Charging and discharging facility power distribution management system and method based on power pooling

Abstract

The invention discloses a charging and discharging facility power distribution management system and method based on power pooling, and relates to the technical field of charging and discharging management, and the method comprises the steps: collecting the operation data of a plurality of distributed power supplies and a plurality of charging and discharging facilities in a target region; obtaining multiple pieces of power supply operation scene information; obtaining a plurality of log iteration associated memories; determining a plurality of distributed power supply operation coefficients; obtaining distributable output power; carrying out charging and discharging demand extraction on the plurality of charging and discharging facility operation logs to obtain a plurality of pieces of charging and discharging demand information; and carrying out power distribution based on the plurality of pieces of charging and discharging demand information, the plurality of demand direction identifiers and the distributable output power to obtain a power distribution scheme, and carrying out power distribution management. The technical problem that in the prior art, the hidden operation condition of the power supply is inaccurately mastered, so that the power distribution reliability of the charging and discharging facility is low is solved, and the technical effect of improving the power distribution management accuracy is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of charge and discharge management, and in particular to a power distribution management system and method for charge and discharge facilities based on power pooling. Background Art

[0002] At present, power system management often only grasps the current power situation that can be provided, and cannot accurately capture the real-time power demand changes and supply capabilities between distributed energy units, which may lead to unbalanced power distribution and affect the stability and reliability of the entire system. The existing technology has the technical problem of inaccurate grasp of the implicit operation status of the power supply, resulting in low reliability of power distribution of charging and discharging facilities. Summary of the invention

[0003] The present application provides a power distribution management system and method for charging and discharging facilities based on power pooling, which is used to solve the technical problem in the prior art that the implicit operating conditions of the power supply are not accurately understood, resulting in low reliability of power distribution for charging and discharging facilities.

[0004] In view of the above problems, the present application provides a power distribution management system and method for charging and discharging facilities based on power pooling.

[0005] In a first aspect of the present application, a method for managing power allocation of charging and discharging facilities based on power pooling is provided, the method comprising:

[0006] Collecting operation data of multiple distributed power sources and multiple charging and discharging facilities in the target area, and obtaining multiple distributed power source operation logs and multiple charging and discharging facility operation logs;

[0007] Using an encoder to perform semantic analysis on the multiple distributed power supply operation logs to obtain multiple power supply operation scenario information;

[0008] Inputting the multiple power supply operation scenario information into the information association module, performing interactive iterative capture of associated words on the multiple distributed power supply operation logs, and obtaining multiple log iterative association memories;

[0009] Calling a decoder to identify the power supply operation status of the multiple power supply operation scenario information and the multiple log iteration association memories respectively, and determining multiple distributed power supply operation coefficients;

[0010] Collecting multiple powers of multiple distributed power sources at the current moment, and performing weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain allocable output power;

[0011] Extracting charging and discharging demands from the operation logs of the multiple charging and discharging facilities to obtain multiple charging and discharging demand information, wherein the multiple charging and discharging demand information has multiple demand direction identifiers;

[0012] Power allocation is performed based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power to obtain a power allocation plan, and power allocation management is performed on the charging and discharging facilities in the target area based on the power allocation plan.

[0013] A second aspect of the present application provides a power distribution management system for charging and discharging facilities based on power pooling, the system comprising:

[0014] An operation log acquisition module is used to collect operation data of multiple distributed power sources and multiple charging and discharging facilities in a target area, and obtain multiple distributed power source operation logs and multiple charging and discharging facility operation logs;

[0015] An operation scenario information acquisition module, used to perform semantic analysis on the multiple distributed power supply operation logs using an encoder to obtain multiple power supply operation scenario information;

[0016] A log iterative association memory acquisition module, used to input the multiple power supply operation scenario information into the information association module, perform interactive iterative capture of association words on the multiple distributed power supply operation logs, and obtain multiple log iterative association memories;

[0017] A distributed power supply operation coefficient determination module is used to call a decoder to identify the power supply operation status of the multiple power supply operation scenario information and the multiple log iteration association memories respectively, and determine multiple distributed power supply operation coefficients;

[0018] The allocable output power acquisition module is used to collect multiple powers of multiple distributed power sources at the current moment, and perform weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain the allocable output power;

[0019] A charging and discharging demand information obtaining module, used to extract charging and discharging demand from the operation logs of the plurality of charging and discharging facilities, and obtain a plurality of charging and discharging demand information, wherein the plurality of charging and discharging demand information has a plurality of demand direction identifiers;

[0020] The power allocation management module is used to allocate power based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power, obtain a power allocation plan, and manage power allocation for the charging and discharging facilities in the target area based on the power allocation plan.

