Building load prediction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a building load prediction method and device, electronic equipment and a storage medium. The method comprises the steps that the influence weight of each influence factor influencing a building load is determined; based on the influence weight and factor data of each influence factor, determining a comprehensive influence index influencing the building load; and inputting the comprehensive influence index into a pre-trained building load prediction model, and obtaining a prediction result of the building load by using the building load prediction model. According to the embodiment of the invention, the parameter quantity and prediction workload of building load pre-test by using the model can be reduced, and the real-time performance of prediction is improved.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a building load forecasting method, device, electronic equipment and storage medium. Background Art

[0002] The load of a building is affected by many factors, such as the climate of the region, the building space, and the people in the building. The existing methods for predicting building loads include taking various influencing factors as inputs and using neural network models and deep learning models for prediction. However, due to the large number of influencing factors and the large amount of corresponding data, the existing technology has problems such as too many parameters, too much prediction workload, and poor real-time performance when using models for load prediction. Summary of the invention

[0003] The embodiments of the present invention provide a building load prediction method, device, electronic device and storage medium, which can reduce the number of parameters and prediction workload for building load pre-testing using a model and improve the real-time performance of the prediction.

[0004] In a first aspect, an embodiment of the present invention provides a building load prediction method, comprising:

[0005] Determine the influence weight of each factor affecting the building load;

[0006] Determining a comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor; and

[0007] The prediction result of the building load is obtained based on the comprehensive impact index through a pre-trained building load prediction model.

[0008] In a second aspect, an embodiment of the present invention provides a building load prediction device, comprising:

[0009] A weight acquisition module is used to determine the influence weights of various influencing factors affecting the building load;

[0010] A comprehensive index acquisition module, used to determine the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor; and

[0011] The prediction result acquisition module is used to obtain the prediction result of the building load based on the comprehensive impact index through a pre-trained building load prediction model.

[0012] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a building load prediction method as described in any one of the embodiments of the present invention is implemented.

[0013] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a building load prediction method as described in any one of the embodiments of the present invention.

[0014] The embodiments of the present invention provide a method, device, electronic device and storage medium for predicting building loads. The method determines the weights of various influencing factors affecting the building loads, and determines the comprehensive influencing index affecting the building loads through factor data based on the weights of the influencing factors. The method further obtains the prediction result of the building load based on the comprehensive influencing index through a pre-trained building load prediction model. The method can reduce the amount of input data when using the model to predict the building load, and further improve the real-time performance of the prediction by reducing the number of parameters and prediction workload for pre-testing the building load using the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of a building load prediction method provided by an embodiment of the present invention;

[0017] Figure 2 is another flow chart of the building load prediction method provided by an embodiment of the present invention;

[0018] Figure 3 It is a flow chart of the training process of the building load prediction model in the building load prediction method provided by an embodiment of the present invention;

[0019] Figure 4 is a structural schematic diagram of a building load prediction device provided by an embodiment of the present invention;

[0020] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] Figure 1 The present invention provides a flow chart of a method for predicting building loads in accordance with an embodiment of the present invention. The method can be performed by the XX device provided in accordance with an embodiment of the present invention. The device can be implemented in software and / or hardware. In a specific embodiment, the device can be integrated into an electronic device, such as a computer, a server, etc. The following embodiments will be described by taking the device integrated into an electronic device as an example. Figure 1 , the method may specifically include the following steps:

[0024] Step 101, determining the influence weights of various influencing factors affecting the building load. This step can help determine the comprehensive influencing factors affecting the building load based on the influence weights.

[0025] Specifically, the above-mentioned building loads may be loads of various buildings such as commercial buildings, office buildings, scenic spot buildings, and sports venues.

[0026] Specifically, the above-mentioned influencing factors may include: climate factors of the area where the building is located, building space factors, personnel factors in the building and historical loads, etc.

[0027] Specifically, the climate factors of the area where the above-mentioned building is located may include information such as the weather, humidity, temperature and pollution predicted for the day in the area.

[0028] Specifically, the above-mentioned building space factors may include: the area, floor, orientation and adjacent building information of the space within the building.

[0029] Specifically, the above-mentioned personnel factors in the building may include: the number of people in the building, changes in the number, age composition, gender composition and other information.

[0030] Optionally, the process of determining the influence weights of various influencing factors affecting the building load includes: manually determining the influence weights of various influencing factors based on experience.

[0031] Optionally, the process of determining the influence weight of each influencing factor affecting the building load includes: using a hierarchical analysis method to determine the influence weight of each influencing factor.

