Method and device for predicting quantity of energy storage cabinets, electronic equipment and medium

Abstract

The invention provides an energy storage cabinet number prediction method and device, electronic equipment and a medium, and relates to the field of energy storage industry investment calculation, and the method comprises the steps: obtaining a plurality of historical electricity bills of a to-be-analyzed enterprise in a first duration and a plurality of load data files of a second duration; for any historical electricity bill, extracting data in the historical electricity bill to obtain key data; for the load data file of any second historical period, extracting data in the load data file to obtain load data; and processing the plurality of key data corresponding to the plurality of historical electricity bills and the plurality of load data corresponding to the plurality of load data files of the second duration by adopting the trained enhanced financial model to obtain a predicted number of energy storage cabinets required by the to-be-analyzed enterprise. According to the invention, a more intelligent and more flexible solution is provided for investment calculation in the energy storage industry, so that the overall investment decision-making efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of investment estimation in the energy storage industry, and specifically, to a method, device, electronic equipment and medium for predicting the number of energy storage cabinets. Background Art

[0002] Currently, the investment estimation tools available on the market mainly rely on traditional financial models and calculation methods. These tools usually estimate investment projects in the energy storage industry by setting a series of parameters, such as initial investment amount, interest rate, investment period, etc.

[0003] The investment estimation software designed specifically for the energy storage industry on the market can handle variables unique to energy storage projects, such as electricity price fluctuations, energy management, energy storage cabinet life, etc. However, these software are usually only targeted at specific types of projects, lack flexibility, and users need to have a high level of industry knowledge to use them effectively.

[0004] In recent years, with the development of machine learning technology, some researchers have begun to try to use machine learning algorithms to estimate the investment of energy storage projects. Such models can be trained with historical data and a large number of samples, thereby improving the accuracy of the estimation to a certain extent. However, traditional machine learning models have limitations when processing language and text data, and it is difficult to integrate unstructured data from industry reports, policies, etc. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method, device, electronic device and medium for predicting the number of energy storage cabinets, so as to solve the above-mentioned problems existing in the prior art and integrate multiple types of data to improve the accuracy and efficiency of measurement.

[0006] In a first aspect, a method for predicting the number of energy storage cabinets is provided, and the method may include:

[0007] Obtain multiple historical electricity bills of the enterprise to be analyzed for the first time period and multiple load data files for the second time period;

[0008] For any historical electricity bill, extract the data in the historical electricity bill to obtain key data;

[0009] For any load data file of the second historical period, extract the data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period;

[0010] The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

[0011] In a possible implementation, the key data includes: electricity price fluctuation data, peak electricity prices, valley electricity prices, electricity consumption time periods, and electricity consumption in each electricity consumption time period.

[0012] In a possible implementation, before obtaining the key data, the method further includes:

[0013] If there is missing electricity price fluctuation data and / or peak electricity prices and / or valley electricity prices in the historical electricity bills of the first historical period, a preset supplement mechanism is used to supplement the missing electricity price fluctuation data and / or peak electricity prices and / or valley electricity prices to obtain the key data.

[0014] In a possible implementation, a trained Seq2Seq model is used to parse data of multiple historical electricity bills to obtain multiple parsed historical electricity bills;

[0015] A well-trained large language model is used to extract data from multiple converted historical electricity bills.

[0016] In a possible implementation, after obtaining the predicted number of energy storage cabinets required by the enterprise to be analyzed,

[0017] The method further comprises:

[0018] An investment return prediction scheme for each energy storage cabinet is determined based on the multiple key data, the multiple load data, the predicted number of energy storage cabinets and the reward mechanism.

[0019] In a possible implementation, after obtaining the predicted number of energy storage cabinets required by the enterprise to be analyzed, the method further includes:

[0020] The predicted return income in future years is determined based on the multiple key data, the multiple load data and the predicted number of energy storage cabinets.

[0021] In a possible implementation, a trained enhanced financial model is used to process a plurality of key data corresponding to a plurality of historical electricity bills and a plurality of load data corresponding to a plurality of load data files of the second time period, including:

[0022] After storing a plurality of key data corresponding to a plurality of historical electricity bills and a plurality of load data corresponding to a plurality of load data files of the second time period in a configured knowledge graph-based database, a data relationship between each key data and the corresponding load data is obtained;

[0023] The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills, multiple load data corresponding to multiple load data files of the second time period, and the data relationship between each key data and the corresponding load data.