[0021] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0022] The present application obtains multiple distributed power supply operation logs and multiple charging and discharging facility operation logs by collecting the operation data of multiple distributed power supply and multiple charging and discharging facilities in the target area, and then uses the encoder to perform semantic analysis on the multiple distributed power supply operation logs to obtain multiple power supply operation scenario information, and then inputs the multiple power supply operation scenario information into the information association module, performs interactive iterative capture of the associated words on the multiple distributed power supply operation logs, obtains multiple log iterative association memories, and identifies the power supply operation status of the multiple power supply operation scenario information and the multiple log iterative association memories by calling the decoder, determines the multiple distributed power supply operation coefficients, and then collects multiple powers of the multiple distributed power supplies at the current moment, and uses the multiple distributed power supply operation coefficients to perform weighted calculation on the multiple powers to obtain the allocable output power, and then extracts the charging and discharging demand from the multiple charging and discharging facility operation logs to obtain multiple charging and discharging demand information, wherein the multiple charging and discharging demand information has multiple demand direction identifiers, and then performs power allocation based on the multiple charging and discharging demand information, the multiple demand direction identifiers and the allocable output power to obtain the power allocation plan, and performs power allocation management on the charging and discharging facilities in the target area based on the power allocation plan. The technical effect of improving the reliability of power allocation management of charging and discharging facilities is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A schematic diagram of a method for managing power allocation of charging and discharging facilities based on power pooling provided in an embodiment of the present application;

[0025] Figure 2 A schematic diagram of the structure of a power distribution management system for charging and discharging facilities based on power pooling provided in an embodiment of the present application.

[0026] Explanation of the accompanying drawings: operation log acquisition module 11, operation scenario information acquisition module 12, log iteration association memory acquisition module 13, distributed power supply operation coefficient determination module 14, allocable output power acquisition module 15, charge and discharge demand information acquisition module 16, power allocation management module 17. DETAILED DESCRIPTION

[0027] The present application provides a power distribution management system and method for charging and discharging facilities based on power pooling, which is used to solve the technical problem in the prior art that the implicit operating conditions of the power supply are not accurately understood, resulting in low reliability of power distribution for charging and discharging facilities.

[0028] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings 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 in 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.

[0029] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment 1

[0031] like Figure 1 As shown, the present application provides a power allocation management method for charging and discharging facilities based on power pooling, wherein the method includes:

[0032] S100: Collecting operation data of multiple distributed power sources and multiple charging and discharging facilities in a target area, and obtaining multiple distributed power source operation logs and multiple charging and discharging facility operation logs;

[0033] In a possible embodiment, the target area is any area determined by a person skilled in the art, which may be an industrial park, a residential area, etc. The distributed power source refers to small power generation equipment distributed in different locations, usually close to the power consumption point, such as solar panels, wind turbines or small gas generators. They can work alone or in collaboration to provide local or overall power supply by being integrated into the power network. The charging and discharging facilities specifically refer to equipment used in energy storage systems, which can store electrical energy for later use (charging) or release the stored electrical energy to the power grid to meet power needs (discharging), such as the battery system of an electric vehicle.

[0034] Optionally, the operation data of the distributed power sources, including information such as power generation, power output, and operation status, are collected through on-site sensors, remote monitoring systems, or smart meters, so as to obtain multiple distributed power source operation logs of the multiple distributed power sources. The multiple distributed power source operation logs reflect the power fluctuations of the multiple distributed power sources.

[0035] The multiple charging and discharging facility operation logs are operation data of the multiple charging and discharging facilities obtained by using the built-in data recording device or monitoring software of the energy storage system, including information such as battery charging and discharging status, charging efficiency, and system health status.

[0036] By obtaining the multiple distributed power supply operation logs and the multiple charging and discharging facility operation logs, the technical effect of providing data support for subsequent reliable charging and discharging facility power allocation management is achieved.

[0037] S200: using an encoder to perform semantic analysis on the multiple distributed power supply operation logs to obtain multiple power supply operation scenario information;

[0038] Furthermore, step S200 of the embodiment of the present application further includes:

[0039] Performing data preprocessing on the multiple distributed power supply operation logs, and when the data preprocessing is completed, performing vector conversion on the text data to obtain multiple feature vector sets;

[0040] Building encoders based on recurrent neural networks;

[0041] The encoder's multi-layer neural network structure is used to perform feature recognition on the multiple feature vector sets to determine the multiple power supply operation scenario information.

[0042] In one embodiment, an encoder is used to perform semantic analysis on the multiple distributed power supply operation logs to analyze the overall operation status of the multiple distributed power supplies, thereby obtaining multiple power supply operation scenario information that can describe the operation status of the distributed power supplies. For example, the power supply operation scenario information can be information describing the operation fluctuation degree, output power increase and decrease of the distributed power supplies.

[0043] In one possible embodiment, the encoder is a neural network structure used to convert input sequences (such as text, time series, etc.) into continuous representations (i.e., feature vectors) in machine learning and natural language processing for subsequent analysis and processing. Data preprocessing operations are completed by cleaning, converting, and normalizing the original data to facilitate subsequent feature extraction and model training. Multiple feature vector sets reflect important features in the data and provide data support for subsequent analysis.

[0044] Optionally, this method usually uses a pre-trained word embedding model (such as Word2Vec, GloVe, etc.) or an embedding representation method automatically learned through training data (such as BERT, ELMo, etc.) to map each word or phrase in the text data into a high-dimensional vector space to obtain the multiple feature vector sets.