[0032] Step 102, based on the influence weights and factor data of each influencing factor, determine the comprehensive influence index affecting the building load. This step can be beneficial to obtain the prediction result of the building load using the building load prediction model based on the comprehensive influence index without too much input data.

[0033] Specifically, the above factor data may be historical data within a preset time period before the current moment. The preset time period may be set based on experience, or may be set based on test results after conducting tests on the effects of different historical time periods on the accuracy of building load prediction results.

[0034] Optionally, the process of determining the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor includes:

[0035] The various influencing factors are grouped according to their relevance to obtain a plurality of mutually unrelated relevant factor groups; and the comprehensive influencing indicators corresponding to the various relevant factor groups are determined.

[0036] Optionally, the process of determining the comprehensive influencing index affecting the building load based on the influencing weights and factor data of each influencing factor includes: using principal component analysis to determine the comprehensive influencing index affecting the building load based on the influencing weights and factor data of each influencing factor.

[0037] Step 103, input the comprehensive influencing index into the pre-trained building load prediction model, and use the building load prediction model to obtain the prediction result of the building load. This step, based on steps 101 and 102, determines the influence weights of various influencing factors affecting the building load, and determines the comprehensive influencing index affecting the building load through factor data based on the influence weights of various influencing factors, and further obtains the prediction result of the building load based on the comprehensive influencing index through the pre-trained building load prediction model; it can reduce the amount of input data when using the model to predict the building load, and further improve the real-time performance of the prediction based on reducing the number of parameters and prediction workload for pre-testing the building load using the model.

[0038] Specifically, the building load prediction model may be an Autoformer model or an improved Autoformer model.

[0039] The following further describes a building load prediction method in another embodiment of the present invention. Figure 2 As shown, Figure 1 Step 102 in may include the following steps:

[0040] Step 1021 , performing de-duplication processing, de-redundancy processing, de-anomaly processing and normalization processing on the factor data to obtain the data to be predicted corresponding to each influencing factor.

[0041] Optionally, the above process of removing abnormalities from the factor data includes: identifying abnormal data in the factor data based on the Laida criterion; and obtaining the average value of the factor data corresponding to the abnormal data, and removing the abnormal data by replacing the abnormal data with the average value.

[0042] Specifically, the process of identifying abnormal data in factor data based on the Laida criterion includes: establishing a normal distribution corresponding to the factor data of each influencing factor, and determining the factor data that is not within the distribution range of the corresponding normal distribution as abnormal data.

[0043] Optionally, the above process of obtaining the average value of the factor data corresponding to the abnormal data and removing the abnormal data by replacing the abnormal data with the average value includes: for the abnormal data of each influencing factor, calculating the average value of all factor data of the current influencing factor, and replacing the abnormal data of the current influencing factor with the average value.

[0044] Step 1022: determine the comprehensive impact index of the building load based on the impact weight of each impact factor and the to-be-predicted data corresponding to each impact factor.

[0045] Optionally, the process of determining the comprehensive impact index corresponding to each relevant factor group includes:

[0046] A group of to-be-predicted data corresponding to a time in the to-be-predicted data corresponding to each relevant factor group is weightedly summed based on corresponding influence weights to obtain a comprehensive influence index corresponding to each relevant factor group.

[0047] For example, a related factor group includes three related influencing factors A, B and C. The corresponding time corresponding group of data to be predicted is a, b and c. The influence weights of A, B and C are K and A , K B and K C , then for a, b and c based on K A , K B and KC The weighted summation is performed using the following formula:

[0048] I=K A a+K B a+K C c

[0049] Wherein, I represents the comprehensive impact index corresponding to the relevant factor group.

[0050] The embodiment of the present invention can avoid the influence of invalid and abnormal data on the building load prediction result by pre-processing the factor data of each influencing factor, thereby improving the accuracy of the prediction result.

[0051] The following further introduces a building load prediction method in another embodiment.

[0052] Optional, such as Figure 3 As shown, the training process of the building load prediction model in the embodiment of the present invention includes:

[0053] Step 301, obtain the sample comprehensive impact index, and input the sample comprehensive impact index into the building load prediction model to be trained, the building load prediction model includes: an encoder and a decoder, the encoder and the decoder are both embedded with a time series decomposition module and an autocorrelation mechanism module.

[0054] Optionally, the process of obtaining the sample comprehensive impact index includes: determining the impact weights of various influencing factors affecting the building load; and determining the sample comprehensive impact index based on the impact weights of various influencing factors and sample factor data.