[0024] In a second aspect, a device for predicting the number of energy storage cabinets is provided, which may include:

[0025] An acquisition unit, used for acquiring a plurality of historical electricity bills of the enterprise to be analyzed for a first period and a plurality of load data files for a second period;

[0026] An extraction unit, for extracting data from any historical electricity bill to obtain key data;

[0027] And, for any load data file of the second historical period, extracting data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period;

[0028] The processing unit is used to use the trained enhanced financial model to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

[0029] In a third aspect, an electronic device is provided, the electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0030] Memory, used to store computer programs;

[0031] The processor is used to implement any method step described in the first aspect when executing the program stored in the memory.

[0032] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any method step described in the first aspect is implemented.

[0033] The present application provides a method for predicting the number of energy storage cabinets, the method comprising: obtaining multiple historical electricity bills of a to-be-analyzed enterprise in a first period and multiple load data files of a second period; for any historical electricity bill, extracting data in the historical electricity bill to obtain key data; for any load data file of a second historical period, extracting data in the load data file to obtain load data; the load data represents the total electric energy consumed by the to-be-analyzed enterprise in the second historical period; using a trained enhanced financial model, processing multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second period, and obtaining the predicted number of energy storage cabinets required by the to-be-analyzed enterprise. The prior art mainly relies on static, structured data input and manual analysis, which is difficult to cope with the changing market environment and complex project requirements. The advanced natural language processing technology, neural network architecture and adaptive optimization algorithm introduced in the present application have been effectively solved, thereby significantly improving the accuracy and efficiency of the measurement. Through a large language model, the use of natural language processing technology can efficiently parse and integrate various forms of data, thereby improving the accuracy and depth of investment measurement. The energy storage investment estimation robot based on the large language model can dynamically analyze market changes and automatically adjust the parameters of the estimation model to provide more flexible and accurate investment advice. Through the solution provided by this application, the estimation robot can guide users to input necessary information and provide intelligent suggestions through a simple natural language interactive interface, which greatly reduces the usage threshold and allows more users to easily perform investment estimation. The shortcomings of the prior art have been significantly improved through the technical means of this application, providing a more intelligent and flexible solution for investment estimation in the energy storage industry, thereby improving the overall investment decision-making efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application 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.

[0035] Figure 1 A system architecture diagram for a method for predicting the number of energy storage cabinets provided in an embodiment of the present application;

[0036] Figure 2 A schematic diagram of a flow chart of a method for predicting the number of energy storage cabinets provided in an embodiment of the present application;

[0037] Figure 3 A schematic diagram of the structure of a device for predicting the number of energy storage cabinets provided in an embodiment of the present application;

[0038] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0040] The method for predicting the number of energy storage cabinets provided in the embodiment of the present application can be applied to Figure 1 In the system architecture shown in Figure 1 As shown, the system may include: a server and a terminal. The server may be a physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and basic cloud computing services such as big data and artificial intelligence platforms. The terminal may be a user equipment (UE) such as a mobile phone, smart phone, laptop, digital broadcast receiver, personal digital assistant (PDA), tablet computer (PAD), handheld device, vehicle-mounted device, wearable device, computing device or other processing device connected to a wireless modem, mobile station (MS), mobile terminal, etc. The terminal and the server may be directly or indirectly connected via wired or wireless communication, which is not limited in this application.

[0041] The terminal is used to obtain multiple historical electricity bills of the enterprise to be analyzed for the first time period and multiple load data files of the second time period, and send the multiple historical electricity bills of the enterprise to be analyzed for the first time period and multiple load data files of the second time period to the server.

[0042] The server is used to receive a plurality of historical electricity bills of a first time period and a plurality of load data files of a second time period of an enterprise to be analyzed, so as to execute a method for predicting the number of energy storage cabinets provided in the present application.