[0045] Optionally, the recurrent neural network (RNN) is a neural network structure with feedback connections, which is particularly suitable for processing sequence data and can capture the time dependency in the sequence. Multiple sample feature vector sets and multiple sample power supply operation scenario information are obtained as training data, and supervised training is performed on the encoder built based on the recurrent neural network to learn the one-to-one mapping relationship between the feature vector set and the power supply operation scenario information until the model converges and the trained encoder is obtained. The trained encoder is used to perform feature recognition on the multiple feature vector sets, and after analysis, the multiple power supply operation scenario information is obtained. The goal of effectively converting the original distributed power supply operation log into semantically rich scenario information is achieved, and the technical effect of providing necessary data support for the subsequent power distribution management system is achieved.

[0046] S300: Inputting the plurality of power supply operation scenario information into an information association module, performing interactive iterative capture of associated words on the plurality of distributed power supply operation logs, and obtaining iterative association memory of a plurality of logs;

[0047] Furthermore, step S300 of the embodiment of the present application further includes:

[0048] Taking time as an index, performing log serialization processing on the multiple distributed power supply operation logs to obtain multiple distributed power supply operation record sequences;

[0049] Extracting a plurality of first distributed power supply operation records that are located at the first position from the plurality of distributed power supply operation record sequences;

[0050] Extracting keywords from the plurality of power supply operation scenario information to obtain a plurality of scenario keyword sets, and storing the plurality of scenario keyword sets into a plurality of first memory vectors;

[0051] Extracting keywords from the plurality of first distributed power supply operation records to obtain a plurality of first keyword sets;

[0052] The multiple first memory vectors are interactively iteratively updated based on the multiple first keyword sets until the iterative capture of the multiple distributed power supply operation record sequences is completed, thereby obtaining the multiple log iterative associated memories.

[0053] Furthermore, step S300 of the embodiment of the present application further includes:

[0054] Capturing associated words based on the multiple first keyword sets and the multiple scene keyword sets, and adding the multiple first capture results into the multiple first memory vectors respectively to obtain multiple second memory vectors;

[0055] Extracting a plurality of second distributed power supply operation records located at the second position from the plurality of distributed power supply operation record sequences;

[0056] Extracting keywords from the plurality of second distributed power supply operation records to obtain a plurality of second keyword sets;

[0057] Capturing associated words based on the multiple second keyword sets, the multiple first keyword sets, the multiple scene keyword sets, the multiple first memory vectors, and the multiple second memory vectors, and adding the multiple second capture results to the multiple second memory vectors, respectively, to obtain multiple third memory vectors;

[0058] Based on the multiple third memory vectors and the multiple distributed power supply operation record sequences, associated words are interactively iterated and captured, and the multiple Mth memory vectors finally obtained are used as multiple log iterative associated memories, where M is the number of distributed power supply operation records in the distributed power supply operation record sequence.

[0059] In one possible embodiment, the information association module is used to capture words with associated information between two distributed power supply operation records to obtain associated words. By inputting multiple power supply operation scenario information into the information association module, the associated words are interactively iteratively captured to obtain multiple log iterative associated memories. The multiple log iterative associated memories reflect the changes in the operating status of multiple distributed power supplies. By obtaining the multiple log iterative associated memories, the goal of analyzing and mastering the overall status and operating trends of multiple distributed power supplies is achieved.

[0060] In a possible embodiment, the plurality of distributed power supply operation logs are serialized in a time-ordered order with time as an index to obtain a plurality of distributed power supply operation record sequences. The plurality of distributed power supply operation record sequences reflect the change of the operation status of the plurality of distributed power supplies over time.

[0061] Optionally, multiple first distributed power supply operation records located in the first position are extracted from the multiple distributed power supply operation record sequences, and then keyword extraction is performed on the multiple power supply operation scenario information to determine keywords that can reflect the power supply operation status of the multiple power supply operation scenario information, obtain multiple scene keyword sets, and store the multiple scene keyword sets into multiple first memory vectors.

[0062] Optionally, the text contents of the plurality of first distributed power supply operation records are divided into words or phrases, and common stop words are removed to reduce interference with keyword extraction. Further, for all operation records of each distributed power supply, the word frequency of each word in the entire corpus is calculated. In addition, the inverse document frequency of each word is calculated, that is, the logarithm of the inverse of the frequency of occurrence of the word in all operation records.

[0063] Then, the word frequency is multiplied by the inverse document frequency to obtain the TF-IDF value of each word, and the words with the highest TF-IDF values ​​in each distributed power supply operation record are selected as keywords according to the TF-IDF value. Thus, multiple first keyword sets are obtained;

[0064] After obtaining the multiple first keyword sets, the multiple first memory vectors are interactively iteratively updated according to the multiple first keyword sets that can reflect the key features of the multiple first distributed power supply operation records, until the iterative capture of the multiple distributed power supply operation record sequences is completed, and the multiple log iterative associated memories are obtained. Thus, the information association module can effectively capture key information from multiple distributed power supply operation logs and integrate this information into iterative associated memories, achieving the technical effect of helping to understand the overall status and trend of power supply operation.