[0055] Specifically, the above-mentioned influencing factors may include: climate factors of the area where the building is located, building space factors, personnel factors in the building and historical loads, etc.; the climate factors of the area where the building is located may specifically include information such as the weather, humidity, temperature and pollution predicted for the day in the area; the above-mentioned building space factors may specifically include: the area, floor, orientation and adjacent building information of the space in the building; the above-mentioned personnel factors in the building may specifically include: the number of people in the building, changes in the number, age composition and gender composition, etc.

[0056] Specifically, the sample factor data may be historical data of various influencing factors.

[0057] Specifically, the above-mentioned sample comprehensive impact indicator may include one or more time series.

[0058] Step 302: Obtain the periodic sequence and trend sequence corresponding to the comprehensive impact index of the sample through the time series decomposition module and the autocorrelation mechanism module of the encoder.

[0059] Optionally, the process of obtaining the periodic sequence and trend sequence corresponding to the comprehensive impact index of the sample through the time series decomposition module and the autocorrelation mechanism module of the encoder includes:

[0060] Aggregate the periodic part in the sample comprehensive influence index through the time series decomposition module of the encoder; decompose the periodic part from the sample comprehensive influence index through the time series decomposition module of the encoder to obtain the periodic term, and combine other parts in the sample comprehensive influence index to obtain the trend term; and encode based on the periodic term to obtain the periodic sequence through the encoder, and encode based on the trend term to obtain the trend sequence.

[0061] Optionally, the sliding average method is used to decompose the sequence. The decomposition process can be expressed as:

[0062]

[0063] Among them, X represents the latent variable to be decomposed, X t represents the trend term, X s represents a periodic item, AvgPool(Padding(X)) represents an average pooling operation, and Padding represents a zero-padding operation.

[0064] Optionally, when the encoder has N coding layers, for the first coding layer, the periodic term may be decomposed based on the following formula:

[0065]

[0066] in, Represents a periodic term, SeriesDecomp represents a time series decomposition operation, Autocorrelation represents an autocorrelation operation, and FeedForward represents a forward propagation operation. represents the original data of the l-1th layer in the encoder, Represents the periodic term of the encoder layer l.

[0067] Optionally, the process of encoding based on periodic items to obtain a periodic sequence and encoding based on trend items to obtain a trend sequence includes: using an encoder to obtain the periodic sequence by position encoding the periodic items and to obtain the trend sequence by position encoding the trend items.

[0068] Step 303: Obtain sample load prediction results based on the periodic sequence and the trend sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder.

[0069] Optionally, the process of obtaining the sample load prediction result based on the periodic sequence and the trend sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder includes: obtaining the periodic features in the periodic sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder, and performing autocorrelation stacking on the periodic features to obtain the decoded periodic features; and

[0070] The trend features in the periodic sequence are obtained through the time series decomposition module and the autocorrelation mechanism module of the decoder, and the trend features are accumulated with the trend sequence to obtain the decoded trend features.

[0071] Optionally, after obtaining the periodic sequence and the trend sequence from the encoder, the periodic sequence and the trend sequence are initialized to obtain an initialized periodic sequence and an initialized trend sequence, and then the initialized periodic sequence and the initialized trend sequence are input into the decoder.

[0072] Specifically, the hidden variables of the period part and the trend part in the initialization period sequence can be gradually split through the time series decomposition module and the autocorrelation mechanism module of the decoder, and the periodic features are gradually autocorrelated and superimposed, and the trend features are accumulated. Specifically, when the decoder has N coding layers, for the lth coding layer, the acquisition process of the above decoding period features and decoding trend features can be expressed as:

[0073]

[0074] in, and represents the periodic term, and represents the trend term, SeriesDecomp represents the time series decomposition operation, and Autocorrelation represents the autocorrelation operation. Represents the original data of layer l-1 in the decoder, Represents the original data of the Nth layer in the encoder, W l,1 , W l,2 , W l,3 are the coefficients of the three lth layers respectively, To decode the periodic features, To decode trend features.

[0075] Step 304: determine the function value of the training loss function based on the sample load prediction result, and adjust the network parameters in the building load prediction model according to the function value of the training loss function.

[0076] In an embodiment of the present invention, by training a building load prediction model including a time series decomposition module and an autocorrelation module, it is possible to obtain a periodic sequence and a trend sequence corresponding to the input feature to be predicted through the time series decomposition module and the autocorrelation module when predicting the building load, and to perform targeted feature extraction on the periodic sequence and the trend sequence respectively, thereby improving the accuracy of building load prediction; and through the autocorrelation module and the time series decomposition module, when extracting periodic features, sequence-level connection and complexity are achieved, breaking the bottleneck of information utilization, better extracting approximately equal periodic features in different periods, further improving the prediction accuracy of periodic features, thereby improving the accuracy of building load prediction.