[0043] In the process of energy storage industry investment estimation, estimation robots based on large language models (LLM) and deep neural networks have shown significant advantages over traditional methods. Existing technologies, such as traditional financial models and dedicated estimation software, can provide basic investment analysis support, but they have the following obvious deficiencies in data processing capabilities, flexibility, and intelligence:

[0044] First, existing technologies have limitations in data integration capabilities. Traditional financial models and dedicated calculation software mainly rely on structured data, such as financial statements and market data, and it is difficult to effectively integrate unstructured data from industry reports, policies, etc. Since the energy storage industry involves many and complex variables, the inability to fully utilize these unstructured data will limit the accuracy and comprehensiveness of the calculation results.

[0045] Secondly, existing calculation software lacks flexibility. These software are usually designed for specific types of projects and are difficult to flexibly respond to changes in market conditions and the unique needs of different projects. When using these tools, investors often need to manually adjust parameters, and the adjustment range is limited, making it difficult to obtain the optimal investment strategy in a complex market environment.

[0046] Finally, existing technologies have high professional requirements for users. Using traditional financial models and calculation software often requires users to have certain financial and industry knowledge, which is a threshold for non-professionals and limits the popularity of the tools.

[0047] Therefore, the present application provides a method for predicting the number of energy storage cabinets to solve the above-mentioned problems existing in the prior art, and can integrate multiple types of data to improve the accuracy and efficiency of measurement.

[0048] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application may be combined with each other if there is no conflict.

[0049] Figure 2 A flowchart of a method for predicting the number of energy storage cabinets provided in an embodiment of the present application. Figure 2 As shown, the method may include:

[0050] Step S210: extract the data in the acquired historical electricity bill and load data file respectively to obtain key data and load data.

[0051] Specifically, step S211, obtain multiple historical electricity bills of the enterprise to be analyzed in the first time period and multiple load data files of the second time period; the second time period is a sub-time period in the first time period. For example: obtain multiple electricity bills of enterprise A in the past year and load data files of every hour on any day in the past year.

[0052] Afterwards, the trained Seq2Seq model including encoder and decoder is used to perform structured conversion on multiple historical electricity bills and load data files to obtain multiple converted historical electricity bills and multiple load data files. In short, since the formats of electricity bills in different provinces are different, it is necessary to convert the format of the historical electricity bills of the enterprises to be analyzed into electricity bills in the unified Josn data format to facilitate subsequent data analysis.

[0053] Step S212: for any historical electricity bill, use a trained large language model based on the attention mechanism to extract the data in the converted historical electricity bill to obtain key data; wherein the key data includes: electricity price fluctuation data, peak electricity price, valley electricity price, electricity consumption time period and electricity consumption in each electricity consumption time period.

[0054] This approach significantly improves the comprehensiveness, flexibility, accuracy and user experience of investment calculations by combining a large language model with the energy storage industry knowledge base. The large language model not only uses a bidirectional long short-term memory network (BiLSTM) but also incorporates the positional encoding technology in the Transformer to process and parse unstructured text data, such as industry reports. These technical innovations are difficult to achieve with existing technologies.

[0055] In some embodiments, before obtaining the key data, the method further includes:

[0056] If there is a lack of electricity price fluctuation data and / or peak electricity price and / or valley electricity price in the historical electricity bill of the first historical period, a preset supplement mechanism is used to supplement the missing electricity price fluctuation data and / or peak electricity price and / or valley electricity price to obtain key data. That is, a recurrent unit (Gated Recurrent Unit, GRU) with a gated mechanism is used to extract and correct electricity price data from the historical electricity bill to ensure data quality.

[0057] The preset supplement mechanism can be understood as: since a historical electricity bill may not show all the key data, for example, the peak electricity price is missing; at this time, it is determined whether there is a peak electricity price in other historical electricity bills, and the peak electricity prices in other historical electricity bills are supplemented to the missing peak electricity prices to make the key data in the historical electricity bill complete.

[0058] and, for any load data file of the second historical period, extracting data from the load data file to obtain load data;

[0059] The load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period.