[0065] In a possible embodiment, the multiple first keyword sets and the multiple scene keyword sets are analyzed to capture associated words of the first keywords associated with the scene keywords, and the multiple first capture results are respectively added to the multiple first memory vectors to obtain multiple second memory vectors.

[0066] Then, extract a plurality of second distributed power supply operation records located in the second position from the plurality of distributed power supply operation record sequences. Based on the same acquisition principle as the plurality of first keyword sets, perform keyword extraction on the plurality of second distributed power supply operation records to obtain a plurality of second keyword sets. Based on the same acquisition principle as the plurality of first capture results, perform associated word capture according to the plurality of second keyword sets, the plurality of first keyword sets, the plurality of scene keyword sets, the plurality of first memory vectors and the plurality of second memory vectors, and respectively add the plurality of second capture results into the plurality of second memory vectors to obtain a plurality of third memory vectors.

[0067] Based on the multiple third memory vectors and the multiple distributed power supply operation record sequences, the associated words are interactively iterated and captured, and the keywords associated with the captured associated words in the distributed power supply operation record sequence are continuously extracted, and the degree of association is gradually deepened, and then the multiple M-th memory vectors finally obtained are used as multiple log iterative associated memories, where M is the number of distributed power supply operation records in the distributed power supply operation record sequence. The goal of deeply understanding the overall operation status and operation trend of distributed power supplies is achieved.

[0068] Furthermore, step S300 of the embodiment of the present application further includes:

[0069] respectively calculating the semantic similarity between any first keyword in the plurality of first keyword sets and the plurality of scene keyword sets to obtain a plurality of first semantic similarity sets;

[0070] respectively extracting the co-occurrence frequency of any first keyword in the plurality of first keyword sets and the scene keywords in the plurality of scene keyword sets to obtain a plurality of first co-occurrence coefficient sets;

[0071] Performing weighted calculation on the multiple first semantic similarity sets and the multiple first co-occurrence coefficient sets to obtain multiple association coefficient sets;

[0072] The first keywords corresponding to the first n correlation coefficients in the plurality of correlation coefficient sets are respectively added to the plurality of first captured results, wherein n is a positive integer.

[0073] In a possible embodiment, any one of the first keywords in the plurality of first keyword sets is extracted, and its semantic similarity with the scene keywords in the corresponding scene keyword set is calculated using the cosine similarity calculation formula, and the maximum value of the semantic similarity calculated is used as the first semantic similarity. Based on the same principle, the semantic similarity of the plurality of first keyword sets and the plurality of scene keyword sets is calculated to obtain a plurality of first semantic similarity sets. The plurality of first semantic similarity sets reflect the importance of the first keywords in the plurality of first keyword sets.

[0074] In one embodiment, the frequencies of co-occurrence of any first keyword in the plurality of first keyword sets and scene keywords in the plurality of scene keyword sets are counted to obtain a plurality of first co-occurrence coefficient sets. The larger the first co-occurrence coefficient, the more important the first keyword.

[0075] Optionally, the plurality of first semantic similarity sets and the plurality of first co-occurrence coefficient sets are weightedly calculated according to weights pre-set by a person skilled in the art to obtain the plurality of correlation coefficient sets. The larger the correlation coefficient, the more important the corresponding first keyword is for the state trend analysis of the distributed power supply.

[0076] By sorting the multiple association coefficient sets in descending order, and then adding the first keywords corresponding to the first n association coefficients to the multiple first capture results, where n is a positive integer, the goal of identifying and extracting keywords associated with the multiple scene keyword sets in the multiple first keyword sets is achieved, and the technical effect of improving the accuracy of capturing associated words is achieved.

[0077] S400: calling a decoder to identify power supply operation states of the plurality of power supply operation scenario information and the plurality of log iteration association memories respectively, and determining a plurality of distributed power supply operation coefficients;

[0078] Furthermore, step S400 in the embodiment of the present application further includes:

[0079] Acquire multiple sample power supply operation scenario information, multiple sample log iterative association memory, and multiple sample distributed power supply operation coefficients as training data;

[0080] The training data is used to train the recurrent neural network to learn the mapping relationship between the power supply operation scenario information and the log iteration associated memory and the distributed power supply operation coefficients until convergence, thereby obtaining the trained decoder.

[0081] In one possible embodiment, the main task of the decoder in step S400 is to identify the power supply operation status of multiple power supply operation scenario information and multiple log iteration association memories, analyze the overall operation status and operation trend of multiple distributed power supplies, and thus determine the operation coefficients of multiple distributed power supplies. These operation coefficients are an important basis for subsequent power allocation because they reflect the operation status and capabilities of each distributed power supply in the current or future period of time. The larger the operation coefficient, the greater the output power of the distributed power supply and the higher the stability.