[0077] Figure 4 FIG. 1 is a structural diagram of a building load prediction device provided by an embodiment of the present invention, and the device is suitable for executing a building load prediction method provided by an embodiment of the present invention. Figure 4 As shown, the device may specifically include:

[0078] The weight acquisition module 401 is used to determine the influence weights of various influencing factors affecting the building load, which can be helpful to determine the comprehensive influencing factors affecting the building load based on the influence weights.

[0079] The comprehensive index acquisition module 402 is used to determine the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor, so as to obtain the prediction result of the building load by using the building load prediction model based on the comprehensive impact index without inputting too much data.

[0080] Optionally, the comprehensive index acquisition module 402 can be specifically used to group various influencing factors according to their relevance to obtain a plurality of mutually unrelated relevant factor groups, and determine a comprehensive influencing index corresponding to each relevant factor group.

[0081] Optionally, the above-mentioned comprehensive index acquisition module 402 can be specifically used to deduplicate, remove redundancy, remove anomalies and normalize the factor data to obtain the data to be predicted corresponding to each influencing factor; and determine the comprehensive influencing index of the building load based on the influence weight of each influencing factor and the data to be predicted corresponding to each influencing factor.

[0082] The prediction result acquisition module 403 is used to obtain the prediction result of the building load based on the comprehensive influencing index through the pre-trained building load prediction model. In combination with module 401 and module 402, by determining the influence weights of various influencing factors affecting the building load, and determining the comprehensive influencing index affecting the building load through the factor data based on the influence weights of various influencing factors, the prediction result of the building load is further obtained based on the comprehensive influencing index through the pre-trained building load prediction model; it can reduce the amount of input data when using the model to predict the building load, and further improve the real-time performance of the prediction based on reducing the number of parameters and prediction workload for pre-testing the building load using the model.

[0083] Optionally, the training process of the above-mentioned building load prediction model includes: obtaining sample comprehensive impact indicators, and inputting the sample comprehensive impact indicators into the building load prediction model to be trained, the building load prediction model includes: an encoder and a decoder, both the encoder and the decoder are embedded with a timing decomposition module and an autocorrelation mechanism module; through the timing decomposition module and the autocorrelation mechanism module of the encoder, the periodic sequence and trend sequence corresponding to the sample comprehensive impact indicators are obtained; through the timing decomposition module and the autocorrelation mechanism module of the decoder, the sample load prediction results are obtained based on the periodic sequence and the trend sequence; and the function value of the training loss function is determined based on the sample load prediction results, and the network parameters in the building load prediction model are adjusted according to the function value of the training loss function.

[0084] Optionally, the process of obtaining the periodic sequence and trend sequence corresponding to the comprehensive impact index of the sample through the time series decomposition module and the autocorrelation mechanism module of the encoder includes:

[0085] Aggregate the periodic part in the sample comprehensive influence index through the time series decomposition module of the encoder; decompose the periodic part from the sample comprehensive influence index through the time series decomposition module of the encoder to obtain the periodic term, and combine other parts in the sample comprehensive influence index to obtain the trend term; and encode based on the periodic term to obtain the periodic sequence through the encoder, and encode based on the trend term to obtain the trend sequence.

[0086] Optionally, the above-mentioned process of obtaining sample load prediction results based on periodic sequence and trend sequence through the timing decomposition module and autocorrelation mechanism module of the decoder includes: obtaining periodic features in the periodic sequence through the timing decomposition module and autocorrelation mechanism module of the decoder, and autocorrelating and stacking the periodic features to obtain decoded periodic features; and obtaining trend features in the periodic sequence through the timing decomposition module and autocorrelation mechanism module of the decoder, and accumulating the trend features with the trend sequence to obtain decoded trend features.

[0087] Optionally, the process of obtaining a prediction result of a building load using a building load prediction model includes:

[0088] Through the time series decomposition module and autocorrelation mechanism module of the trained building load forecasting model, the periodic sequence and trend sequence corresponding to the comprehensive influencing index are obtained;

[0089] Through the time series decomposition module and the autocorrelation mechanism module of the decoder, the sample load prediction results are obtained based on the periodic sequence and trend sequence corresponding to the comprehensive influencing index.

[0090] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the functional modules described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0091] An embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the building load prediction method provided in any of the above embodiments is implemented.

