is the normalized sample data, σ _j is the deviation of the jth sample component, D _ij is the jth sample component of the i sample, m is the sample number of the jth sample component,

is the average value of the jth sample component in the m samples.

According to multiple sample data, input into the Stacking integrated model to train the Stacking integrated model. Among them, the Stacking integrated model includes two-layer models, the first-layer model includes support vector machine model, extreme gradient boosting model and backpropagation model, and the second-layer model includes linear regression model.

Among them, the data set of the sample data is first divided into a training set and a test set, and the training set is subjected to 5-fold cross-validation.

(2) Perform the following operations on each strong learning model (here SVM is taken as an example), as shown in Figure 3:

①For the training set: each cross-validation obtains the predicted value of the verification set, and the five predicted values are combined into the feature column output by the SVM in the new training set, and combined with the label feature column in the training set to form the input data of the second-layer model;

②For the test set: input the test set into each cross-validation model to obtain the predicted value, take the average value of five times, and use it as the feature column output by the SVM for the new test set.

(3) The feature column output by SVM, the feature column output by xgboost and the feature column output by BP constitute the output of the first layer model, and the label (label) column is used as the input of the second layer model, and the label column is the energy of the target time period consumption measurement data.

(4) Modeling using liner regression.

According to the error between the energy consumption prediction data of the target time period output by the liner regress in the Stacking integration model and the energy consumption measurement data of the target time period, adjust the parameters of multiple models in the Stacking integration model until the target output by the Stacking integration model If the error between the energy consumption prediction data in the time period and the energy consumption measurement data in the target time period is less than or equal to the preset error threshold, it is determined that the Stacking integrated model training is completed.

When the Stacking integrated model training is completed, based on the process shown in Figure 4, energy consumption prediction data is obtained. The building energy consumption prediction method may include: S410, collecting energy consumption correlation data and energy consumption measurement data. S420. Generate sample data of a specified structure according to the energy consumption correlation data and the energy consumption measurement data. S430, performing normalization processing on the generated sample data. S440. Process the normalized sample data through the vector machine model, the extreme gradient boosting model and the backpropagation model in the Stacking integrated model to obtain output data. S450, input the output data into the linear regression model in the Stacking integrated model for processing. S460. Obtain energy consumption prediction data output by the linear regression model, that is, energy consumption prediction data output by the Stacking integrated model.

Using the integrated model including the first-layer model of SVM, xgboost and BP and the second-layer model of the Stacking integrated model of liner regression, it is possible to predict the power consumption prediction data for each hour of the next day, next week or next month, and realize Predicting the energy consumption of buildings will help relevant technicians allocate resources rationally and avoid waste of resources.

In the embodiment of the present application, by using the Stacking integrated model as the energy consumption prediction model, the advantages of various models are integrated, the prediction deviation is reduced, the generalization ability of the Stacking integrated model is increased, and the practicability is enhanced. At the same time, the prediction accuracy is improved through the strong learning model rate, combined with the weak learning model to improve the learning ability of features, and improve the prediction accuracy of the Stacking integrated model.

FIG. 5 is a schematic structural diagram of a building energy consumption prediction device provided by an embodiment of the present application. This embodiment is a corresponding device for implementing the method for predicting building energy consumption provided by the above embodiments. The device can be implemented in the form of software and/or hardware, and can generally be integrated into computer equipment. Building energy consumption prediction devices include:

The data acquisition module 510 is configured to acquire the energy consumption related data of the building target time period, and the energy consumption measurement data of the historical time period;

The energy consumption prediction module 520 is configured to input the energy consumption related data and the energy consumption measurement data into the energy consumption prediction model, and obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.

The energy consumption prediction model is a Stacking integrated model, the Stacking integrated model includes a first layer model and a second layer model, the first layer model includes at least one strong learning model, and the second layer model includes a weak learning model .

The energy consumption-related data includes temperature data and humidity data; the data acquisition module 510 is configured to: send a weather forecast data acquisition request to the meteorological system; and receive temperature data and humidity data of a target time period fed back by the meteorological system.

The temperature data includes the average temperature of the building in the target time period and the instantaneous temperature of each unit time period in the target time period, and the unit time period is obtained by dividing the target time period time period.

The duration of the historical time period is longer than the duration of the target time period, and the time sequence of the historical time period is before the time sequence of the target time period.

The energy consumption measurement data of the historical time period includes the energy consumption measurement data of each unit time period of the building in the historical time period; the energy consumption forecast data of the target time period includes the building in the Energy consumption prediction data for each unit time period within the target time period.

The above-mentioned device can execute the building energy consumption prediction method provided in the embodiment of the present application, and has corresponding functional modules and beneficial effects for executing the building energy consumption prediction method.

FIG. 4 is a schematic structural diagram of a computer device provided by an embodiment of the present application. FIG. 4 shows a block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present application. The computer device 12 shown in FIG. 4 is only an example, and should not limit the functions and scope of use of this embodiment of the present application.

As shown in FIG. 4, computer device 12 takes the form of a general-purpose computing device. Components of computer device 12 may include, but are not limited to: one or more processors or processing units 16 , system memory 28 , bus 18 connecting various system components including system memory 28 and processing unit 16 . The computer device 12 may be a bus-attached device.