[0082] To achieve this goal, the system first needs a trained decoder. This decoder is obtained by learning a large amount of sample data, which includes multiple sample power supply operation scenario information, multiple sample log iterative association memory, and their corresponding multiple sample distributed power supply operation coefficients. These sample data constitute the training data set for training the recurrent neural network (RNN).

[0083] By collecting and organizing a large amount of sample data, including log records of power supply in different operating scenarios, log information after iterative associative memory processing, and the actual operating coefficients of distributed power supply in these scenarios, these data need to be labeled and organized into a format suitable for neural network training.

[0084] Then, a neural network model capable of processing sequence data is constructed based on recurrent neural network (RNN). This model will serve as a decoder to receive power supply operation scenario information and log iteration association memory as input, and output the operation coefficient of distributed power supply.

[0085] The prepared training data is input into the neural network model, and the difference (i.e., loss) between the model output and the true operating coefficient is calculated through forward propagation. Then, the parameters in the model are adjusted using the backpropagation algorithm to reduce this difference. This process is repeated many times until the performance of the model reaches a certain standard (such as the loss function converges to a lower value), at which point the model is considered to have been trained. During or after training, a portion of independent validation data can be used to evaluate the performance of the model. Based on the evaluation results, the model can be further tuned to improve its accuracy and generalization ability.

[0086] Furthermore, the trained decoder is deployed into the system to process the power supply operation scenario information and log iteration association memory in real time or periodically, and output the operation coefficient of the distributed power supply.

[0087] Through this process, the system can accurately identify the status of distributed power sources in different operating scenarios and determine their operating coefficients accordingly, thus providing a reliable basis for subsequent power allocation and contributing to the technical effect of achieving more reasonable and efficient energy management.

[0088] S500: collecting multiple powers of multiple distributed power sources at the current moment, and performing weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain allocable output power;

[0089] In one possible embodiment, multiple powers of multiple distributed power sources at the current moment are collected, and these powers are weightedly calculated using the operating coefficients of these distributed power sources to determine the distributable output power that meets the actual operating status and operating trend of each distributed power source. This achieves the technical effect of providing reliable data support for subsequent refined energy management and scheduling.

[0090] Optionally, the current output power of all distributed power sources (such as solar photovoltaic panels, wind turbines, energy storage batteries, etc.) in the target area is collected. These power data reflect the actual power generation or discharge capacity of the distributed power sources at the current moment. The collection method may include directly reading the output data of the power supply device, interacting with the power supply device through a communication protocol to obtain data, etc.

[0091] According to the distributed power supply operation coefficient determined in the previous step S400, a weighted calculation is performed on multiple powers. The operation coefficient is a quantitative evaluation of the performance and stability of the distributed power supply in the current operation scenario, which reflects the deviation or adjustment factor between the actual output capacity of the power supply and the ideal output capacity.

[0092] Optionally, the ratio of the operating coefficients of the multiple distributed power sources to the sum of the operating coefficients of the multiple distributed power sources is used as multiple weights. The multiple weights reflect the priority or importance of the multiple distributed power sources relative to other power sources under current conditions. Then, the multiple powers are multiplied by the multiple weights respectively, and the multiplication results are added to obtain the overall output power of the multiple distributed power sources in the target area, that is, the allocable output power.

[0093] The allocable output power obtained through weighted calculation not only reflects the actual power generation or discharge capacity of distributed power sources, but also takes into account their performance and stability in specific operating scenarios, achieving the technical effect of improving the rationality of power allocation management.

[0094] S600: extracting charging and discharging demands from the operation logs of the plurality of charging and discharging facilities to obtain a plurality of charging and discharging demand information, wherein the plurality of charging and discharging demand information has a plurality of demand direction identifiers;

[0095] Furthermore, step S600 in the embodiment of the present application further includes:

[0096] When the demand direction is marked as a positive sign, it indicates that the charging and discharging facilities need to be charged;

[0097] When the demand direction is marked as negative, it indicates that the charging and discharging facilities need to discharge.

[0098] S700: Perform power allocation based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power, obtain a power allocation plan, and perform power allocation management on the charging and discharging facilities in the target area based on the power allocation plan.

[0099] In one possible embodiment, effective power allocation is achieved through steps S600 and S700. In step S600, the system first needs to conduct a detailed analysis of the operation logs of multiple charging and discharging facilities in the target area. These operation logs record the historical usage, current status and possible needs of the charging and discharging facilities. By deeply mining these logs, the system can extract the charging and discharging demand information of each charging and discharging facility. These demand information may include but is not limited to: required power, demand time, priority, etc. While extracting the demand information, the system also assigns a demand direction identifier to each demand. This identifier is used to clearly indicate whether the current demand of the charging and discharging facility is charging or discharging. Specifically, when the demand direction identifier is a positive identifier (such as "+" or "charging"), it indicates that the charging and discharging facility currently needs to be charged to meet its energy needs; and when the demand direction identifier is a negative identifier (such as "-" or "discharging"), it indicates that the facility has surplus electric energy to be released, that is, it needs to be discharged, and the power allocated to the charging and discharging facility can be reduced at this time.