[0060] In some embodiments, a graph convolutional network (GCN)-based algorithm is used to determine whether load data exists, that is, to determine whether the data in the load file data is an outlier: if an outlier exists, it is determined that load data has not been extracted, and only key data is subsequently processed. If no outlier exists, it is determined that load data has been extracted, and key data and load data are subsequently processed simultaneously.

[0061] Step S220: store the load data and key data in a configured knowledge graph-based database to obtain the data relationship between each key data and the corresponding load data.

[0062] Among them, Neo4j can be used to configure a knowledge graph-based database.

[0063] Specifically, the server reads the data relationship between each key data and the corresponding load data from the energy storage measurement database by using a distributed deep learning framework (such as Horovod).

[0064] Step S230: Using the trained enhanced financial model, multiple key data corresponding to multiple historical electricity bills, multiple load data corresponding to multiple load data files of the second time period, and the data relationship between each key data and the corresponding load data are processed to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

[0065] Specifically, the trained enhanced financial model is adopted and time series analysis methods (such as ARIMA, LSTM, etc.) are used to predict future load data to obtain predicted load data.

[0066] This step is to better understand future electricity demand and plan the size of the energy storage cabinet accordingly.

[0067] Similarly, if the electricity price has obvious peak and valley characteristics, the same method can be used to predict the electricity price, which helps to optimize the charging and discharging strategy of the energy storage cabinet.

[0068] According to the time period to be predicted, future time points are generated and corresponding features (such as hours, days of the week, months, etc.) are extracted.

[0069] The generated future time series data is input into the trained enhanced financial model to obtain the future load forecast value.

[0070] The predicted future load values ​​are accumulated to obtain the total load demand.

[0071] In order to cope with possible emergencies, a certain safety margin is usually added to the total demand.

[0072] Assuming that the capacity of each energy storage cabinet is C (unit: kWh), the number of energy storage cabinets N required can be calculated by the following formula:

[0073] N = (total load demand + safety margin) / C

[0074] In some embodiments, a return on investment prediction scheme for each energy storage cabinet is determined based on a plurality of key data, a plurality of load data, the predicted number of energy storage cabinets, and an incentive mechanism.

[0075] Among them, the reward mechanism can be understood as a reward for participating in a demand response (DR) program; for example, when a neighboring enterprise of the enterprise to be analyzed needs electricity, the enterprise to be analyzed distributes electricity from the energy storage cabinet to the neighboring enterprise and receives a corresponding reward.

[0076] Specifically, assuming that the total cost of each energy storage cabinet is C and the number of energy storage cabinets required is predicted to be N, the initial investment I0 is:

[0077] I0=C×N

[0078] Assume that the daily peak electricity price is P peak , the valley electricity price is P valley , the daily average load is L, the daily charge and discharge capacity of the energy storage cabinet is E, then the daily electricity cost S saved is:

[0079] S=(P peak -P valley )×E

[0080] Annual electricity savings year for:

[0081] S year =S×365

[0082] Assume that the annual demand response reward is R, the annual energy conservation and emission reduction reward is G, and the annual operating cost is O;

[0083] The total annual revenue T is:

[0084] T=S year +R+G

[0085] The annual net income W is:

[0086] W=TO

[0087] That is, based on the plurality of key data, the plurality of load data and the predicted number of energy storage cabinets, the predicted return income (annual net income) in future years is determined.

[0088] Afterwards, the natural language generation (NLG) technology can be used to automatically generate a project plan document based on the predicted return, the number of energy storage cabinets required, and the return on investment prediction plan.

[0089] This application uses an integrated Adaptive Bayesian Optimization algorithm to dynamically adjust the calculation parameters based on real-time data and automatically generate investment plans, reducing manual intervention and improving efficiency. Through a simplified operating interface and intelligent suggestions, the system greatly reduces the threshold for use, allowing non-professional users to easily perform investment calculations.