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (Industry Standard Architecture, ISA) bus, Micro Channel Architecture (Micro Channel Architecture, MCA) bus, Enhanced ISA bus, Video Electronics Standards Association (Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12 and include both volatile and nonvolatile media, removable and non-removable media.

System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and/or cache memory 32 . Computer device 12 may include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 4, commonly referred to as a "hard drive"). Although not shown in FIG. 4, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk") may be provided, as well as a disk drive for a removable non-volatile disk (such as a Compact Disk ROM (Compact Disk). Disc Read-Only Memory, CD-ROM), Digital Video Disc (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical media) CD-ROM drive. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (eg, at least one) of program components configured to perform the functions of various embodiments of the present application.

Programs/utilities 40 may be stored, for example, in system memory 28 as a set (at least one) of program components 42 including, but not limited to, an operating system, one or more application programs, other program components, and program data, each or some combination of these examples may include the implementation of the network environment. Program components 42 generally perform the functions and/or methodologies of the embodiments described herein.

The computer device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with the computer device 12, and/or with Any device (eg, network card, modem, etc.) that enables the computing device 12 to communicate with one or more other computing devices. This communication can be performed through an input/output (Input/Output, I/O) interface 22 . And, computer device 12 can also communicate with one or more networks (such as local area network (Local Area Network, LAN), wide area network (Wide Area Network, WAN)) by network adapter 20. As shown in the figure, network adapter 20 communicates with by bus 18 other components of computer device 12. It should be appreciated that although not shown in FIG. 4, other hardware and/or software components may be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant Disk drive array (Redundant Arrays of Inexpensive Disks, RAID) system, tape drive and data backup storage system, etc.

The processing unit 16 executes a variety of functional applications and data processing by running the programs stored in the system memory 28 , such as realizing the building energy consumption prediction method provided by any embodiment of the present application.

An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the building energy consumption prediction method provided in all application embodiments of the present application is implemented:

That is to say, when the program is executed by the processor, it realizes: acquiring the energy consumption correlation data of the building target time period, and the energy consumption measurement data of the historical time period; inputting the energy consumption correlation data and the energy consumption measurement data into the energy consumption In the energy consumption prediction model, the energy consumption prediction data output by the energy consumption prediction model for the target time period is obtained.

The computer storage medium in the embodiments of the present application may use any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections with one or more conductors, portable computer disks, hard disks, RAM, Read Only Memory (ROM), erasable Programmable Read Only Memory (Erasable Programmable Read Only Memory, EPROM), flash memory, optical fiber, portable CD-ROM, optical storage device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. . The computer readable storage medium may be a non-transitory computer readable storage medium.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wire, optical cable, Radio Frequency (RF), etc., or any suitable combination of the foregoing.

Computer program code for performing the operations of the present application may be written in one or more programming languages or combinations thereof, including object-oriented programming languages—such as Java, Smalltalk, C++, and conventional Procedural Programming Language - such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. Where a remote computer is involved, the remote computer can be connected to the user computer through any kind of network, including a LAN or WAN, or, alternatively, can be connected to an external computer (eg, via the Internet using an Internet service provider).

Claims

A method for predicting building energy consumption, comprising:

Obtain the energy consumption related data of the building target time period, and the energy consumption measurement data of the historical time period;

The energy consumption correlation data and the energy consumption measurement data are input into an energy consumption prediction model to obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.
The method according to claim 1, wherein the energy consumption prediction model is a Stacking integrated model, the Stacking integrated model includes a first layer model and a second layer model, and the first layer model includes at least one strong learning model , the second layer model includes a weak learning model.
The method according to claim 2, wherein the strong learning model includes a support vector machine model, an extreme gradient boosting model, and a backpropagation model, and the weak learning model includes a linear regression model.
The method according to claim 1, wherein said acquiring the energy consumption-related data of the building target time period includes at least one of the following: temperature data, humidity data, illumination data and time type data.
The method according to claim 4, wherein the energy consumption associated data includes temperature data and humidity data;

The acquisition of the energy consumption related data of the building target time period includes:

Send weather forecast data acquisition request to the meteorological system;

The temperature data and humidity data of the target time period fed back by the meteorological system are received.
The method according to claim 4, wherein, the target time period includes a plurality of unit time periods; the temperature data includes the average temperature of the building in the target time period and the temperature of each unit in the target time period. The instantaneous temperature of unit time period.
The method according to claim 1, wherein the historical time period includes a plurality of unit time periods, and the target time period includes a plurality of unit time periods; the energy consumption measurement data of the historical time period includes the building in The energy consumption measurement data of each unit time period in the historical time period; the energy consumption prediction data of the target time period, including the energy consumption prediction of the building in each unit time period in the target time period data.
A building energy consumption prediction device, comprising:

The data acquisition module is configured to acquire the energy consumption related data of the building target time period, and the energy consumption measurement data of the historical time period;

The energy consumption prediction module is configured to input the energy consumption related data and the energy consumption measurement data into the energy consumption prediction model, and obtain the energy consumption prediction data of the target time period output by the energy consumption prediction model.
A computer device comprising:

at least one processor;

a storage device configured to store at least one program;

When the at least one program is executed by the one or more processors, the at least one processor implements the building energy consumption prediction method according to any one of claims 1-7.
A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for predicting building energy consumption according to any one of claims 1-7 is realized.