[0100] Optionally, a reasonable power allocation plan is formulated based on the multiple charging and discharging demand information extracted in step S600, the multiple demand direction identifiers, and the allocable output power determined in steps S400 to S500. This plan needs to comprehensively consider multiple factors, including but not limited to: the urgency of the demand of each charging and discharging facility, the demand direction (charging or discharging), the required power, the total amount of allocable output power, and the allocation efficiency.

[0101] Optionally, a power allocation network layer is constructed, and the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power are input into the power allocation network layer for power allocation to obtain the power allocation scheme. The power allocation scheme refers to a plan or strategy for reasonably allocating and scheduling the allocable output power according to the needs of each charging and discharging facility and the overall allocable output power of multiple distributed energy sources in the target area, including the power allocated to each charging and discharging facility.

[0102] After obtaining the power allocation scheme, actual power allocation management is performed on the charging and discharging facilities in the target area according to the scheme. For facilities that need to be charged, the system will ensure that they can obtain sufficient electricity to meet their needs; and for facilities that need to be discharged, they will be guided to release surplus electricity in a reasonable manner. In this way, the system can achieve energy balance between charging and discharging facilities, improve energy utilization, and ensure the stable operation of the entire system. The technical effect of effective power allocation management and scheduling of charging and discharging facilities in the target area, optimizing energy configuration, and improving energy utilization efficiency is achieved.

[0103] In summary, the embodiments of the present application have at least the following technical effects:

[0104] The present application collects the operation data of multiple distributed power sources and multiple charging and discharging facilities in the target area, obtains multiple distributed power source operation logs and multiple charging and discharging facility operation logs, and achieves the goal of mastering the operation status of distributed power sources and charging and discharging facilities in the target area. Then, an encoder is used to perform semantic analysis on the multiple distributed power source operation logs to obtain multiple power source operation scenario information, and then the multiple power source operation scenario information is input into the information association module, and the multiple distributed power source operation logs are interactively iteratively captured to obtain multiple log iterative association memories. The power supply operation status is identified by calling the decoder to identify the multiple power source operation scenario information and the multiple log iterative association memories, and the multiple distributed power source operation coefficients are determined. Then, multiple powers of the multiple distributed power sources at the current moment are collected, and the multiple powers are weightedly calculated using the multiple distributed power source operation coefficients to obtain the allocable output power, and then the charging and discharging demand is extracted from the multiple charging and discharging facility operation logs to obtain multiple charging and discharging demand information, wherein the multiple charging and discharging demand information has multiple demand direction identifiers, and then power allocation is performed based on the multiple charging and discharging demand information, the multiple demand direction identifiers and the allocable output power to obtain a power allocation plan, and power allocation management is performed on the charging and discharging facilities in the target area based on the power allocation plan. The technical effect of effective power distribution management and scheduling of charging and discharging facilities in the target area has been achieved.

[0105] Embodiment 2

[0106] Based on the same inventive concept as the power allocation management method for charging and discharging facilities based on power pooling in the aforementioned embodiment, Figure 2 As shown, the present application provides a power distribution management system for charging and discharging facilities based on power pooling, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. The system includes:

[0107] An operation log acquisition module 11 is used to collect operation data of multiple distributed power sources and multiple charging and discharging facilities in a target area, and obtain multiple distributed power source operation logs and multiple charging and discharging facility operation logs;

[0108] An operation scenario information obtaining module 12 is used to perform semantic analysis on the multiple distributed power supply operation logs using an encoder to obtain multiple power supply operation scenario information;

[0109] The log iterative association memory acquisition module 13 is used to input the multiple power supply operation scenario information into the information association module, perform interactive iterative capture of association words on the multiple distributed power supply operation logs, and obtain multiple log iterative association memories;

[0110] A distributed power supply operation coefficient determination module 14 is used to call a decoder to identify the power supply operation status of the multiple power supply operation scenario information and the multiple log iteration association memories respectively, and determine multiple distributed power supply operation coefficients;

[0111] The allocable output power acquisition module 15 is used to collect multiple powers of multiple distributed power sources at the current moment, and perform weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain the allocable output power;

[0112] A charging and discharging demand information obtaining module 16 is used to extract charging and discharging demand from the plurality of charging and discharging facility operation logs to obtain a plurality of charging and discharging demand information, wherein the plurality of charging and discharging demand information has a plurality of demand direction identifiers;

[0113] The power allocation management module 17 is used to allocate power based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power, obtain a power allocation plan, and manage power allocation for the charging and discharging facilities in the target area based on the power allocation plan.

[0114] Furthermore, the execution steps of the operation scenario information obtaining module 12 also include:

[0115] Performing data preprocessing on the multiple distributed power supply operation logs, and when the data preprocessing is completed, performing vector conversion on the text data to obtain multiple feature vector sets;

[0116] Building encoders based on recurrent neural networks;

[0117] The encoder's multi-layer neural network structure is used to perform feature recognition on the multiple feature vector sets to determine the multiple power supply operation scenario information.