[0092] An embodiment of the present invention further provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the building load prediction method provided by any of the above embodiments is implemented.

[0093] The embodiment of the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the building load prediction method as described in any one of the embodiments of the present invention.

[0094] Reference below Figure 5 , which shows a schematic diagram of the structure of a computer system 500 of an electronic device suitable for implementing an embodiment of the present invention. Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0095] like Figure 5As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0096] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.

[0097] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the system of the present invention are executed.

[0098] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0099] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0100] The modules and / or units involved in the embodiments of the present invention may be implemented in software or hardware. The modules and / or units described may also be provided in a processor, for example, may be described as: a processor including a weight acquisition module, a comprehensive index acquisition module, and a prediction result acquisition module. The names of these modules do not, in some cases, constitute limitations on the modules themselves.

[0101] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device implements: determining the influence weights of various influencing factors affecting the building load; determining a comprehensive influence index affecting the building load based on the influence weights and factor data of the various influencing factors; and inputting the comprehensive influence index into a pre-trained building load prediction model, and using the building load prediction model to obtain the prediction result of the building load.

[0102] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A building load prediction method, characterized in that: include: Determine the influence weight of each factor affecting the building load; Determining a comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor; and The comprehensive impact index is input into a pre-trained building load prediction model, and the building load prediction model is used to obtain the prediction result of the building load.

2. The building load prediction method according to claim 1, characterized in that: Determining the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor includes: The influencing factors are grouped according to their relevance to obtain a plurality of mutually unrelated relevant factor groups, and the comprehensive influencing index corresponding to each relevant factor group is determined.

3. The building load prediction method according to claim 1, characterized in that: Determining the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor includes: Performing de-duplication, de-redundancy, de-anomaly and normalization processing on the factor data to obtain the data to be predicted corresponding to each influencing factor; and The comprehensive impact index of the building load is determined based on the impact weights of the various impact factors and the data to be predicted corresponding to the various impact factors.

4. The building load prediction method according to claim 3, characterized in that: The step of removing abnormalities from the factor data includes: Identifying abnormal data in the factor data based on the Laida criterion; and The average value of the factor data corresponding to the abnormal data is obtained, and the abnormal data is removed by replacing the abnormal data with the average value.

5. The building load prediction method according to claim 1, characterized in that: The training process of the building load prediction model includes: Obtaining a sample comprehensive impact index, and inputting the sample comprehensive impact index into a building load forecasting model to be trained, wherein the building load forecasting model comprises: an encoder and a decoder, wherein both the encoder and the decoder are embedded with a timing decomposition module and an autocorrelation mechanism module; Obtaining the periodic sequence and trend sequence corresponding to the comprehensive impact index of the sample through the time series decomposition module and the autocorrelation mechanism module of the encoder; Obtaining a sample load prediction result based on the period sequence and the trend sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder; and The function value of the training loss function is determined based on the sample load prediction result, and the network parameters in the building load prediction model are adjusted according to the function value of the training loss function.

6. The building load prediction method according to claim 5, characterized in that: The step of obtaining the periodic sequence and trend sequence corresponding to the comprehensive impact index of the sample through the time series decomposition module and the autocorrelation mechanism module of the encoder includes: Aggregating the periodic part of the sample comprehensive impact index by the timing decomposition module of the encoder; Decomposing the periodic part from the sample comprehensive impact index by the time series decomposition module of the encoder to obtain a periodic term, and combining other parts of the sample comprehensive impact index to obtain a trend term; and The encoder performs encoding based on the periodic item to obtain a periodic sequence, and performs encoding based on the trend item to obtain a trend sequence.

7. The building load prediction method according to claim 5, characterized in that: The obtaining of sample load prediction results based on the periodic sequence and the trend sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder includes: Acquire the periodic features in the periodic sequence through the time series decomposition module and the autocorrelation mechanism module of the decoder, and perform autocorrelation stacking on the periodic features to obtain decoded periodic features; and The trend features in the periodic sequence are acquired through the time series decomposition module and the autocorrelation mechanism module of the decoder, and the trend features are accumulated with the trend sequence to obtain the decoded trend features.

8. A building load prediction device, characterized in that: include: A weight acquisition module is used to determine the influence weights of various influencing factors affecting the building load; A comprehensive index acquisition module, used to determine the comprehensive impact index affecting the building load based on the impact weights and factor data of each impact factor; and The prediction result acquisition module is used to obtain the prediction result of the building load based on the comprehensive impact index through a pre-trained building load prediction model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the building load prediction method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the building load prediction method as described in any one of claims 1 to 7 is implemented.

Images