[0090] The present application provides a method for predicting the number of energy storage cabinets, the method comprising: obtaining multiple historical electricity bills of a to-be-analyzed enterprise in a first period and multiple load data files of a second period; for any historical electricity bill, extracting data in the historical electricity bill to obtain key data; for any load data file of a second historical period, extracting data in the load data file to obtain load data; the load data represents the total electric energy consumed by the to-be-analyzed enterprise in the second historical period; using a trained enhanced financial model, processing multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second period, and obtaining the predicted number of energy storage cabinets required by the to-be-analyzed enterprise. The prior art mainly relies on static, structured data input and manual analysis, which is difficult to cope with the changing market environment and complex project requirements. The advanced natural language processing technology, neural network architecture and adaptive optimization algorithm introduced in the present application have been effectively solved, thereby significantly improving the accuracy and efficiency of the measurement. Through a large language model, the use of natural language processing technology can efficiently parse and integrate various forms of data, thereby improving the accuracy and depth of investment measurement. The energy storage investment estimation robot based on the large language model can dynamically analyze market changes and automatically adjust the parameters of the estimation model to provide more flexible and accurate investment advice. Through the solution provided by this application, the estimation robot can guide users to input necessary information and provide intelligent suggestions through a simple natural language interactive interface, which greatly reduces the usage threshold and allows more users to easily perform investment estimation. The shortcomings of the prior art have been significantly improved through the technical means of this application, providing a more intelligent and flexible solution for investment estimation in the energy storage industry, thereby improving the overall investment decision-making efficiency and accuracy.

[0091] This application achieves more intelligent and automated investment measurement by introducing the Multi-Head Attention mechanism and Dynamic Convolutional Neural Network (DCNN) under the Transformer architecture. It can process various data forms including unstructured data and dynamically adapt to different investment scenarios through time series forecasting (Time-Series Forecasting).

[0092] By integrating the energy storage industry knowledge base with the Large Language Model (LLM), the accuracy and efficiency of investment estimation are significantly improved. Compared with the prior art, this application can not only efficiently process structured data, but also use natural language processing (NLP) technology to deeply analyze unstructured data such as industry reports. By introducing the multi-head attention mechanism (Multi-Head Attention) and bidirectional long short-term memory network (BiLSTM) in the Transformer architecture, the system can extract key information in a complex context to ensure the comprehensiveness and accuracy of the estimation results. In addition, the application of technology based on conditional generative adversarial networks (Conditional GAN) and positional encoding (Positional Encoding) makes the estimation process more automated and intelligent, reduces the need for human intervention, significantly lowers the threshold for use, and enables non-professional users to easily perform complex investment estimation. Through the integration of the adaptive Bayesian optimization algorithm (Adaptive Bayesian Optimization), the system can adjust the estimation parameters in real time to cope with the changing investment needs of the energy storage industry, ensure the reliability and adaptability of the estimation results, and provide users with strong decision support.

[0093] Corresponding to the above method, the embodiment of the present application also provides a device for predicting the number of energy storage cabinets, such as Figure 3 As shown, the device comprises:

[0094] An acquisition unit 310 is used to acquire a plurality of historical electricity bills of the enterprise to be analyzed for a first time period and a plurality of load data files for a second time period;

[0095] The extraction unit 320 is used to extract data from any historical electricity bill to obtain key data;

[0096] And, for any load data file of the second historical period, extracting data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period;

[0097] The processing unit 330 is used to use the trained enhanced financial model to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

[0098] The functions of the various functional units of the device for predicting the number of energy storage cabinets provided in the above-mentioned embodiment of the present application can be realized through the above-mentioned method steps. Therefore, the specific working process and beneficial effects of each unit in the device for predicting the number of energy storage cabinets provided in the embodiment of the present application are not repeated here.

[0099] The present application also provides an electronic device, such as Figure 4 As shown, it includes a processor 410 , a communication interface 420 , a memory 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other via the communication bus 440 .

[0100] Memory 430, for storing computer programs;

[0101] The processor 410 is used to execute the program stored in the memory 430 to implement the following steps:

[0102] Obtain multiple historical electricity bills of the enterprise to be analyzed for the first time period and multiple load data files for the second time period;

[0103] For any historical electricity bill, extract the data in the historical electricity bill to obtain key data;

[0104] For any load data file of the second historical period, extract the data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period;

[0105] The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

[0106] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0107] The communication interface is used for communication between the above electronic device and other devices.

[0108] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0109] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0110] The implementation methods and beneficial effects of the components of the electronic device in the above embodiments to solve the problems can be seen in Figure 2 The various steps in the illustrated embodiment are implemented, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiment of the present application are not repeated here.