[0118] Furthermore, the execution steps of the log iteration associated memory acquisition module 13 also include:

[0119] Taking time as an index, performing log serialization processing on the multiple distributed power supply operation logs to obtain multiple distributed power supply operation record sequences;

[0120] Extracting a plurality of first distributed power supply operation records that are located at the first position from the plurality of distributed power supply operation record sequences;

[0121] Extracting keywords from the plurality of power supply operation scenario information to obtain a plurality of scenario keyword sets, and storing the plurality of scenario keyword sets into a plurality of first memory vectors;

[0122] Extracting keywords from the plurality of first distributed power supply operation records to obtain a plurality of first keyword sets;

[0123] The multiple first memory vectors are interactively iteratively updated based on the multiple first keyword sets until the iterative capture of the multiple distributed power supply operation record sequences is completed, thereby obtaining the multiple log iterative associated memories.

[0124] Furthermore, the execution steps of the log iteration associated memory acquisition module 13 also include:

[0125] Capturing associated words based on the multiple first keyword sets and the multiple scene keyword sets, and adding the multiple first capture results into the multiple first memory vectors respectively to obtain multiple second memory vectors;

[0126] Extracting a plurality of second distributed power supply operation records located at the second position from the plurality of distributed power supply operation record sequences;

[0127] Extracting keywords from the plurality of second distributed power supply operation records to obtain a plurality of second keyword sets;

[0128] Capturing associated words based on the multiple second keyword sets, the multiple first keyword sets, the multiple scene keyword sets, the multiple first memory vectors, and the multiple second memory vectors, and adding the multiple second capture results to the multiple second memory vectors, respectively, to obtain multiple third memory vectors;

[0129] Based on the multiple third memory vectors and the multiple distributed power supply operation record sequences, associated words are interactively iterated and captured, and the multiple Mth memory vectors finally obtained are used as multiple log iterative associated memories, where M is the number of distributed power supply operation records in the distributed power supply operation record sequence.

[0130] Furthermore, the execution steps of the log iteration associated memory acquisition module 13 also include:

[0131] respectively calculating the semantic similarity between any first keyword in the plurality of first keyword sets and the plurality of scene keyword sets to obtain a plurality of first semantic similarity sets;

[0132] respectively extracting the co-occurrence frequency of any first keyword in the plurality of first keyword sets and the scene keywords in the plurality of scene keyword sets to obtain a plurality of first co-occurrence coefficient sets;

[0133] Performing weighted calculation on the multiple first semantic similarity sets and the multiple first co-occurrence coefficient sets to obtain multiple association coefficient sets;

[0134] The first keywords corresponding to the first n correlation coefficients in the plurality of correlation coefficient sets are respectively added to the plurality of first captured results, wherein n is a positive integer.

[0135] Furthermore, the execution steps of the distributed power supply operation coefficient determination module 14 also include:

[0136] Acquire multiple sample power supply operation scenario information, multiple sample log iterative association memory, and multiple sample distributed power supply operation coefficients as training data;

[0137] The training data is used to train the recurrent neural network to learn the mapping relationship between the power supply operation scenario information and the log iteration associated memory and the distributed power supply operation coefficients until convergence, thereby obtaining the trained decoder.

[0138] Furthermore, the execution steps of the charging and discharging demand information obtaining module 16 also include:

[0139] When the demand direction is marked as a positive sign, it indicates that the charging and discharging facilities need to be charged;

[0140] When the demand direction is marked as negative, it indicates that the charging and discharging facilities need to discharge.

[0141] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0142] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0143] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A power distribution management system for charging and discharging facilities based on power pooling, characterized in that: The system comprises: An operation log acquisition module is used to collect operation data of multiple distributed power sources and multiple charging and discharging facilities in a target area, and obtain multiple distributed power source operation logs and multiple charging and discharging facility operation logs; An operation scenario information acquisition module, used to perform semantic analysis on the multiple distributed power supply operation logs using an encoder to obtain multiple power supply operation scenario information; A log iterative association memory acquisition module, used to input the multiple power supply operation scenario information into the information association module, perform interactive iterative capture of association words on the multiple distributed power supply operation logs, and obtain multiple log iterative association memories; A distributed power supply operation coefficient determination module is used to call a decoder to identify the power supply operation status of the multiple power supply operation scenario information and the multiple log iteration association memories respectively, and determine multiple distributed power supply operation coefficients; The allocable output power acquisition module is used to collect multiple powers of multiple distributed power sources at the current moment, and perform weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain the allocable output power; A charging and discharging demand information obtaining module, used to extract charging and discharging demand from the operation logs of the plurality of charging and discharging facilities, and obtain a plurality of charging and discharging demand information, wherein the plurality of charging and discharging demand information has a plurality of demand direction identifiers; The power allocation management module is used to allocate power based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power, obtain a power allocation plan, and manage power allocation for the charging and discharging facilities in the target area based on the power allocation plan.

2. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 1, characterized in that: The execution steps of the operation scene information obtaining module also include: Performing data preprocessing on the multiple distributed power supply operation logs, and when the data preprocessing is completed, performing vector conversion on the text data to obtain multiple feature vector sets; Building encoders based on recurrent neural networks; The encoder's multi-layer neural network structure is used to perform feature recognition on the multiple feature vector sets to determine the multiple power supply operation scenario information.

3. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 1, characterized in that: The execution steps of the log iteration associated memory acquisition module also include: Taking time as an index, performing log serialization processing on the multiple distributed power supply operation logs to obtain multiple distributed power supply operation record sequences; Extracting a plurality of first distributed power supply operation records that are located at the first position from the plurality of distributed power supply operation record sequences; Extracting keywords from the plurality of power supply operation scenario information to obtain a plurality of scenario keyword sets, and storing the plurality of scenario keyword sets into a plurality of first memory vectors; Extracting keywords from the plurality of first distributed power supply operation records to obtain a plurality of first keyword sets; The multiple first memory vectors are interactively iteratively updated based on the multiple first keyword sets until the iterative capture of the multiple distributed power supply operation record sequences is completed, thereby obtaining the multiple log iterative associated memories.

4. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 3, characterized in that: The execution steps of the log iteration associated memory acquisition module also include: Capturing associated words based on the multiple first keyword sets and the multiple scene keyword sets, and adding the multiple first capture results into the multiple first memory vectors respectively to obtain multiple second memory vectors; Extracting a plurality of second distributed power supply operation records located at the second position from the plurality of distributed power supply operation record sequences; Extracting keywords from the plurality of second distributed power supply operation records to obtain a plurality of second keyword sets; Capturing associated words based on the multiple second keyword sets, the multiple first keyword sets, the multiple scene keyword sets, the multiple first memory vectors, and the multiple second memory vectors, and adding the multiple second capture results to the multiple second memory vectors, respectively, to obtain multiple third memory vectors; Based on the multiple third memory vectors and the multiple distributed power supply operation record sequences, associated words are interactively iterated and captured, and the multiple Mth memory vectors finally obtained are used as multiple log iterative associated memories, where M is the number of distributed power supply operation records in the distributed power supply operation record sequence.

5. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 4, characterized in that: The execution steps of the log iteration associated memory acquisition module also include: respectively calculating the semantic similarity between any first keyword in the plurality of first keyword sets and the plurality of scene keyword sets to obtain a plurality of first semantic similarity sets; Respectively extracting the co-occurrence frequency of any first keyword in the plurality of first keyword sets and the scene keywords in the plurality of scene keyword sets to obtain a plurality of first co-occurrence coefficient sets; Performing weighted calculation on the multiple first semantic similarity sets and the multiple first co-occurrence coefficient sets to obtain multiple association coefficient sets; The first keywords corresponding to the first n correlation coefficients in the plurality of correlation coefficient sets are respectively added to the plurality of first capture results, wherein n is a positive integer.

6. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 1, characterized in that: The execution steps of the distributed power supply operation coefficient determination module also include: Acquire multiple sample power supply operation scenario information, multiple sample log iterative association memory, and multiple sample distributed power supply operation coefficients as training data; The training data is used to train the recurrent neural network to learn the mapping relationship between the power supply operation scenario information and the log iteration associated memory and the distributed power supply operation coefficients until convergence, thereby obtaining the trained decoder.

7. The power distribution management system for charging and discharging facilities based on power pooling as claimed in claim 1, characterized in that: The execution steps of the charging and discharging demand information obtaining module also include: When the demand direction is marked as a positive sign, it indicates that the charging and discharging facilities need to be charged; When the demand direction is marked as negative, it indicates that the charging and discharging facilities need to discharge.

8. A power allocation management method for charging and discharging facilities based on power pooling, characterized in that: The method comprises: Collecting operation data of multiple distributed power sources and multiple charging and discharging facilities in the target area, and obtaining multiple distributed power source operation logs and multiple charging and discharging facility operation logs; Using an encoder to perform semantic analysis on the multiple distributed power supply operation logs to obtain multiple power supply operation scenario information; Inputting the multiple power supply operation scenario information into the information association module, performing interactive iterative capture of associated words on the multiple distributed power supply operation logs, and obtaining multiple log iterative association memories; Calling a decoder to identify the power supply operation status of the multiple power supply operation scenario information and the multiple log iteration association memories respectively, and determining multiple distributed power supply operation coefficients; Collecting multiple powers of multiple distributed power sources at the current moment, and performing weighted calculation on the multiple powers using the multiple distributed power source operation coefficients to obtain allocable output power; Extracting charging and discharging demands from the operation logs of the multiple charging and discharging facilities to obtain multiple charging and discharging demand information, wherein the multiple charging and discharging demand information has multiple demand direction identifiers; Power allocation is performed based on the multiple charging and discharging demand information, multiple demand direction identifiers and the allocable output power to obtain a power allocation plan, and power allocation management is performed on the charging and discharging facilities in the target area based on the power allocation plan.

Images