[0111] In another embodiment provided in the present application, a computer-readable storage medium is provided, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes a method for predicting the number of energy storage cabinets described in any of the above embodiments.

[0112] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute a method for predicting the number of energy storage cabinets described in any one of the above embodiments.

[0113] Those skilled in the art will appreciate that the embodiments in the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt a complete hardware embodiment, a complete software embodiment, or a form of an embodiment combining software and hardware. Moreover, the present application may adopt a form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0117] Unless otherwise defined, the technical terms or scientific terms used in this application should be understood by people with ordinary skills in the field to which the present invention belongs. "First", "second" and similar words used in this application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect", "couple" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0118] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the present application embodiments are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application embodiments.

[0119] Obviously, those skilled in the art can make various changes and modifications to the embodiments in the present application without departing from the spirit and scope of the embodiments in the present application. Thus, if these modifications and variations of the embodiments in the present application are within the scope of the embodiments in the present application and their equivalents, the embodiments in the present application are also intended to include these modifications and variations.

Claims

1. A method for predicting the number of energy storage cabinets, characterized in that: The method comprises: Obtain multiple historical electricity bills of the enterprise to be analyzed for the first time period and multiple load data files for the second time period; For any historical electricity bill, extract the data in the historical electricity bill to obtain key data; For any load data file of the second historical period, extract the data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period; The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

2. The method according to claim 1, characterized in that The key data include: electricity price fluctuation data, peak electricity prices, valley electricity prices, electricity consumption time periods and electricity consumption in each electricity consumption time period.

3. The method according to claim 2, characterized in that Before obtaining the key data, the method further includes: If there is missing electricity price fluctuation data and / or peak electricity prices and / or valley electricity prices in the historical electricity bills of the first historical period, a preset supplement mechanism is used to supplement the missing electricity price fluctuation data and / or peak electricity prices and / or valley electricity prices to obtain the key data.

4. The method according to claim 1, characterized in that Before extracting the data in the historical electricity bill, the method further includes: The trained Seq2Seq model is used to parse data of multiple historical electricity bills to obtain multiple parsed historical electricity bills. A well-trained large language model is used to extract data from multiple converted historical electricity bills.

5. The method according to claim 1, characterized in that After obtaining the predicted number of energy storage cabinets required by the enterprise to be analyzed, the method further includes: An investment return prediction scheme for each energy storage cabinet is determined based on the multiple key data, the multiple load data, the predicted number of energy storage cabinets and the reward mechanism.

6. The method according to claim 1, characterized in that After obtaining the predicted number of energy storage cabinets required by the enterprise to be analyzed, the method further includes: The predicted return income in future years is determined based on the multiple key data, the multiple load data and the predicted number of energy storage cabinets.

7. The method according to claim 1, characterized in that The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple second-time load data files, including: After storing a plurality of key data corresponding to a plurality of historical electricity bills and a plurality of load data corresponding to a plurality of load data files of the second time period in a configured knowledge graph-based database, a data relationship between each key data and the corresponding load data is obtained; The trained enhanced financial model is used to process multiple key data corresponding to multiple historical electricity bills, multiple load data corresponding to multiple load data files of the second time period, and the data relationship between each key data and the corresponding load data.

8. A device for predicting the number of energy storage cabinets, characterized in that: The device comprises: An acquisition unit, used for acquiring a plurality of historical electricity bills of the enterprise to be analyzed for a first period and a plurality of load data files for a second period; An extraction unit, for extracting data from any historical electricity bill to obtain key data; And, for any load data file of the second historical period, extracting data in the load data file to obtain load data; the load data represents the total electric energy consumed by the enterprise to be analyzed in the second historical period; The processing unit is used to use the trained enhanced financial model to process multiple key data corresponding to multiple historical electricity bills and multiple load data corresponding to multiple load data files of the second time period to obtain the predicted number of energy storage cabinets required by the enterprise to be analyzed.

9. An electronic device, characterized in that: The electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 7 when executing a program stored in a memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in any one of claims 1 to 7 are implemented.

Images